Compare commits

...
83 Commits
Author SHA1 Message Date
Oran Agra db09f6eb2e Redis 6.2.5
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
2021-07-21 21:06:49 +03:00
Huang Zhw 835d15b536 On 32 bit platform, the bit position of GETBIT/SETBIT/BITFIELD/BITCOUNT,BITPOS may overflow (see CVE-2021-32761) (#9191)
GETBIT, SETBIT may access wrong address because of wrap.
BITCOUNT and BITPOS may return wrapped results.
BITFIELD may access the wrong address but also allocate insufficient memory and segfault (see CVE-2021-32761).

This commit uses `uint64_t` or `long long` instead of `size_t`.
related https://github.com/redis/redis/pull/8096

At 32bit platform:
> setbit bit 4294967295 1
(integer) 0
> config set proto-max-bulk-len 536870913
OK
> append bit "\xFF"
(integer) 536870913
> getbit bit 4294967296
(integer) 0

When the bit index is larger than 4294967295, size_t can't hold bit index. In the past,  `proto-max-bulk-len` is limit to 536870912, so there is no problem.

After this commit, bit position is stored in `uint64_t` or `long long`. So when `proto-max-bulk-len > 536870912`, 32bit platforms can still be correct.

For 64bit platform, this problem still exists. The major reason is bit pos 8 times of byte pos. When proto-max-bulk-len is very larger, bit pos may overflow.
But at 64bit platform, we don't have so long string. So this bug may never happen.

Additionally this commit add a test cost `512MB` memory which is tag as `large-memory`. Make freebsd ci and valgrind ci ignore this test.

(cherry picked from commit 71d452876e)
2021-07-21 21:06:49 +03:00
Oran Agra bae0512c8a longer timeout in replication test (#8963)
the test normally passes. but we saw one failure in a valgrind run in github actions

(cherry picked from commit 8458baf6a9)
2021-07-21 21:06:49 +03:00
Huang Zhw 8d6134952a Remove testmodule in src/modules/Makefile. (#9250)
src/modules make failed. As in #3718 testmodule.c was removed. But the makefile was not updated

(cherry picked from commit d54c9086c2)
2021-07-21 21:06:49 +03:00
Oran Agra 1d7c0e5949 Fix failing basics moduleapi test on 32bit CI (#9140)
(cherry picked from commit 5ffdbae1f6)
2021-07-21 21:06:49 +03:00
Oran Agra 37b0f3617d Adjustments to recent RM_StringTruncate fix (#3718) (#9125)
- Introduce a new sdssubstr api as a building block for sdsrange.
  The API of sdsrange is many times hard to work with and also has
  corner case that cause bugs. sdsrange is easy to work with and also
  simplifies the implementation of sdsrange.
- Revert the fix to RM_StringTruncate and just use sdssubstr instead of
  sdsrange.
- Solve valgrind warnings from the new tests introduced by the previous
  PR.

(cherry picked from commit ae418eca24)
2021-07-21 21:06:49 +03:00
Huang Zhw 6866117194 Fix missing separator in module info line (usedby and using lists) (#9241)
Fix module info genModulesInfoStringRenderModulesList lack separator when there's more than one module in the list.

Co-authored-by: Oran Agra <oran@redislabs.com>
(cherry picked from commit 1895e134a7)
2021-07-21 21:06:49 +03:00
Binbin b622537199 SMOVE only notify dstset when the addition is successful. (#9244)
in case dest key already contains the member, the dest key isn't modified, so the command shouldn't invalidate watch.

(cherry picked from commit 11dc4e59b3)
2021-07-21 21:06:49 +03:00
qetu3790 355b1b6a57 Set TCP keepalive on inbound clusterbus connections (#9230)
Set TCP keepalive on inbound clusterbus connections to prevent memory leak

(cherry picked from commit f03af47a34)
2021-07-21 21:06:49 +03:00
Yossi Gottlieb d6f4273241 Fix compatibility with OpenSSL 1.1.0. (#9233)
(cherry picked from commit 277e4dc203)
2021-07-21 21:06:49 +03:00
Oran Agra de1b19ea88 fix valgrind issues with recently added test in modules/blockonbackground (#9192)
fixes test issue introduced in #9167

1. invalid reads due to accessing non-retained string (passed as unblock context).
2. leaking module blocked client context, see #6922 for info.

(cherry picked from commit a8518cce95)
2021-07-21 21:06:49 +03:00
Yossi Gottlieb 79fa5618f1 Fix CLIENT UNBLOCK crashing modules. (#9167)
Modules that use background threads with thread safe contexts are likely
to use RM_BlockClient() without a timeout function, because they do not
set up a timeout.

Before this commit, `CLIENT UNBLOCK` would result with a crash as the
`NULL` timeout callback is called. Beyond just crashing, this is also
logically wrong as it may throw the module into an unexpected client
state.

This commits makes `CLIENT UNBLOCK` on such clients behave the same as
any other client that is not in a blocked state and therefore cannot be
unblocked.

(cherry picked from commit aa139e2f02)
2021-07-21 21:06:49 +03:00
Huang Zhw 91bf2ab86d redis-cli cluster import command may issue wrong MIGRATE command. (#8945)
In clusterManagerCommandImport strcat was used to concat COPY and
REPLACE, the space maybe not enough.
If we use --cluster-replace but not --cluster-copy, the MIGRATE
command contained COPY instead of REPLACE.

(cherry picked from commit a049f6295a)
2021-07-21 21:06:49 +03:00
Binbin 88655019cc Fix accidental deletion of sinterstore command when we meet wrong type error. (#9032)
SINTERSTORE would have deleted the dest key right away,
even when later on it is bound to fail on an (WRONGTYPE) error.

With this change it first picks up all the input keys, and only later
delete the dest key if one is empty.

Also add more tests for some commands.
Mainly focus on
- `wrong type error`:
	expand test case (base on sinter bug) in non-store variant
	add tests for store variant (although it exists in non-store variant, i think it would be better to have same tests)
- the dstkey result when we meet `non-exist key (empty set)` in *store

sdiff:
- improve test case about wrong type error (the one we found in sinter, although it is safe in sdiff)
- add test about using non-exist key (treat it like an empty set)
sdiffstore:
- according to sdiff test case, also add some tests about `wrong type error` and `non-exist key`
- the different is that in sdiffstore, we will consider the `dstkey` result

sunion/sunionstore add more tests (same as above)

sinter/sinterstore also same as above ...

(cherry picked from commit b8a5da80c4)
2021-07-21 21:06:49 +03:00
Jason Elbaum fad44611dc Change return value type for ZPOPMAX/MIN in RESP3 (#8981)
When using RESP3, ZPOPMAX/ZPOPMIN should return nested arrays for consistency
with other commands (e.g. ZRANGE).

We do that only when COUNT argument is present (similarly to how LPOP behaves).
for reasoning see https://github.com/redis/redis/issues/8824#issuecomment-855427955

This is a breaking change only when RESP3 is used, and COUNT argument is present!

(cherry picked from commit 7f342020dc)
2021-07-21 21:06:49 +03:00
Mikhail Fesenko 8884971223 Direct redis-cli repl prints to stderr, because --rdb can print to stdout. fflush stdout after responses (#9136)
1. redis-cli can output --rdb data to stdout
   but redis-cli also write some messages to stdout which will mess up the rdb.

2. Make redis-cli flush stdout when printing a reply
  This was needed in order to fix a hung in redis-cli test that uses
  --replica.
   Note that printf does flush when there's a newline, but fwrite does not.

3. fix the redis-cli --replica test which used to pass previously
   because it didn't really care what it read, and because redis-cli
   used printf to print these other things to stdout.

4. improve redis-cli --replica test to run with both diskless and disk-based.

Co-authored-by: Oran Agra <oran@redislabs.com>
Co-authored-by: Viktor Söderqvist <viktor@zuiderkwast.se>
(cherry picked from commit 1eb4baa5b8)
2021-07-21 21:06:49 +03:00
Rob Snyder 07e1248686 Fix ziplist length updates on bigendian platforms (#2080)
Adds call to intrev16ifbe to ensure ZIPLIST_LENGTH is compared correctly

(cherry picked from commit eaa52719a3)
2021-07-21 21:06:49 +03:00
Oran Agra 6cd84b64f0 Test infra, handle RESP3 attributes and big-numbers and bools (#9235)
- promote the code in DEBUG PROTOCOL to addReplyBigNum
- DEBUG PROTOCOL ATTRIB skips the attribute when client is RESP2
- networking.c addReply for push and attributes generate assertion when
  called on a RESP2 client, anything else would produce a broken
  protocol that clients can't handle.

(cherry picked from commit 6a5bac309e)
2021-07-21 21:06:49 +03:00
Binbin c6b3966d02 hrandfield and zrandmember with count should return emptyarray when key does not exist. (#9178)
due to a copy-paste bug, it used to reply with null response rather than empty array.
this commit includes new tests that are looking at the RESP response directly in
order to be able to tell the difference between them.

Co-authored-by: Oran Agra <oran@redislabs.com>
(cherry picked from commit a418a2d3fc)
2021-07-21 21:06:49 +03:00
Oran Agra 432e056659 Tests: add a way to read raw RESP protocol reponses (#9193)
This makes it possible to distinguish between null response and an empty
array (currently the tests infra translates both to an empty string/list)

(cherry picked from commit 7103367ad4)
2021-07-21 21:06:49 +03:00
Mikhail Fesenko bbb6c2fb53 redis-cli --rdb: fix broken fsync/ftruncate for stdout (#9135)
A change in redis 6.2 caused redis-cli --rdb that's directed to stdout to fail because fsync fails.
This commit avoids doing ftruncate (fails with a warning) and fsync (fails with an error) when the
output file is `-`, and adds the missing documentation that `-` means stdout.

Co-authored-by: Oran Agra <oran@redislabs.com>
Co-authored-by: Wang Yuan <wangyuancode@163.com>
(cherry picked from commit 74fe15b360)
2021-07-21 21:06:49 +03:00
Leibale Eidelman 095a6e5937 fix ZRANGESTORE - should return 0 when src points to an empty key (#9089)
mistakenly it used to return an empty array rather than 0.

Co-authored-by: Oran Agra <oran@redislabs.com>
(cherry picked from commit 95274f1f8a)
2021-07-21 21:06:49 +03:00
guybe7 56020fd821 Include sizeof(struct stream) in objectComputeSize (#9164)
Affects MEMORY USAGE

(cherry picked from commit 4434cebbb3)
2021-07-21 21:06:49 +03:00
Binbin 11235c0e40 ZRANDMEMBER WITHSCORES with negative COUNT may return bad score (#9162)
Return a bad score when used with negative count (or count of 1), and non-ziplist encoded zset.
Also add test to validate the return value and cover the issue.

(cherry picked from commit 4bc5a8324d)
2021-07-21 21:06:49 +03:00
Evan 406caa5f56 modules: Add newlen == 0 handling to RM_StringTruncate (#3717) (#3718)
Previously, passing 0 for newlen would not truncate the string at all.
This adds handling of this case, freeing the old string and creating a new empty string.

Other changes:
- Move `src/modules/testmodule.c` to `tests/modules/basics.c`
- Introduce that basic test into the test suite
- Add tests to cover StringTruncate
- Add `test-modules` build target for the main makefile
- Extend `distclean` build target to clean modules too

(cherry picked from commit 1ccf2ca2f4)
2021-07-21 21:06:49 +03:00
yvette903 8bd5e39102 Fix coredump after Client Unpause command when threaded I/O is enabled (#9041)
Fix crash when using io-threads-do-reads and issuing CLIENT PAUSE and
CLIENT UNPAUSE.
This issue was introduced in redis 6.2 together with the FAILOVER command.

(cherry picked from commit 096c5fd5d2)
2021-07-21 21:06:49 +03:00
guybe7 c547dff650 Module API for current command name (#8792)
sometimes you can be very deep in the call stack, without access to argv.
once you're there you may want your reply/log to contain the command name.

(cherry picked from commit e16d3eb998)
2021-07-21 21:06:49 +03:00
Huang Zhw e6df2f6210 Fix XTRIM or XADD with LIMIT may delete more entries than Count. (#9048)
The decision to stop trimming due to LIMIT in XADD and XTRIM was after the limit was reached.
i.e. the code was deleting **at least** that count of records (from the LIMIT argument's perspective, not the MAXLEN),
instead of **up to** that count of records.
see #9046

(cherry picked from commit eaa7a7bb93)
2021-07-21 21:06:49 +03:00
Huang Zhw 312ccae003 XTRIM call streamParseAddOrTrimArgsOrReply use wrong arg xadd. (#9047)
xtrimCommand call streamParseAddOrTrimArgsOrReply should use xadd==0.

When the syntax is valid, it does not cause any bugs because the params of XADD is superset of XTRIM.
Just XTRIM will not respond with error on invalid syntax. The syntax of XADD will also be accpeted by XTRIM.

(cherry picked from commit 91f3689bf5)
2021-07-21 21:06:49 +03:00
YaacovHazan ff27217639 stabilize tests that involved with load handlers (#8967)
When test stop 'load handler' by killing the process that generating the load,
some commands that already in the input buffer, still might be processed by the server.
This may cause some instability in tests, that count on that no more commands
processed after we stop the `load handler'

In this commit, new proc 'wait_load_handlers_disconnected' added, to verify that no more
cammands from any 'load handler' prossesed, by checking that the clients who
genreate the load is disconnceted.

Also, replacing check of dbsize with wait_for_ofs_sync before comparing debug digest, as
it would fail in case the last key the workload wrote was an overridden key (not a new one).

Affected tests
Race fix:
- failover command to specific replica works
- Connect multiple replicas at the same time (issue #141), master diskless=$mdl, replica diskless=$sdl
- AOF rewrite during write load: RDB preamble=$rdbpre

Cleanup and speedup:
- Test replication with blocking lists and sorted sets operations
- Test replication with parallel clients writing in different DBs
- Test replication partial resync: $descr (diskless: $mdl, $sdl, reconnect: $reconnect

(cherry picked from commit 32a2584e07)
2021-07-21 21:06:49 +03:00
perryitay 3f4f9b6331 Fail EXEC command in case a watched key is expired (#9194)
There are two issues fixed in this commit: 
1. we want to fail the EXEC command in case there is a watched key that's logically
   expired but not yet deleted by active expire or lazy expire.
2. we saw that currently cache time is update in every `call()` (including nested calls),
   this time is being also being use for the isKeyExpired comparison, we want to update
   the cache time only in the first call (execCommand)

Co-authored-by: Oran Agra <oran@redislabs.com>
(cherry picked from commit ac8b1df885)
2021-07-21 21:06:49 +03:00
Huang Zhw 5b8e395174 Do not install a file event to send data to rewrite child when parent stop sending diff to child in aof rewrite. (#8767)
In aof rewrite, when parent stop sending data to child, if there is
new rewrite data, aofChildWriteDiffData write event will be installed.
Then this event is issued and deletes the file event without do anyting.
This will happen over and over again until aof rewrite finish.

This bug used to waste a few system calls per excessive wake-up
(epoll_ctl and epoll_wait) per cycle, each cycle triggered by receiving
a write command from a client.

(cherry picked from commit cb961d8c8e)
2021-07-21 21:06:49 +03:00
Omer Shadmi 9914b676b3 Avoid exiting to allow diskless loading to recover from RDB short read on module AUX data (#9199)
Currently a replica is able to recover from a short read (when diskless loading
is enabled) and avoid crashing/exiting, replying to the master and then the rdb
could be sent again by the master for another load attempt by the replica.
There were a few scenarios that were not behaving similarly, such as when
there is no end-of-file marker, or when module aux data failed to load, which
should be allowed to occur due to a short read.

(cherry picked from commit f06d782f5a)
2021-07-21 21:06:49 +03:00
Oran Agra abd44c8393 Fix race in client side tracking (#9116)
The `Tracking gets notification of expired keys` test in tracking.tcl
used to hung in valgrind CI quite a lot.

It turns out the reason is that with valgrind and a busy machine, the
server cron active expire cycle could easily run in the same event loop
as the command that created `mykey`, so that when they key got expired,
there were two change events to broadcast, one that set the key and one
that expired it, but since we used raxTryInsert, the client that was
associated with the "last" change was the one that created the key, so
the NOLOOP filtered that event.

This commit adds a test that reproduces the problem by using lazy expire
in a multi-exec which makes sure the key expires in the same event loop
as the one that added it.

(cherry picked from commit 9b564b525d)
2021-07-21 21:06:49 +03:00
Maxim Galushka f70916f8b0 redis-cli: support for REDIS_REPLY_SET in CSV and RAW output. (#7338)
Fixes #6792. Added support of REDIS_REPLY_SET in raw and csv output of `./redis-cli`

Test:

run commands to test:
  ./redis-cli -3 --csv COMMAND
  ./redis-cli -3 --raw COMMAND

Now they are returning resuts, were failing with: "Unknown reply type: 10" before the change.

(cherry picked from commit 96bb078577)
2021-07-21 21:06:49 +03:00
M Sazzadul Hoque ef3ad68020 Fix 6.2.4 release month to June (#9027) 2021-06-01 19:03:06 +03:00
Oran Agra f9470c9a17 Redis 6.2.4
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
2021-06-01 17:03:36 +03:00
Oran Agra e9a1438ac4 Fix integer overflow in STRALGO LCS (CVE-2021-32625) (#9011)
An integer overflow bug in Redis version 6.0 or newer can be exploited using the
STRALGO LCS command to corrupt the heap and potentially result with remote code
execution. This is a result of an incomplete fix by CVE-2021-29477.

(cherry picked from commit 1ddecf1958)
2021-06-01 17:03:36 +03:00
YaacovHazan 5102c0da92 unregister AE_READABLE from the read pipe in backgroundSaveDoneHandlerSocket (#8991)
In diskless replication, we create a read pipe for the RDB, between the child and the parent.
When we close this pipe (fd), the read handler also needs to be removed from the event loop (if it still registered).
Otherwise, next time we will use the same fd, the registration will be fail (panic), because
we will use EPOLL_CTL_MOD (the fd still register in the event loop), on fd that already removed from epoll_ctl

(cherry picked from commit 501d775583)
2021-06-01 17:03:36 +03:00
Wen Hui d68b775df5 fix sentinel test failure (#8983)
fix for recent breakage from #8958, did a mistake when updating the pr.

(cherry picked from commit be6ce8a92a)
2021-06-01 17:03:36 +03:00
Wen Hui ed995ffa03 [SENTINEL] reset sentinel-user/pass to NULL when user config with empty string (#8958)
(cherry picked from commit ae6f58690b)
2021-06-01 17:03:36 +03:00
Madelyn Olson 875a1f07d8 Hide migrate command from slowlog if they include auth (#8859)
Redact commands that include sensitive data from slowlog and monitor

(cherry picked from commit a59e75a475)
2021-06-01 17:03:36 +03:00
patpatbear d070546568 sinterstore: add missing keyspace del event when any source set not exists. (#8949)
this patch fixes sinterstore by add missing keyspace del event when any source set not exists.

Co-authored-by: srzhao <srzhao@sysnew.com>
(cherry picked from commit 46d9f31e94)
2021-06-01 17:03:36 +03:00
Oran Agra 511be70eba Fix crash unlinking a stream with groups rax and no groups (#8932)
When estimating the effort for unlink, we try to compute the effort of
the first group and extrapolate.
If there's a groups rax that's empty, there'a an assertion.

reproduce:
xadd s * a b
xgroup create s bla $
xgroup destroy s bla
unlink s

(cherry picked from commit 97108845e2)
2021-06-01 17:03:36 +03:00
Oran Agra 840f7f61c5 fix redis-benchmark to ignore unsupported configs (#8916)
Redis Enterprise supports the CONFIG GET command, but it replies with am
empty array since the save and appendonly configs are not supported.
before this fix redis-benchmark would segfault for trying to access the
error string on an array type reply.
see #8869

(cherry picked from commit 4d1094e8be)
2021-06-01 17:03:36 +03:00
Wang Yuan 9d69d6e8fb Fix wrong COW memory in log (#8917)
Always 0 MB of memory used by copy-on-write, introduced in #8645.

(cherry picked from commit 81e2d7272b)
2021-06-01 17:03:36 +03:00
yoav-steinberg 15c078df61 Enforce client output buffer soft limit when no traffic. (#8833)
When client breached the output buffer soft limit but then went idle,
we didn't disconnect on soft limit timeout, now we do.
Note this also resolves some sporadic test failures in due to Linux
buffering data which caused tests to fail if during the test we went
back under the soft COB limit.

Co-authored-by: Oran Agra <oran@redislabs.com>
Co-authored-by: sundb <sundbcn@gmail.com>
(cherry picked from commit 152fce5e2c)
2021-06-01 17:03:36 +03:00
Oran Agra e90e5640e7 Redis 6.2.3
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
2021-05-03 22:57:00 +03:00
Oran Agra 2df6695f2b Resolve nonsense static analysis warnings
(cherry picked from commit fd7d51c353)
2021-05-03 22:57:00 +03:00
Oran Agra 92e3b1802f Fix integer overflow in STRALGO LCS (CVE-2021-29477)
An integer overflow bug in Redis version 6.0 or newer could be exploited using
the STRALGO LCS command to corrupt the heap and potentially result with remote
code execution.

(cherry picked from commit f0c5f920d0)
2021-05-03 22:57:00 +03:00
Oran Agra 0463520693 Fix integer overflow in intset (CVE-2021-29478)
An integer overflow bug in Redis 6.2 could be exploited to corrupt the heap and
potentially result with remote code execution.

The vulnerability involves changing the default set-max-intset-entries
configuration value, creating a large set key that consists of integer values
and using the COPY command to duplicate it.

The integer overflow bug exists in all versions of Redis starting with 2.6,
where it could result with a corrupted RDB or DUMP payload, but not exploited
through COPY (which did not exist before 6.2).

(cherry picked from commit 29900d4e6b)
2021-05-03 22:57:00 +03:00
sundb 922e3bf59f Fix memory leak in moduleDefragGlobals (#8853)
(cherry picked from commit 5100ef9f82)
2021-05-03 22:57:00 +03:00
Huang Zhw 42f2ad0516 Improve redis-cli help. When help command, we only match command (#8879)
prefix args not all args. So when we help commands with subcommands,
all subcommands will be output.

(cherry picked from commit 0b1b9edb28)
2021-05-03 22:57:00 +03:00
Binbin 1eab6202ac redis-benchmark: Add zfree(data) and fix lrange size / text mismatch (#8872)
missing zfree(data) in redis-benchmark.

And also correct the wrong size in lrange.
the text mentioned 500, but size was 450, changed to 500

(cherry picked from commit 1eff8564c7)
2021-05-03 22:57:00 +03:00
Binbin d30ee6c44a redis-cli: Do not use hostsocket when we got redirected in cluster mode (#8870)
When redis-cli was used with both -c (cluster) and -s (unix socket),
it would have kept trying to use that unix socket, even if it got
redirected by the cluster (resulting in an infinite loop).

(cherry picked from commit 416f277339)
2021-05-03 22:57:00 +03:00
filipe oliveira 4f2da00e94 redis-benchmark: Error/Warning handling updates. (#8869)
- Immediately exit on errors that are not related to topology updates.
- Deprecates the `-e` option ( retro compatible ) and warns that we now
  exit immediately on errors that are not related to topology updates.
- Fixed wrongfully failing on config fetch error (warning only). This only affects RE.

Bottom line:
- MOVED and ASK errors will not show any warning (unlike the throttled error with `-e` before).
- CLUSTERDOWN still prints an error unconditionally and sleeps for 1 second.
- other errors are fatal.

(cherry picked from commit ef6f902372)
2021-05-03 22:57:00 +03:00
Huang Zhw 34b9a3fa2e Fix potential CONFIG SET bind test failure. (#8875)
Use an invalid IP address to trigger CONFIG SET bind failure, instead of DNS which is not guaranteed to always fail.

(cherry picked from commit 2b22fffc78)
2021-05-03 22:57:00 +03:00
yoav-steinberg 5c7b869e61 Bump freebsd-vm version to fix CI failures (#8876)
Specifically we had issues with NTP sync failure which was resolved here: https://github.com/vmactions/freebsd-vm/commit/457af7345642e154a79d219971a2d4a7c7fe2118

(cherry picked from commit 2e88b06396)
2021-05-03 22:57:00 +03:00
Oran Agra 6cbea7d29b Prevent replicas from sending commands that interact with keyspace (#8868)
This solves an issue reported in #8712 in which a replica would bypass
the client write pause check and cause an assertion due to executing a
write command during failover.

The fact is that we don't expect replicas to execute any command other
than maybe REPLCONF and PING, etc. but matching against the ADMIN
command flag is insufficient, so instead i just block keyspace access
for now.

(cherry picked from commit 46f4ebbe84)
2021-05-03 22:57:00 +03:00
Yossi Gottlieb 8cfa37fc21 Remove redundant -latomic on arm64. (#8867)
(cherry picked from commit ebfbb09109)
2021-05-03 22:57:00 +03:00
Yang Bodong 16c53085f7 When the password is wrong, redis-benchmark should exit (#8855)
(cherry picked from commit 8423b77f14)
2021-05-03 22:57:00 +03:00
Istemi Ekin Akkus 2ccb926314 Modules: Fix RM_GetClusterNodeInfo() to correctly populate the master_id (#8846)
(cherry picked from commit af035c1e1d)
2021-05-03 22:57:00 +03:00
zyxwvu Shi 6384fe3414 Use monotonic clock to check for Lua script timeout. (#8812)
This prevents a case where NTP moves the system clock
forward resulting in a false detection of a busy script.

Signed-off-by: zyxwvu Shi <i@shiyc.cn>
(cherry picked from commit f61c37cec9)
2021-05-03 22:57:00 +03:00
Wang Yuan ef64333e63 Expire key firstly and then notify keyspace event (#8830)
Modules event subscribers may get wrong things in notifyKeyspaceEvent callback,
such as wrong number of keys, or be able to lookup this key.
This commit changes the order to be like the one in evict.c.

Cleanup:
Since we know the key exists (it expires now), db*Delete is sure to return 1,
so there's no need to check it's output (misleading).

(cherry picked from commit 63acfe4b00)
2021-05-03 22:57:00 +03:00
bugwz 0851705304 Print the number of abnormal line in AOF (#8823)
When redis-check-aof finds an error, it prints the line number for faster troubleshooting. 

(cherry picked from commit 761d7d2771)
2021-05-03 22:57:00 +03:00
Huang Zhw b97a4ad7f2 Fix migrateCommand may migrate wrong value. (#8815)
This scene is hard to happen. When first attempt some keys expired,
only kv position is updated not ov. Then socket err happens, second
attempt is taken. This time kv items may be mismatching with ov items.

(cherry picked from commit 080d4579db)
2021-05-03 22:57:00 +03:00
Madelyn Olson d01afe885c Fix memory leak when doing lazyfreeing client tracking table (#8822)
Interior rax pointers were not being freed

(cherry picked from commit c73b4ddfd9)
2021-05-03 22:57:00 +03:00
Oran Agra 959d6035e5 Merge 6.2.2 release
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
Release 6.2.2
2021-04-20 08:03:58 +03:00
Oran Agra aa730ef1ea Redis 6.2.2 2021-04-19 21:39:40 +03:00
Oran Agra f5ca1f9ee9 Merge unstable into 6.2 2021-04-19 21:36:00 +03:00
Oran Agra 92bde124ca Merge Redis 6.2.1
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
Redis 6.2.1
2021-03-02 08:14:39 +02:00
Oran Agra 7018ad69a2 Redis 6.2.1 2021-03-01 21:02:05 +02:00
Oran Agra 46df4db87b Merge unstable into 6.2 2021-03-01 20:54:02 +02:00
Oran Agra 445aa844b9 Merge Redis 6.2.0 GA
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
Redis 6.2.0 GA
2021-02-22 23:23:58 +02:00
Oran Agra f098fe319b Redis 6.2.0 GA 2021-02-22 15:56:55 +02:00
Oran Agra 2ab6fef099 Merge origin/unstable into 6.2 2021-02-22 15:48:42 +02:00
Oran Agra 2dba1e391d Merge 6.2 RC3
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
CI / build-freebsd (push) Has been cancelled
2021-02-01 20:11:42 +02:00
Oran Agra 95338f9cc4 Redis 6.2 RC3 2021-01-31 19:55:20 +02:00
Oran Agra d0854927fc Merge branch 'unstable' into 6.2 2021-01-31 12:20:34 +02:00
Oran Agra ec2d180739 Merge 6.2 RC2
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
2021-01-12 16:21:03 +02:00
Oran Agra 049f2f0805 Redis 6.2 RC2 2021-01-12 16:17:58 +02:00
Oran Agra b34fc03e5e Merge unstable into 6.2 2021-01-12 10:14:04 +02:00
Oran Agra b8c67ce41b Redis 6.2 RC1.
CI / test-ubuntu-latest (push) Has been cancelled
CI / build-ubuntu-old (push) Has been cancelled
CI / build-macos-latest (push) Has been cancelled
CI / build-32bit (push) Has been cancelled
CI / build-libc-malloc (push) Has been cancelled
CI / build-centos7-jemalloc (push) Has been cancelled
2020-12-14 20:54:10 +02:00
75 changed files with 2087 additions and 445 deletions
+4 -4
View File
@@ -151,7 +151,7 @@ jobs:
run: |
sudo apt-get update
sudo apt-get install tcl8.6 valgrind -y
./runtest --valgrind --verbose --clients 1 --dump-logs
./runtest --valgrind --verbose --clients 1 --tags -large-memory --dump-logs
- name: module api test
run: ./runtest-moduleapi --valgrind --no-latency --verbose --clients 1
- name: unittest
@@ -171,7 +171,7 @@ jobs:
run: |
sudo apt-get update
sudo apt-get install tcl8.6 valgrind -y
./runtest --valgrind --verbose --clients 1 --dump-logs
./runtest --valgrind --verbose --clients 1 --tags -large-memory --dump-logs
- name: module api test
run: ./runtest-moduleapi --valgrind --no-latency --verbose --clients 1
@@ -253,14 +253,14 @@ jobs:
steps:
- uses: actions/checkout@v2
- name: test
uses: vmactions/freebsd-vm@v0.1.2
uses: vmactions/freebsd-vm@v0.1.4
with:
usesh: true
sync: rsync
prepare: pkg install -y bash gmake lang/tcl86
run: >
gmake &&
./runtest --accurate --verbose --no-latency --dump-logs &&
./runtest --accurate --verbose --no-latency --tags -large-memory --dump-logs &&
MAKE=gmake ./runtest-moduleapi --verbose &&
./runtest-sentinel &&
./runtest-cluster
+633 -11
View File
@@ -1,16 +1,638 @@
Hello! This file is just a placeholder, since this is the "unstable" branch
of Redis, the place where all the development happens.
Redis 6.2 release notes
=======================
There is no release notes for this branch, it gets forked into another branch
every time there is a partial feature freeze in order to eventually create
a new stable release.
--------------------------------------------------------------------------------
Upgrade urgency levels:
Usually "unstable" is stable enough for you to use it in development environments
however you should never use it in production environments. It is possible
to download the latest stable release here:
LOW: No need to upgrade unless there are new features you want to use.
MODERATE: Program an upgrade of the server, but it's not urgent.
HIGH: There is a critical bug that may affect a subset of users. Upgrade!
CRITICAL: There is a critical bug affecting MOST USERS. Upgrade ASAP.
SECURITY: There are security fixes in the release.
--------------------------------------------------------------------------------
http://download.redis.io/releases/redis-stable.tar.gz
================================================================================
Redis 6.2.5 Released Wed Jul 21 16:32:19 IDT 2021
================================================================================
More information is available at https://redis.io
Upgrade urgency: SECURITY, contains fixes to security issues that affect
authenticated client connections on 32-bit versions. MODERATE otherwise.
Happy hacking!
Fix integer overflow in BITFIELD on 32-bit versions (CVE-2021-32761).
An integer overflow bug in Redis version 2.2 or newer can be exploited using the
BITFIELD command to corrupt the heap and potentially result with remote code
execution.
Bug fixes that involve behavior changes:
* Change reply type for ZPOPMAX/MIN with count in RESP3 to nested array (#8981).
Was using a flat array like in RESP2 instead of a nested array like ZRANGE does.
* Fix reply type for HRANDFIELD and ZRANDMEMBER when key is missing (#9178).
Was using a null array instead of an empty array.
* Fix reply type for ZRANGESTORE when source key is missing (#9089).
Was using an empty array like ZRANGE instead of 0 (used in the STORE variant).
Bug fixes that are only applicable to previous releases of Redis 6.2:
* ZRANDMEMBER WITHSCORES with negative COUNT may return bad score (#9162)
* Fix crash after CLIENT UNPAUSE when threaded I/O config is enabled (#9041)
* Fix XTRIM or XADD with LIMIT may delete more entries than the limit (#9048)
* Fix build issue with OpenSSL 1.1.0 (#9233)
Other bug fixes:
* Fail EXEC command in case a watched key is expired (#9194)
* Fix SMOVE not to invalidate dest key (WATCH and tracking) when member already exists (#9244)
* Fix SINTERSTORE not to delete dest key when getting a wrong type error (#9032)
* Fix overflows on 32-bit versions in GETBIT, SETBIT, BITCOUNT, BITPOS, and BITFIELD (#9191)
* Improve MEMORY USAGE on stream keys (#9164)
* Set TCP keepalive on inbound cluster bus connections (#9230)
* Fix diskless replica loading to recover from RDB short read on module AUX data (#9199)
* Fix race in client side tracking (#9116)
* Fix ziplist length updates on big-endian platforms (#2080)
CLI tools:
* redis-cli cluster import command may issue wrong MIGRATE command, sending COPY instead of REPLACE (#8945)
* redis-cli --rdb fixes when using "-" to write to stdout (#9136, #9135)
* redis-cli support for RESP3 set type in CSV and RAW output (#7338)
Modules:
* Module API for getting current command name (#8792)
* Fix RM_StringTruncate when newlen is 0 (#3718)
* Fix CLIENT UNBLOCK crashing modules without timeout callback (#9167)
================================================================================
Redis 6.2.4 Released Tue June 1 12:00:00 IST 2021
================================================================================
Upgrade urgency: SECURITY, Contains fixes to security issues that affect
authenticated client connections. MODERATE otherwise.
Fix integer overflow in STRALGO LCS (CVE-2021-32625)
An integer overflow bug in Redis version 6.0 or newer can be exploited using the
STRALGO LCS command to corrupt the heap and potentially result with remote code
execution. This is a result of an incomplete fix by CVE-2021-29477.
Bug fixes that are only applicable to previous releases of Redis 6.2:
* Fix crash after a diskless replication fork child is terminated (#8991)
* Fix redis-benchmark crash on unsupported configs (#8916)
Other bug fixes:
* Fix crash in UNLINK on a stream key with deleted consumer groups (#8932)
* SINTERSTORE: Add missing keyspace del event when none of the sources exist (#8949)
* Sentinel: Fix CONFIG SET of empty string sentinel-user/sentinel-pass configs (#8958)
* Enforce client output buffer soft limit when no traffic (#8833)
Improvements:
* Hide AUTH passwords in MIGRATE command from slowlog (#8859)
================================================================================
Redis 6.2.3 Released Mon May 3 19:00:00 IST 2021
================================================================================
Upgrade urgency: SECURITY, Contains fixes to security issues that affect
authenticated client connections. LOW otherwise.
Integer overflow in STRALGO LCS command (CVE-2021-29477):
An integer overflow bug in Redis version 6.0 or newer could be exploited using
the STRALGO LCS command to corrupt the heap and potentially result in remote
code execution. The integer overflow bug exists in all versions of Redis
starting with 6.0.
Integer overflow in COPY command for large intsets (CVE-2021-29478):
An integer overflow bug in Redis 6.2 could be exploited to corrupt the heap and
potentially result with remote code execution. The vulnerability involves
changing the default set-max-intset-entries configuration value, creating a
large set key that consists of integer values and using the COPY command to
duplicate it. The integer overflow bug exists in all versions of Redis starting
with 2.6, where it could result with a corrupted RDB or DUMP payload, but not
exploited through COPY (which did not exist before 6.2).
Bug fixes that are only applicable to previous releases of Redis 6.2:
* Fix memory leak in moduleDefragGlobals (#8853)
* Fix memory leak when doing lazy freeing client tracking table (#8822)
* Block abusive replicas from sending command that could assert and crash redis (#8868)
Other bug fixes:
* Use a monotonic clock to check for Lua script timeout (#8812)
* redis-cli: Do not use unix socket when we got redirected in cluster mode (#8870)
Modules:
* Fix RM_GetClusterNodeInfo() to correctly populate master id (#8846)
================================================================================
Redis 6.2.2 Released Mon April 19 19:00:00 IST 2021
================================================================================
Upgrade urgency: HIGH, if you're using ACL and pub/sub, CONFIG REWRITE, or
suffering from performance regression. see below.
Bug fixes for regressions in previous releases of Redis 6.2:
* Fix BGSAVE, AOFRW, and replication slowdown due to child reporting CoW (#8645)
* Fix short busy loop when timer event is about to fire (#8764)
* Fix default user, overwritten and reset users losing pubsub channel permissions (#8723)
* Fix config rewrite with an empty `save` config resulsing in default `save` values (#8719)
* Fix not starting on alpine/libmusl without IPv6 (#8655)
* Fix issues with propagation and MULTI/EXEC in modules (#8617)
Several issues around nested calls and thread safe contexts
Bug fixes that are only applicable to previous releases of Redis 6.2:
* ACL Pub/Sub channels permission handling for save/load scenario (#8794)
* Fix early rejection of PUBLISH inside MULTI-EXEC transaction (#8534)
* Fix missing SLOWLOG records for blocked commands (#8632)
* Allow RESET command during busy scripts (#8629)
* Fix some error replies were not counted on stats (#8659)
Bug fixes:
* Add a timeout mechanism for replicas stuck in fullsync (#8762)
* Process HELLO command even if the default user has no permissions (#8633)
* Client issuing a long running script and using a pipeline, got disconnected (#8715)
* Fix script kill to work also on scripts that use `pcall` (#8661)
* Fix list-compress-depth may compress more node than required (#8311)
* Fix redis-cli handling of rediss:// URL scheme (#8705)
* Cluster: Skip unnecessary check which may prevent failure detection (#8585)
* Cluster: Fix hang manual failover when replica just started (#8651)
* Sentinel: Fix info-refresh time field before sentinel get first response (#8567)
* Sentinel: Fix possible crash on failed connection attempt (#8627)
* Systemd: Send the readiness notification when a replica is ready to accept connections (#8409)
Command behavior changes:
* ZADD: fix wrong reply when INCR used with GT/LT which blocked the update (#8717)
It was responding with the incremented value rather than nil
* XAUTOCLAIM: fix response to return the next available id as the cursor (#8725)
Previous behavior was retuning the last one which was already scanned
* XAUTOCLAIM: fix JUSTID to prevent incrementing delivery_count (#8724)
New config options:
* Add cluster-allow-replica-migration config option (#5285)
* Add replica-announced config option (#8653)
* Add support for plaintext clients in TLS cluster (#8587)
* Add support for reading encrypted keyfiles (#8644)
Improvements:
* Fix performance regression in BRPOP on Redis 6.0 (#8689)
* Avoid adding slowlog entries for config with sensitive data (#8584)
* Improve redis-cli non-binary safe string handling (#8566)
* Optimize CLUSTER SLOTS reply (#8541)
* Handle remaining fsync errors (#8419)
Info fields and introspection changes:
* Strip % sign from current_fork_perc info field (#8628)
* Fix RSS memory info on FreeBSD (#8620)
* Fix client_recent_max_input/output_buffer in 'INFO CLIENTS' when all clients drop (#8588)
* Fix invalid master_link_down_since_seconds in info replication (#8785)
Platform and deployment-related changes:
* Fix FreeBSD <12.x builds (#8603)
Modules:
* Add macros for RedisModule_log logging levels (#4246)
* Add RedisModule_GetAbsExpire / RedisModule_SetAbsExpire (#8564)
* Add a module type for key space notification (#8759)
* Set module eviction context flag only in masters (#8631)
* Fix unusable RedisModule_IsAOFClient API (#8596)
* Fix missing EXEC on modules propagation after failed EVAL execution (#8654)
* Fix edge-case when a module client is unblocked (#8618)
================================================================================
Redis 6.2.1 Released Mon Mar 1 17:51:36 IST 2021
================================================================================
Upgrade urgency: LOW.
Here is a comprehensive list of changes in this release compared to 6.2.0,
each one includes the PR number that added it, so you can get more details
at https://github.com/redis/redis/pull/<number>
Bug fixes:
* Fix sanitize-dump-payload for stream with deleted records (#8568)
* Prevent client-query-buffer-limit config from being set to lower than 1mb (#8557)
Improvements:
* Make port, tls-port and bind config options modifiable at runtime (#8510)
Platform and deployment-related changes:
* Fix compilation error on non-glibc systems if jemalloc is not used (#8533)
* Improved memory consumption and memory usage tracking on FreeBSD (#8545)
* Fix compilation on ARM64 MacOS with jemalloc (#8458)
Modules:
* New Module API for getting user name of a client (#8508)
* Optimize RM_Call by utilizing a shared reusable client (#8516)
* Fix crash running CLIENT INFO via RM_Call (#8560)
================================================================================
Redis 6.2.0 GA Released Tue Feb 22 14:00:00 IST 2021
================================================================================
Upgrade urgency: SECURITY if you use 32bit build of redis (see bellow), MODERATE
if you used earlier versions of Redis 6.2, LOW otherwise.
Integer overflow on 32-bit systems (CVE-2021-21309):
Redis 4.0 or newer uses a configurable limit for the maximum supported bulk
input size. By default, it is 512MB which is a safe value for all platforms.
If the limit is significantly increased, receiving a large request from a client
may trigger several integer overflow scenarios, which would result with buffer
overflow and heap corruption.
Here is a comprehensive list of changes in this release compared to 6.2 RC3,
each one includes the PR number that added it, so you can get more details
at https://github.com/redis/redis/pull/<number>
Bug fixes:
* Avoid 32-bit overflows when proto-max-bulk-len is set high (#8522)
* Fix broken protocol in client tracking tracking-redir-broken message (#8456)
* Avoid unsafe field name characters in INFO commandstats, errorstats, modules (#8492)
* XINFO able to access expired keys during CLIENT PAUSE WRITE (#8436)
* Fix allowed length for REPLCONF ip-address, needed due to Sentinel's support for hostnames (#8517)
* Fix broken protocol in redis-benchmark when used with -a or --dbnum (#8486)
* XADD counts deleted records too when considering switching to a new listpack (#8390)
Bug fixes that are only applicable to previous releases of Redis 6.2:
* Fixes in GEOSEARCH bybox (accuracy and mismatch between width and height) (#8445)
* Fix risk of OOM panic in HRANDFIELD, ZRANDMEMBER commands with huge negative count (#8429)
* Fix duplicate replicas issue in Sentinel, needed due to hostname support (#8481)
* Fix Sentinel configuration rewrite, an improvement of #8271 (#8480)
Command behavior changes:
* SRANDMEMBER uses RESP3 array type instead of set type (#8504)
* EXPIRE, EXPIREAT, SETEX, GETEX: Return error when provided expire time overflows (#8287)
Other behavior changes:
* Remove ACL subcommand validation if fully added command exists. (#8483)
Improvements:
* Optimize sorting in GEORADIUS / GEOSEARCH with COUNT (#8326)
* Optimize HRANDFIELD and ZRANDMEMBER case 4 when ziplist encoded (#8444)
* Optimize in-place replacement of elements in HSET, HINCRBY, LSET (#8493)
* Remove redundant list to store pubsub patterns (#8472)
* Add --insecure option to command line tools (#8416)
Info fields and introspection changes:
* Add INFO fields to track progress of BGSAVE, AOFRW, replication (#8414)
Modules:
* RM_ZsetRem: Delete key if empty, the bug could leave empty zset keys (#8453)
* RM_HashSet: Add COUNT_ALL flag and set errno (#8446)
================================================================================
Redis 6.2 RC3 Released Tue Feb 1 14:00:00 IST 2021
================================================================================
Upgrade urgency LOW: This is the third Release Candidate of Redis 6.2.
Here is a comprehensive list of changes in this release compared to 6.2 RC2,
each one includes the PR number that added it, so you can get more details
at https://github.com/redis/redis/pull/<number>
New commands / args:
* Add HRANDFIELD and ZRANDMEMBER commands (#8297)
* Add FAILOVER command (#8315)
* Add GETEX, GETDEL commands (#8327)
* Add PXAT/EXAT arguments to SET command (#8327)
* Add SYNC arg to FLUSHALL and FLUSHDB, and ASYNC/SYNC arg to SCRIPT FLUSH (#8258)
Sentinel:
* Add hostname support to Sentinel (#8282)
* Prevent file descriptors from leaking into Sentinel scripts (#8242)
* Fix config file line order dependency and config rewrite sequence (#8271)
New configuration options:
* Add set-proc-title config option to disable changes to the process title (#3623)
* Add proc-title-template option to control what's shown in the process title (#8397)
* Add lazyfree-lazy-user-flush config option to control FLUSHALL, FLUSHDB and SCRIPT FLUSH (#8258)
Bug fixes:
* AOF: recover from last write error by turning on/off appendonly config (#8030)
* Exit on fsync error when the AOF fsync policy is 'always' (#8347)
* Avoid assertions (on older kernels) when testing arm64 CoW bug (#8405)
* CONFIG REWRITE should honor umask settings (#8371)
* Fix firstkey,lastkey,step in COMMAND command for some commands (#8367)
Special considerations:
* Fix misleading description of the save configuration directive (#8337)
Improvements:
* A way to get RDB file via replication without excessive replication buffers (#8303)
* Optimize performance of clusterGenNodesDescription for large clusters (#8182)
Info fields and introspection changes:
* SLOWLOG and LATENCY monitor include unblocking time of blocked commands (#7491)
Modules:
* Add modules API for streams (#8288)
* Add event for fork child birth and termination (#8289)
* Add RM_BlockedClientMeasureTime* etc, to track background processing in commandstats (#7491)
* Fix bug in v6.2, wrong value passed to the new unlink callback (#8381)
* Fix bug in v6.2, modules blocked on keys unblock on commands like LPUSH (#8356)
================================================================================
Redis 6.2 RC2 Released Tue Jan 12 16:17:20 IST 2021
================================================================================
Upgrade urgency LOW: This is the second Release Candidate of Redis 6.2.
IMPORTANT: If you're running Redis on ARM64 or a big-endian system, upgrade may
have significant implications. Please be sure to read the notes below.
Here is a comprehensive list of changes in this release compared to 6.2 RC1,
each one includes the PR number that added it, so you can get more details
at https://github.com/redis/redis/pull/<number>
New commands / args:
* Add the REV, BYLEX and BYSCORE arguments to ZRANGE, and the ZRANGESTORE command (#7844)
* Add the XAUTOCLAIM command (#7973)
* Add the MINID trimming strategy and the LIMIT argument to XADD and XTRIM (#8169)
* Add the ANY argument to GEOSEARCH and GEORADIUS (#8259)
* Add the CH, NX, XX arguments to GEOADD (#8227)
* Add the COUNT argument to LPOP and RPOP (#8179)
* Add the WRITE argument to CLIENT PAUSE for pausing write commands exclusively (#8170)
* Change the proto-ver argument of HELLO to optional (#7377)
* Add the CLIENT TRACKINGINFO subcommand (#7309)
Command behavior changes:
* CLIENT TRACKING yields an error when given overlapping BCAST prefixes (#8176)
* SWAPDB invalidates WATCHed keys (#8239)
* SORT command behaves differently when used on a writable replica (#8283)
Other behavior changes:
* Avoid propagating MULTI/EXEC for read-only transactions (#8216)
* Remove the read-only flag from TIME, ECHO, ROLE, LASTSAVE (#8216)
* Fix the command flags of PFDEBUG (#8222)
* Tracking clients will no longer receive unnecessary key invalidation messages after FLUSHDB (#8039)
* Sentinel: Fix missing updates to the config file after SENTINEL SET command (#8229)
Bug fixes with compatibility implications (bugs introduced in Redis 6.0):
* Fix RDB CRC64 checksum on big-endian systems (#8270)
If you're using big-endian please consider the compatibility implications with
RESTORE, replication and persistence.
* Fix wrong order of key/value in Lua's map response (#8266)
If your scripts use redis.setresp() or return a map (new in Redis 6.0), please
consider the implications.
Bug fixes that are only applicable to previous releases of Redis 6.2:
* Resolve rare assertions in active defragmentation while loading (#8284, #8281)
Bug fixes:
* Fix the selection of a random element from large hash tables (#8133)
* Fix an issue where a forked process deletes the parent's pidfile (#8231)
* Fix crashes when enabling io-threads-do-reads (#8230)
* Fix a crash in redis-cli after executing cluster backup (#8267)
* Fix redis-benchmark to use an IP address for the first cluster node (#8154)
* Fix saving of strings larger than 2GB into RDB files (#8306)
Additional improvements:
* Improve replication handshake time (#8214)
* Release client tracking table memory asynchronously in cases where the DB is also freed asynchronously (#8039)
* Avoid wasteful transient memory allocation in certain cases (#8286, #5954)
* Handle binary string values by the 'requirepass' and 'masterauth' configs (#8200)
Platform and deployment-related changes:
* Install redis-check-rdb and redis-check-aof as symlinks to redis-server (#5745)
* Add a check for an ARM64 Linux kernel bug (#8224)
Due to the potential severity of this issue, Redis will refuse to run on
affected platforms by default.
Info fields and introspection changes:
* Add the errorstats section to the INFO command (#8217)
* Add the failed_calls and rejected_calls fields INFO's commandstats section (#8217)
* Report child copy-on-write metrics continuously (#8264)
Module API changes:
* Add the RedisModule_SendChildCOWInfo API (#8264)
* Add the may-replicate command flag (#8170)
================================================================================
Redis 6.2 RC1 Released Mon Dec 14 11:50:00 IST 2020
================================================================================
Upgrade urgency LOW: This is the first Release Candidate of Redis 6.2.
Introduction to the Redis 6.2 release
=====================================
This release is the first significant Redis release managed by the core team
under the new project governance model.
Redis 6.2 includes many new commands and improvements, but no big features. It
mainly makes Redis more complete and addresses issues that have been requested
by many users frequently or for a long time.
Many of these changes were not eligible for 6.0.x for several reasons:
1. They are not backward compatible, which is always the case with new or
extended commands (that cannot be replicated to an older replica).
2. They require a longer release-candidate test cycle.
Here is a comprehensive list of changes in this release compared to 6.0.9,
each one includes the PR number that added it, so you can get more details
at https://github.com/redis/redis/pull/<number>
New commands / args:
* Add SMISMEMBER command that checks multiple members (#7615)
* Add ZMSCORE command that returns an array of scores (#7593)
* Add LMOVE and BLMOVE commands that pop and push arbitrarily (#6929)
* Add RESET command that resets client connection state (#7982)
* Add COPY command that copies keys (#7953)
* Add ZDIFF and ZDIFFSTORE commands (#7961)
* Add ZINTER and ZUNION commands (#7794)
* Add GEOSEARCH/GEOSEARCHSTORE commands for bounding box spatial queries (#8094)
* Add GET parameter to SET command, for more powerful GETSET (#7852)
* Add exclusive range query to XPENDING (#8130)
* Add exclusive range query to X[REV]RANGE (#8072)
* Add GT and LT options to ZADD for conditional score updates (#7818)
* Add CLIENT INFO and CLIENT LIST for specific ids (#8113)
* Add IDLE argument to XPENDING command (#7972)
* Add local address to CLIENT LIST, and a CLIENT KILL filter. (#7913)
* Add NOMKSTREAM option to XADD command (#7910)
* Add command introspection to Sentinel (#7940)
* Add SENTINEL MYID subcommand (#7858)
New features:
* Dump payload sanitization: prevent corrupt payload causing crashes (#7807)
Has flags to enable full O(N) validation (disabled by default).
* ACL patterns for Pub/Sub channels (#7993)
* Support ACL for Sentinel mode (#7888)
* Support getting configuration from both stdin and file at the same time (#7893)
Lets you avoid storing secrets on the disk.
New features in CLI tools:
* redis-cli RESP3 push support (#7609)
* redis-cli cluster import support source and target that require auth (#7994)
* redis-cli URIs able to provide user name in addition to password (#8048)
* redis-cli/redis-benchmark allow specifying the prefered ciphers/ciphersuites (#8005)
* redis-cli add -e option to exit with code when command execution fails (#8136)
Command behavior changes:
* EXISTS should not alter LRU (#8016)
In Redis 5.0 and 6.0 it would have touched the LRU/LFU of the key.
* OBJECT should not reveal logically expired keys (#8016)
Will now behave the same TYPE or any other non-DEBUG command.
* Improve db id range check for SELECT and MOVE (#8085)
Changes the error message text on a wrong db index.
* Modify AUTH / HELLO error message (#7648)
Changes the error message text when the user isn't found or is disabled.
* BITOPS length limited to proto_max_bulk_len rather than 512MB (#8096)
The limit is now configurable like in SETRANGE, and APPEND.
* GEORADIUS[BYMEMBER] can fail with -OOM if Redis is over the memory limit (#8107)
Other behavior changes:
* Optionally (default) fail to start if requested bind address is not available (#7936)
If you rely on Redis starting successfully even if one of the bind addresses
is not available, you'll need to tune the new config.
* Limit the main db dictionaries expansion to prevent key eviction (#7954)
In the past big dictionary rehashing could result in massive data eviction.
Now this rehashing is delayed (up to a limit), which can result in performance
loss due to hash collisions.
* CONFIG REWRITE is atomic and safer, but requires write access to the config file's folder (#7824, #8051)
This change was already present in 6.0.9, but was missing from the release
notes.
* A new incremental eviction mechanism that reduces latency on eviction spikes (#7653)
In pathological cases this can cause memory to grow uncontrolled and may require
specific tuning.
* Not resetting "save" config when Redis is started with command line arguments. (#7092)
In case you provide command line arguments without "save" and count on it
being disabled, Now the defaults "save" config will kick in.
* Update memory metrics for INFO during loading (#7690)
* When "supervised" config is enabled, it takes precedence over "daemonize". (#8036)
* Assertion and panic, print crash log without generating SIGSEGV (#7585)
* Added crash log report on SIGABRT, instead of silently exiting (#8004)
* Disable THP (Transparent Huge Pages) if enabled (#7381)
If you deliberately enabled it, you'll need to config Redis to keep it.
Bug fixes:
* Handle output buffer limits for module blocked clients (#8141)
Could result in a module sending reply to a blocked client to go beyond the
limit.
* Fix setproctitle related crashes. (#8150, #8088)
Caused various crashes on startup, mainly on Apple M1 chips or under
instrumentation.
* A module doing RM_Call could cause replicas to get nested MULTI (#8097).
* Backup/restore cluster mode keys to slots map for repl-diskless-load=swapdb (#8108)
In cluster mode with repl-diskless-load, when loading failed, slot map
wouldn't have been restored.
* Fix oom-score-adj-values range, and bug when used in config file (#8046)
Enabling setting this in the config file in a line after enabling it, would
have been buggy.
* Reset average ttl when empty databases (#8106)
Just causing misleading metric in INFO
* Disable rehash when Redis has child process (#8007)
This could have caused excessive CoW during BGSAVE, replication or AOFRW.
* Further improved ACL algorithm for picking categories (#7966)
Output of ACL GETUSER is now more similar to the one provided by ACL SETUSER.
* Fix bug with module GIL being released prematurely (#8061)
Could in theory (and rarely) cause multi-threaded modules to corrupt memory.
* Fix cluster redirect for module command with no firstkey. (#7539)
* Reduce effect of client tracking causing feedback loop in key eviction (#8100)
* Kill disk-based fork child when all replicas drop and 'save' is not enabled (#7819)
* Rewritten commands (modified for propagation) are logged as their original command (#8006)
* Fix cluster access to unaligned memory (SIGBUS on old ARM) #7958
* If diskless repl child is killed, make sure to reap the child pid (#7742)
* Broadcast a PONG message when slot's migration is over, may reduce MOVED responses (#7571)
Other improvements:
* TLS Support in redis-benchmark (#7959)
* Accelerate diskless master connections, and general re-connections (#6271)
* Run active defrag while blocked / loading (#7726)
* Performance and memory reporting improvement - sds take control of its internal fragmentation (#7875)
* Speedup cluster failover. (#7948)
Platform / toolchain support related improvements:
* Optionally (not by default) use H/W Monotonic clock for faster time sampling (#7644)
* Remove the requirements for C11 and _Atomic supporting compiler (#7707)
This would allow to more easily build and use Redis on older systems and
compilers again.
* Fix crash log registers output on ARM. (#8020)
* Raspberry build fix. (#8095)
* Setting process title support for Haiku. (#8060)
* DragonFlyBSD RSS memory sampling support. (#8023)
New configuration options:
* Enable configuring OpenSSL using the standard openssl.cnf (#8143)
* oom-score-adj-values config can now take absolute values (besides relative ones) (#8046)
* TLS: Add different client cert support. (#8076)
* Note that a few other changes listed above added their config options.
Info fields and introspection changes:
* Add INFO fields to track diskless and disk-based replication progress (#7981)
* Add INFO field for main thread cpu time, and scrape system time. (#8132)
* Add total_forks to INFO STATS (#8155)
* Add maxclients and cluster_connections to INFO CLIENTS (#7979)
* Add tracking bcast flag and client redirection in client list (#7995)
* Fixed INFO client_recent_max_input_buffer includes argv array (#8065, see #7874)
* Note that a few other changes listed above added their info fields.
Module API changes:
* Add CTX_FLAGS_DENY_BLOCKING as a unified the way to know if blocking is allowed (#8025)
* Add data type callbacks for lazy free effort, and unlink (#7912)
* Add data type callback for COPY command (#8112)
* Add callbacks for defrag support. (#8149)
* Add module event for repl-diskless-load swapdb (#8153)
Module related fixes:
* Moved RMAPI_FUNC_SUPPORTED so that it's usable (#8037)
* Improve timer accuracy (#7987)
* Allow '\0' inside of result of RM_CreateStringPrintf (#6260)
Thanks to all the users and developers who made this release possible.
We'll follow up with more RC releases, until the code looks production ready
and we don't get reports of serious issues for a while.
A special thank you for the amount of work put into this release by:
- Oran Agra
- Yossi Gottlieb
- Viktor Söderqvist
- Yang Bodong
- Filipe Oliveira
- Guy Benoish
- Itamar Haber
- Madelyn Olson
- Wang Yuan
- Felipe Machado
- Wen Hui
- Tatsuya Arisawa
- Jonah H. Harris
- Raghav Muddur
- Jim Brunner
- Yaacov Hazan
- Allen Farris
- Chen Yang
- Nitai Caro
- sundb
- Meir Shpilraien
- maohuazhu
- Valentino Geron
- Zhao Zhao
- Qu Chen
- George Prekas
- Tyson Andre
- Uri Yagelnik
- Michael Grunder
- Huang Zw
- alexronke-channeladvisor
- Andy Pan
- Wu Yunlong
- Wei Kukey
- Yoav Steinberg
- Greg Femec
- Uri Shachar
- Nykolas Laurentino de Lima
- xhe
- zhenwei pi
- David CARLIER
Migrating from 6.0 to 6.2
=========================
Redis 6.2 is mostly a strict superset of 6.0, you should not have any problem
upgrading your application from 6.0 to 6.2. However there are some small changes
of behavior listed above, please make sure you are not badly affected by any of
them.
Specifically these sections:
* Command behavior changes
* Other behavior changes
--------------------------------------------------------------------------------
Cheers,
The Redis team
@@ -235,7 +235,7 @@ iget_defrag_hint(tsdn_t *tsdn, void* ptr) {
int free_in_slab = extent_nfree_get(slab);
if (free_in_slab) {
const bin_info_t *bin_info = &bin_infos[binind];
int curslabs = bin->stats.curslabs;
unsigned long curslabs = bin->stats.curslabs;
size_t curregs = bin->stats.curregs;
if (bin->slabcur) {
/* remove slabcur from the overall utilization */
+1
View File
@@ -16,6 +16,7 @@ fi
$MAKE -C tests/modules && \
$TCLSH tests/test_helper.tcl \
--single unit/moduleapi/commandfilter \
--single unit/moduleapi/basics \
--single unit/moduleapi/fork \
--single unit/moduleapi/testrdb \
--single unit/moduleapi/infotest \
+6 -5
View File
@@ -93,14 +93,10 @@ FINAL_LDFLAGS=$(LDFLAGS) $(REDIS_LDFLAGS) $(DEBUG)
FINAL_LIBS=-lm
DEBUG=-g -ggdb
# Linux ARM needs -latomic at linking time
ifneq (,$(filter aarch64 armv,$(uname_M)))
FINAL_LIBS+=-latomic
else
# Linux ARM32 needs -latomic at linking time
ifneq (,$(findstring armv,$(uname_M)))
FINAL_LIBS+=-latomic
endif
endif
ifeq ($(uname_S),SunOS)
# SunOS
@@ -379,6 +375,8 @@ clean:
distclean: clean
-(cd ../deps && $(MAKE) distclean)
-(cd modules && $(MAKE) clean)
-(cd ../tests/modules && $(MAKE) clean)
-(rm -f .make-*)
.PHONY: distclean
@@ -386,6 +384,9 @@ distclean: clean
test: $(REDIS_SERVER_NAME) $(REDIS_CHECK_AOF_NAME) $(REDIS_CLI_NAME) $(REDIS_BENCHMARK_NAME)
@(cd ..; ./runtest)
test-modules: $(REDIS_SERVER_NAME)
@(cd ..; ./runtest-moduleapi)
test-sentinel: $(REDIS_SENTINEL_NAME) $(REDIS_CLI_NAME)
@(cd ..; ./runtest-sentinel)
+9 -4
View File
@@ -1892,10 +1892,6 @@ void addACLLogEntry(client *c, int reason, int argpos, sds username) {
void aclCommand(client *c) {
char *sub = c->argv[1]->ptr;
if (!strcasecmp(sub,"setuser") && c->argc >= 3) {
/* Consider information about passwords or permissions
* to be sensitive, which will be the arguments for this
* subcommand. */
preventCommandLogging(c);
sds username = c->argv[2]->ptr;
/* Check username validity. */
if (ACLStringHasSpaces(username,sdslen(username))) {
@@ -1912,6 +1908,12 @@ void aclCommand(client *c) {
user *u = ACLGetUserByName(username,sdslen(username));
if (u) ACLCopyUser(tempu, u);
/* Initially redact all of the arguments to not leak any information
* about the user. */
for (int j = 2; j < c->argc; j++) {
redactClientCommandArgument(c, j);
}
for (int j = 3; j < c->argc; j++) {
if (ACLSetUser(tempu,c->argv[j]->ptr,sdslen(c->argv[j]->ptr)) != C_OK) {
const char *errmsg = ACLSetUserStringError();
@@ -2245,6 +2247,8 @@ void authCommand(client *c) {
addReplyErrorObject(c,shared.syntaxerr);
return;
}
/* Always redact the second argument */
redactClientCommandArgument(c, 1);
/* Handle the two different forms here. The form with two arguments
* will just use "default" as username. */
@@ -2264,6 +2268,7 @@ void authCommand(client *c) {
} else {
username = c->argv[1];
password = c->argv[2];
redactClientCommandArgument(c, 2);
}
if (ACLAuthenticateUser(c,username,password) == C_OK) {
+3 -1
View File
@@ -162,7 +162,9 @@ void aofRewriteBufferAppend(unsigned char *s, unsigned long len) {
/* Install a file event to send data to the rewrite child if there is
* not one already. */
if (aeGetFileEvents(server.el,server.aof_pipe_write_data_to_child) == 0) {
if (!server.aof_stop_sending_diff &&
aeGetFileEvents(server.el,server.aof_pipe_write_data_to_child) == 0)
{
aeCreateFileEvent(server.el, server.aof_pipe_write_data_to_child,
AE_WRITABLE, aofChildWriteDiffData, NULL);
}
+16 -16
View File
@@ -37,8 +37,8 @@
/* Count number of bits set in the binary array pointed by 's' and long
* 'count' bytes. The implementation of this function is required to
* work with an input string length up to 512 MB or more (server.proto_max_bulk_len) */
size_t redisPopcount(void *s, long count) {
size_t bits = 0;
long long redisPopcount(void *s, long count) {
long long bits = 0;
unsigned char *p = s;
uint32_t *p4;
static const unsigned char bitsinbyte[256] = {0,1,1,2,1,2,2,3,1,2,2,3,2,3,3,4,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,1,2,2,3,2,3,3,4,2,3,3,4,3,4,4,5,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,2,3,3,4,3,4,4,5,3,4,4,5,4,5,5,6,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,3,4,4,5,4,5,5,6,4,5,5,6,5,6,6,7,4,5,5,6,5,6,6,7,5,6,6,7,6,7,7,8};
@@ -98,11 +98,11 @@ size_t redisPopcount(void *s, long count) {
* no zero bit is found, it returns count*8 assuming the string is zero
* padded on the right. However if 'bit' is 1 it is possible that there is
* not a single set bit in the bitmap. In this special case -1 is returned. */
long redisBitpos(void *s, unsigned long count, int bit) {
long long redisBitpos(void *s, unsigned long count, int bit) {
unsigned long *l;
unsigned char *c;
unsigned long skipval, word = 0, one;
long pos = 0; /* Position of bit, to return to the caller. */
long long pos = 0; /* Position of bit, to return to the caller. */
unsigned long j;
int found;
@@ -410,7 +410,7 @@ void printBits(unsigned char *p, unsigned long count) {
* If the 'hash' argument is true, and 'bits is positive, then the command
* will also parse bit offsets prefixed by "#". In such a case the offset
* is multiplied by 'bits'. This is useful for the BITFIELD command. */
int getBitOffsetFromArgument(client *c, robj *o, size_t *offset, int hash, int bits) {
int getBitOffsetFromArgument(client *c, robj *o, uint64_t *offset, int hash, int bits) {
long long loffset;
char *err = "bit offset is not an integer or out of range";
char *p = o->ptr;
@@ -435,7 +435,7 @@ int getBitOffsetFromArgument(client *c, robj *o, size_t *offset, int hash, int b
return C_ERR;
}
*offset = (size_t)loffset;
*offset = loffset;
return C_OK;
}
@@ -477,7 +477,7 @@ int getBitfieldTypeFromArgument(client *c, robj *o, int *sign, int *bits) {
* so that the 'maxbit' bit can be addressed. The object is finally
* returned. Otherwise if the key holds a wrong type NULL is returned and
* an error is sent to the client. */
robj *lookupStringForBitCommand(client *c, size_t maxbit) {
robj *lookupStringForBitCommand(client *c, uint64_t maxbit) {
size_t byte = maxbit >> 3;
robj *o = lookupKeyWrite(c->db,c->argv[1]);
if (checkType(c,o,OBJ_STRING)) return NULL;
@@ -527,7 +527,7 @@ unsigned char *getObjectReadOnlyString(robj *o, long *len, char *llbuf) {
void setbitCommand(client *c) {
robj *o;
char *err = "bit is not an integer or out of range";
size_t bitoffset;
uint64_t bitoffset;
ssize_t byte, bit;
int byteval, bitval;
long on;
@@ -566,7 +566,7 @@ void setbitCommand(client *c) {
void getbitCommand(client *c) {
robj *o;
char llbuf[32];
size_t bitoffset;
uint64_t bitoffset;
size_t byte, bit;
size_t bitval = 0;
@@ -888,7 +888,7 @@ void bitposCommand(client *c) {
addReplyLongLong(c, -1);
} else {
long bytes = end-start+1;
long pos = redisBitpos(p+start,bytes,bit);
long long pos = redisBitpos(p+start,bytes,bit);
/* If we are looking for clear bits, and the user specified an exact
* range with start-end, we can't consider the right of the range as
@@ -897,11 +897,11 @@ void bitposCommand(client *c) {
* So if redisBitpos() returns the first bit outside the range,
* we return -1 to the caller, to mean, in the specified range there
* is not a single "0" bit. */
if (end_given && bit == 0 && pos == bytes*8) {
if (end_given && bit == 0 && pos == (long long)bytes<<3) {
addReplyLongLong(c,-1);
return;
}
if (pos != -1) pos += start*8; /* Adjust for the bytes we skipped. */
if (pos != -1) pos += (long long)start<<3; /* Adjust for the bytes we skipped. */
addReplyLongLong(c,pos);
}
}
@@ -933,12 +933,12 @@ struct bitfieldOp {
* GET subcommand is allowed, other subcommands will return an error. */
void bitfieldGeneric(client *c, int flags) {
robj *o;
size_t bitoffset;
uint64_t bitoffset;
int j, numops = 0, changes = 0;
struct bitfieldOp *ops = NULL; /* Array of ops to execute at end. */
int owtype = BFOVERFLOW_WRAP; /* Overflow type. */
int readonly = 1;
size_t highest_write_offset = 0;
uint64_t highest_write_offset = 0;
for (j = 2; j < c->argc; j++) {
int remargs = c->argc-j-1; /* Remaining args other than current. */
@@ -1128,9 +1128,9 @@ void bitfieldGeneric(client *c, int flags) {
* object boundaries. */
memset(buf,0,9);
int i;
size_t byte = thisop->offset >> 3;
uint64_t byte = thisop->offset >> 3;
for (i = 0; i < 9; i++) {
if (src == NULL || i+byte >= (size_t)strlen) break;
if (src == NULL || i+byte >= (uint64_t)strlen) break;
buf[i] = src[i+byte];
}
+1 -1
View File
@@ -93,7 +93,7 @@ void sendChildInfoGeneric(childInfoType info_type, size_t keys, double progress,
if (cow) {
serverLog((info_type == CHILD_INFO_TYPE_CURRENT_INFO) ? LL_VERBOSE : LL_NOTICE,
"%s: %zu MB of memory used by copy-on-write",
pname, data.cow / (1024 * 1024));
pname, cow / (1024 * 1024));
}
}
+6 -1
View File
@@ -707,6 +707,7 @@ void clusterAcceptHandler(aeEventLoop *el, int fd, void *privdata, int mask) {
}
connNonBlock(conn);
connEnableTcpNoDelay(conn);
connKeepAlive(conn,server.cluster_node_timeout * 2);
/* Use non-blocking I/O for cluster messages. */
serverLog(LL_VERBOSE,"Accepting cluster node connection from %s:%d", cip, cport);
@@ -5361,13 +5362,16 @@ void migrateCommand(client *c) {
}
j++;
password = c->argv[j]->ptr;
redactClientCommandArgument(c,j);
} else if (!strcasecmp(c->argv[j]->ptr,"auth2")) {
if (moreargs < 2) {
addReplyErrorObject(c,shared.syntaxerr);
return;
}
username = c->argv[++j]->ptr;
redactClientCommandArgument(c,j);
password = c->argv[++j]->ptr;
redactClientCommandArgument(c,j);
} else if (!strcasecmp(c->argv[j]->ptr,"keys")) {
if (sdslen(c->argv[3]->ptr) != 0) {
addReplyError(c,
@@ -5465,9 +5469,10 @@ try_again:
if (ttl < 1) ttl = 1;
}
/* Relocate valid (non expired) keys into the array in successive
/* Relocate valid (non expired) keys and values into the array in successive
* positions to remove holes created by the keys that were present
* in the first lookup but are now expired after the second lookup. */
ov[non_expired] = ov[j];
kv[non_expired++] = kv[j];
serverAssertWithInfo(c,NULL,
+1 -1
View File
@@ -726,7 +726,7 @@ void configSetCommand(client *c) {
(config->alias && !strcasecmp(c->argv[2]->ptr,config->alias))))
{
if (config->flags & SENSITIVE_CONFIG) {
preventCommandLogging(c);
redactClientCommandArgument(c,3);
}
if (!config->interface.set(config->data,o->ptr,1,&errstr)) {
goto badfmt;
+8 -5
View File
@@ -1480,7 +1480,7 @@ int keyIsExpired(redisDb *db, robj *key) {
* script execution, making propagation to slaves / AOF consistent.
* See issue #1525 on Github for more information. */
if (server.lua_caller) {
now = server.lua_time_start;
now = server.lua_time_snapshot;
}
/* If we are in the middle of a command execution, we still want to use
* a reference time that does not change: in that case we just use the
@@ -1541,14 +1541,17 @@ int expireIfNeeded(redisDb *db, robj *key) {
if (checkClientPauseTimeoutAndReturnIfPaused()) return 1;
/* Delete the key */
if (server.lazyfree_lazy_expire) {
dbAsyncDelete(db,key);
} else {
dbSyncDelete(db,key);
}
server.stat_expiredkeys++;
propagateExpire(db,key,server.lazyfree_lazy_expire);
notifyKeyspaceEvent(NOTIFY_EXPIRED,
"expired",key,db->id);
int retval = server.lazyfree_lazy_expire ? dbAsyncDelete(db,key) :
dbSyncDelete(db,key);
if (retval) signalModifiedKey(NULL,db,key);
return retval;
signalModifiedKey(NULL,db,key);
return 1;
}
/* -----------------------------------------------------------------------------
+8 -6
View File
@@ -721,7 +721,7 @@ NULL
} else if (!strcasecmp(name,"double")) {
addReplyDouble(c,3.14159265359);
} else if (!strcasecmp(name,"bignum")) {
addReplyProto(c,"(1234567999999999999999999999999999999\r\n",40);
addReplyBigNum(c,"1234567999999999999999999999999999999",37);
} else if (!strcasecmp(name,"null")) {
addReplyNull(c);
} else if (!strcasecmp(name,"array")) {
@@ -737,11 +737,13 @@ NULL
addReplyBool(c, j == 1);
}
} else if (!strcasecmp(name,"attrib")) {
addReplyAttributeLen(c,1);
addReplyBulkCString(c,"key-popularity");
addReplyArrayLen(c,2);
addReplyBulkCString(c,"key:123");
addReplyLongLong(c,90);
if (c->resp >= 3) {
addReplyAttributeLen(c,1);
addReplyBulkCString(c,"key-popularity");
addReplyArrayLen(c,2);
addReplyBulkCString(c,"key:123");
addReplyLongLong(c,90);
}
/* Attributes are not real replies, so a well formed reply should
* also have a normal reply type after the attribute. */
addReplyBulkCString(c,"Some real reply following the attribute");
+1 -1
View File
@@ -281,7 +281,7 @@ uint32_t intsetLen(const intset *is) {
/* Return intset blob size in bytes. */
size_t intsetBlobLen(intset *is) {
return sizeof(intset)+intrev32ifbe(is->length)*intrev32ifbe(is->encoding);
return sizeof(intset)+(size_t)intrev32ifbe(is->length)*intrev32ifbe(is->encoding);
}
/* Validate the integrity of the data structure.
+3 -4
View File
@@ -39,12 +39,11 @@ void lazyfreeFreeSlotsMap(void *args[]) {
atomicIncr(lazyfreed_objects,len);
}
/* Release the rax mapping Redis Cluster keys to slots in the
* lazyfree thread. */
/* Release the key tracking table. */
void lazyFreeTrackingTable(void *args[]) {
rax *rt = args[0];
size_t len = rt->numele;
raxFree(rt);
freeTrackingRadixTree(rt);
atomicDecr(lazyfree_objects,len);
atomicIncr(lazyfreed_objects,len);
}
@@ -110,7 +109,7 @@ size_t lazyfreeGetFreeEffort(robj *key, robj *obj) {
/* Every consumer group is an allocation and so are the entries in its
* PEL. We use size of the first group's PEL as an estimate for all
* others. */
if (s->cgroups) {
if (s->cgroups && raxSize(s->cgroups)) {
raxIterator ri;
streamCG *cg;
raxStart(&ri,s->cgroups);
+2 -2
View File
@@ -94,8 +94,8 @@ lwCanvas *lwCreateCanvas(int width, int height, int bgcolor) {
lwCanvas *canvas = zmalloc(sizeof(*canvas));
canvas->width = width;
canvas->height = height;
canvas->pixels = zmalloc(width*height);
memset(canvas->pixels,bgcolor,width*height);
canvas->pixels = zmalloc((size_t)width*height);
memset(canvas->pixels,bgcolor,(size_t)width*height);
return canvas;
}
+1 -1
View File
@@ -71,7 +71,7 @@ void memtest_progress_start(char *title, int pass) {
printf("\x1b[H\x1b[2K"); /* Cursor home, clear current line. */
printf("%s [%d]\n", title, pass); /* Print title. */
progress_printed = 0;
progress_full = ws.ws_col*(ws.ws_row-3);
progress_full = (size_t)ws.ws_col*(ws.ws_row-3);
fflush(stdout);
}
+33 -5
View File
@@ -2542,7 +2542,7 @@ int RM_StringTruncate(RedisModuleKey *key, size_t newlen) {
if (newlen > curlen) {
key->value->ptr = sdsgrowzero(key->value->ptr,newlen);
} else if (newlen < curlen) {
sdsrange(key->value->ptr,0,newlen-1);
sdssubstr(key->value->ptr,0,newlen);
/* If the string is too wasteful, reallocate it. */
if (sdslen(key->value->ptr) < sdsavail(key->value->ptr))
key->value->ptr = sdsRemoveFreeSpace(key->value->ptr);
@@ -5362,8 +5362,8 @@ int moduleTryServeClientBlockedOnKey(client *c, robj *key) {
* reply_callback: called after a successful RedisModule_UnblockClient()
* call in order to reply to the client and unblock it.
*
* timeout_callback: called when the timeout is reached in order to send an
* error to the client.
* timeout_callback: called when the timeout is reached or if `CLIENT UNBLOCK`
* is invoked, in order to send an error to the client.
*
* free_privdata: called in order to free the private data that is passed
* by RedisModule_UnblockClient() call.
@@ -5380,6 +5380,12 @@ int moduleTryServeClientBlockedOnKey(client *c, robj *key) {
* In these cases, a call to RedisModule_BlockClient() will **not** block the
* client, but instead produce a specific error reply.
*
* A module that registers a timeout_callback function can also be unblocked
* using the `CLIENT UNBLOCK` command, which will trigger the timeout callback.
* If a callback function is not registered, then the blocked client will be
* treated as if it is not in a blocked state and `CLIENT UNBLOCK` will return
* a zero value.
*
* Measuring background time: By default the time spent in the blocked command
* is not account for the total command duration. To include such time you should
* use RM_BlockedClientMeasureTimeStart() and RM_BlockedClientMeasureTimeEnd() one,
@@ -5649,6 +5655,17 @@ void moduleHandleBlockedClients(void) {
pthread_mutex_unlock(&moduleUnblockedClientsMutex);
}
/* Check if the specified client can be safely timed out using
* moduleBlockedClientTimedOut().
*/
int moduleBlockedClientMayTimeout(client *c) {
if (c->btype != BLOCKED_MODULE)
return 1;
RedisModuleBlockedClient *bc = c->bpop.module_blocked_handle;
return (bc && bc->timeout_callback != NULL);
}
/* Called when our client timed out. After this function unblockClient()
* is called, and it will invalidate the blocked client. So this function
* does not need to do any cleanup. Eventually the module will call the
@@ -6168,7 +6185,7 @@ int RM_GetClusterNodeInfo(RedisModuleCtx *ctx, const char *id, char *ip, char *m
/* If the information is not available, the function will set the
* field to zero bytes, so that when the field can't be populated the
* function kinda remains predictable. */
if (node->flags & CLUSTER_NODE_MASTER && node->slaveof)
if (node->flags & CLUSTER_NODE_SLAVE && node->slaveof)
memcpy(master_id,node->slaveof->name,REDISMODULE_NODE_ID_LEN);
else
memset(master_id,0,REDISMODULE_NODE_ID_LEN);
@@ -8679,8 +8696,9 @@ sds genModulesInfoStringRenderModulesList(list *l) {
while((ln = listNext(&li))) {
RedisModule *module = ln->value;
output = sdscat(output,module->name);
if (ln != listLast(l))
output = sdscat(output,"|");
}
output = sdstrim(output,"|");
output = sdscat(output,"]");
return output;
}
@@ -9006,6 +9024,14 @@ int *RM_GetCommandKeys(RedisModuleCtx *ctx, RedisModuleString **argv, int argc,
return res;
}
/* Return the name of the command currently running */
const char *RM_GetCurrentCommandName(RedisModuleCtx *ctx) {
if (!ctx || !ctx->client || !ctx->client->cmd)
return NULL;
return (const char*)ctx->client->cmd->name;
}
/* --------------------------------------------------------------------------
* ## Defrag API
* -------------------------------------------------------------------------- */
@@ -9205,6 +9231,7 @@ long moduleDefragGlobals(void) {
module->defrag_cb(&defrag_ctx);
defragged += defrag_ctx.defragged;
}
dictReleaseIterator(di);
return defragged;
}
@@ -9466,6 +9493,7 @@ void moduleRegisterCoreAPI(void) {
REGISTER_API(GetServerVersion);
REGISTER_API(GetClientCertificate);
REGISTER_API(GetCommandKeys);
REGISTER_API(GetCurrentCommandName);
REGISTER_API(GetTypeMethodVersion);
REGISTER_API(RegisterDefragFunc);
REGISTER_API(DefragAlloc);
+1 -6
View File
@@ -13,7 +13,7 @@ endif
.SUFFIXES: .c .so .xo .o
all: helloworld.so hellotype.so helloblock.so testmodule.so hellocluster.so hellotimer.so hellodict.so hellohook.so helloacl.so
all: helloworld.so hellotype.so helloblock.so hellocluster.so hellotimer.so hellodict.so hellohook.so helloacl.so
.c.xo:
$(CC) -I. $(CFLAGS) $(SHOBJ_CFLAGS) -fPIC -c $< -o $@
@@ -58,10 +58,5 @@ helloacl.xo: ../redismodule.h
helloacl.so: helloacl.xo
$(LD) -o $@ $< $(SHOBJ_LDFLAGS) $(LIBS) -lc
testmodule.xo: ../redismodule.h
testmodule.so: testmodule.xo
$(LD) -o $@ $< $(SHOBJ_LDFLAGS) $(LIBS) -lc
clean:
rm -rf *.xo *.so
+23 -12
View File
@@ -153,8 +153,7 @@ void execCommandAbort(client *c, sds error) {
/* Send EXEC to clients waiting data from MONITOR. We did send a MULTI
* already, and didn't send any of the queued commands, now we'll just send
* EXEC so it is clear that the transaction is over. */
if (listLength(server.monitors) && !server.loading)
replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
}
void execCommand(client *c) {
@@ -169,6 +168,11 @@ void execCommand(client *c) {
return;
}
/* EXEC with expired watched key is disallowed*/
if (isWatchedKeyExpired(c)) {
c->flags |= (CLIENT_DIRTY_CAS);
}
/* Check if we need to abort the EXEC because:
* 1) Some WATCHed key was touched.
* 2) There was a previous error while queueing commands.
@@ -179,7 +183,7 @@ void execCommand(client *c) {
addReply(c, c->flags & CLIENT_DIRTY_EXEC ? shared.execaborterr :
shared.nullarray[c->resp]);
discardTransaction(c);
goto handle_monitor;
return;
}
uint64_t old_flags = c->flags;
@@ -266,15 +270,6 @@ void execCommand(client *c) {
}
server.in_exec = 0;
handle_monitor:
/* Send EXEC to clients waiting data from MONITOR. We do it here
* since the natural order of commands execution is actually:
* MUTLI, EXEC, ... commands inside transaction ...
* Instead EXEC is flagged as CMD_SKIP_MONITOR in the command
* table, and we do it here with correct ordering. */
if (listLength(server.monitors) && !server.loading)
replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
}
/* ===================== WATCH (CAS alike for MULTI/EXEC) ===================
@@ -352,6 +347,22 @@ void unwatchAllKeys(client *c) {
}
}
/* iterates over the watched_keys list and
* look for an expired key . */
int isWatchedKeyExpired(client *c) {
listIter li;
listNode *ln;
watchedKey *wk;
if (listLength(c->watched_keys) == 0) return 0;
listRewind(c->watched_keys,&li);
while ((ln = listNext(&li))) {
wk = listNodeValue(ln);
if (keyIsExpired(wk->db, wk->key)) return 1;
}
return 0;
}
/* "Touch" a key, so that if this key is being WATCHed by some client the
* next EXEC will fail. */
void touchWatchedKey(redisDb *db, robj *key) {
+56 -25
View File
@@ -333,7 +333,7 @@ void _addReplyProtoToList(client *c, const char *s, size_t len) {
listAddNodeTail(c->reply, tail);
c->reply_bytes += tail->size;
asyncCloseClientOnOutputBufferLimitReached(c);
closeClientOnOutputBufferLimitReached(c, 1);
}
}
@@ -616,7 +616,7 @@ void setDeferredReply(client *c, void *node, const char *s, size_t length) {
listNodeValue(ln) = buf;
c->reply_bytes += buf->size;
asyncCloseClientOnOutputBufferLimitReached(c);
closeClientOnOutputBufferLimitReached(c, 1);
}
}
@@ -649,14 +649,13 @@ void setDeferredSetLen(client *c, void *node, long length) {
}
void setDeferredAttributeLen(client *c, void *node, long length) {
int prefix = c->resp == 2 ? '*' : '|';
if (c->resp == 2) length *= 2;
setDeferredAggregateLen(c,node,length,prefix);
serverAssert(c->resp >= 3);
setDeferredAggregateLen(c,node,length,'|');
}
void setDeferredPushLen(client *c, void *node, long length) {
int prefix = c->resp == 2 ? '*' : '>';
setDeferredAggregateLen(c,node,length,prefix);
serverAssert(c->resp >= 3);
setDeferredAggregateLen(c,node,length,'>');
}
/* Add a double as a bulk reply */
@@ -685,6 +684,16 @@ void addReplyDouble(client *c, double d) {
}
}
void addReplyBigNum(client *c, const char* num, size_t len) {
if (c->resp == 2) {
addReplyBulkCBuffer(c, num, len);
} else {
addReplyProto(c,"(",1);
addReplyProto(c,num,len);
addReply(c,shared.crlf);
}
}
/* Add a long double as a bulk reply, but uses a human readable formatting
* of the double instead of exposing the crude behavior of doubles to the
* dear user. */
@@ -756,14 +765,13 @@ void addReplySetLen(client *c, long length) {
}
void addReplyAttributeLen(client *c, long length) {
int prefix = c->resp == 2 ? '*' : '|';
if (c->resp == 2) length *= 2;
addReplyAggregateLen(c,length,prefix);
serverAssert(c->resp >= 3);
addReplyAggregateLen(c,length,'|');
}
void addReplyPushLen(client *c, long length) {
int prefix = c->resp == 2 ? '*' : '>';
addReplyAggregateLen(c,length,prefix);
serverAssert(c->resp >= 3);
addReplyAggregateLen(c,length,'>');
}
void addReplyNull(client *c) {
@@ -949,7 +957,7 @@ void AddReplyFromClient(client *dst, client *src) {
src->bufpos = 0;
/* Check output buffer limits */
asyncCloseClientOnOutputBufferLimitReached(dst);
closeClientOnOutputBufferLimitReached(dst, 1);
}
/* Copy 'src' client output buffers into 'dst' client output buffers.
@@ -1663,9 +1671,6 @@ void resetClient(client *c) {
c->flags |= CLIENT_REPLY_SKIP;
c->flags &= ~CLIENT_REPLY_SKIP_NEXT;
}
/* Always clear the prevent logging field. */
c->flags &= ~CLIENT_PREVENT_LOGGING;
}
/* This function is used when we want to re-enter the event loop but there
@@ -2672,7 +2677,7 @@ NULL
if (getLongLongFromObjectOrReply(c,c->argv[2],&id,NULL)
!= C_OK) return;
struct client *target = lookupClientByID(id);
if (target && target->flags & CLIENT_BLOCKED) {
if (target && target->flags & CLIENT_BLOCKED && moduleBlockedClientMayTimeout(target)) {
if (unblock_error)
addReplyError(target,
"-UNBLOCKED client unblocked via CLIENT UNBLOCK");
@@ -2967,7 +2972,8 @@ void helloCommand(client *c) {
int moreargs = (c->argc-1) - j;
const char *opt = c->argv[j]->ptr;
if (!strcasecmp(opt,"AUTH") && moreargs >= 2) {
preventCommandLogging(c);
redactClientCommandArgument(c, j+1);
redactClientCommandArgument(c, j+2);
if (ACLAuthenticateUser(c, c->argv[j+1], c->argv[j+2]) == C_ERR) {
addReplyError(c,"-WRONGPASS invalid username-password pair or user is disabled.");
return;
@@ -3054,6 +3060,15 @@ static void retainOriginalCommandVector(client *c) {
}
}
/* Redact a given argument to prevent it from being shown
* in the slowlog. This information is stored in the
* original_argv array. */
void redactClientCommandArgument(client *c, int argc) {
retainOriginalCommandVector(c);
decrRefCount(c->argv[argc]);
c->original_argv[argc] = shared.redacted;
}
/* Rewrite the command vector of the client. All the new objects ref count
* is incremented. The old command vector is freed, and the old objects
* ref count is decremented. */
@@ -3223,18 +3238,33 @@ int checkClientOutputBufferLimits(client *c) {
*
* Note: we need to close the client asynchronously because this function is
* called from contexts where the client can't be freed safely, i.e. from the
* lower level functions pushing data inside the client output buffers. */
void asyncCloseClientOnOutputBufferLimitReached(client *c) {
if (!c->conn) return; /* It is unsafe to free fake clients. */
* lower level functions pushing data inside the client output buffers.
* When `async` is set to 0, we close the client immediately, this is
* useful when called from cron.
*
* Returns 1 if client was (flagged) closed. */
int closeClientOnOutputBufferLimitReached(client *c, int async) {
if (!c->conn) return 0; /* It is unsafe to free fake clients. */
serverAssert(c->reply_bytes < SIZE_MAX-(1024*64));
if (c->reply_bytes == 0 || c->flags & CLIENT_CLOSE_ASAP) return;
if (c->reply_bytes == 0 || c->flags & CLIENT_CLOSE_ASAP) return 0;
if (checkClientOutputBufferLimits(c)) {
sds client = catClientInfoString(sdsempty(),c);
freeClientAsync(c);
serverLog(LL_WARNING,"Client %s scheduled to be closed ASAP for overcoming of output buffer limits.", client);
if (async) {
freeClientAsync(c);
serverLog(LL_WARNING,
"Client %s scheduled to be closed ASAP for overcoming of output buffer limits.",
client);
} else {
freeClient(c);
serverLog(LL_WARNING,
"Client %s closed for overcoming of output buffer limits.",
client);
}
sdsfree(client);
return 1;
}
return 0;
}
/* Helper function used by performEvictions() in order to flush slaves
@@ -3632,7 +3662,7 @@ int postponeClientRead(client *c) {
if (server.io_threads_active &&
server.io_threads_do_reads &&
!ProcessingEventsWhileBlocked &&
!(c->flags & (CLIENT_MASTER|CLIENT_SLAVE|CLIENT_PENDING_READ)))
!(c->flags & (CLIENT_MASTER|CLIENT_SLAVE|CLIENT_PENDING_READ|CLIENT_BLOCKED)))
{
c->flags |= CLIENT_PENDING_READ;
listAddNodeHead(server.clients_pending_read,c);
@@ -3696,6 +3726,7 @@ int handleClientsWithPendingReadsUsingThreads(void) {
c->flags &= ~CLIENT_PENDING_READ;
listDelNode(server.clients_pending_read,ln);
serverAssert(!(c->flags & CLIENT_BLOCKED));
if (processPendingCommandsAndResetClient(c) == C_ERR) {
/* If the client is no longer valid, we avoid
* processing the client later. So we just go
+2 -2
View File
@@ -836,7 +836,7 @@ size_t objectComputeSize(robj *o, size_t sample_size) {
if (samples) asize += (double)elesize/samples*dictSize(d);
} else if (o->encoding == OBJ_ENCODING_INTSET) {
intset *is = o->ptr;
asize = sizeof(*o)+sizeof(*is)+is->encoding*is->length;
asize = sizeof(*o)+sizeof(*is)+(size_t)is->encoding*is->length;
} else {
serverPanic("Unknown set encoding");
}
@@ -881,7 +881,7 @@ size_t objectComputeSize(robj *o, size_t sample_size) {
}
} else if (o->type == OBJ_STREAM) {
stream *s = o->ptr;
asize = sizeof(*o);
asize = sizeof(*o)+sizeof(*s);
asize += streamRadixTreeMemoryUsage(s->rax);
/* Now we have to add the listpacks. The last listpack is often non
+6 -3
View File
@@ -2484,8 +2484,10 @@ int rdbLoadRio(rio *rdb, int rdbflags, rdbSaveInfo *rsi) {
int when_opcode = rdbLoadLen(rdb,NULL);
int when = rdbLoadLen(rdb,NULL);
if (rioGetReadError(rdb)) goto eoferr;
if (when_opcode != RDB_MODULE_OPCODE_UINT)
if (when_opcode != RDB_MODULE_OPCODE_UINT) {
rdbReportReadError("bad when_opcode");
goto eoferr;
}
moduleType *mt = moduleTypeLookupModuleByID(moduleid);
char name[10];
moduleTypeNameByID(name,moduleid);
@@ -2509,7 +2511,7 @@ int rdbLoadRio(rio *rdb, int rdbflags, rdbSaveInfo *rsi) {
if (mt->aux_load(&io,moduleid&1023, when) != REDISMODULE_OK || io.error) {
moduleTypeNameByID(name,moduleid);
serverLog(LL_WARNING,"The RDB file contains module AUX data for the module type '%s', that the responsible module is not able to load. Check for modules log above for additional clues.", name);
exit(1);
goto eoferr;
}
if (io.ctx) {
moduleFreeContext(io.ctx);
@@ -2518,7 +2520,7 @@ int rdbLoadRio(rio *rdb, int rdbflags, rdbSaveInfo *rsi) {
uint64_t eof = rdbLoadLen(rdb,NULL);
if (eof != RDB_MODULE_OPCODE_EOF) {
serverLog(LL_WARNING,"The RDB file contains module AUX data for the module '%s' that is not terminated by the proper module value EOF marker", name);
exit(1);
goto eoferr;
}
continue;
} else {
@@ -2691,6 +2693,7 @@ static void backgroundSaveDoneHandlerSocket(int exitcode, int bysignal) {
}
if (server.rdb_child_exit_pipe!=-1)
close(server.rdb_child_exit_pipe);
aeDeleteFileEvent(server.el, server.rdb_pipe_read, AE_READABLE);
close(server.rdb_pipe_read);
server.rdb_child_exit_pipe = -1;
server.rdb_pipe_read = -1;
+46 -44
View File
@@ -99,7 +99,6 @@ static struct config {
int randomkeys_keyspacelen;
int keepalive;
int pipeline;
int showerrors;
long long start;
long long totlatency;
const char *title;
@@ -307,7 +306,9 @@ static redisContext *getRedisContext(const char *ip, int port,
fprintf(stderr, "Node %s:%d replied with error:\n%s\n", ip, port, reply->str);
else
fprintf(stderr, "Node %s replied with error:\n%s\n", hostsocket, reply->str);
goto cleanup;
freeReplyObject(reply);
redisFree(ctx);
exit(1);
}
freeReplyObject(reply);
return ctx;
@@ -366,9 +367,16 @@ fail:
fprintf(stderr, "ERROR: failed to fetch CONFIG from ");
if (hostsocket == NULL) fprintf(stderr, "%s:%d\n", ip, port);
else fprintf(stderr, "%s\n", hostsocket);
int abort_test = 0;
if (reply && reply->type == REDIS_REPLY_ERROR &&
(!strncmp(reply->str,"NOAUTH",5) ||
!strncmp(reply->str,"WRONGPASS",9) ||
!strncmp(reply->str,"NOPERM",5)))
abort_test = 1;
freeReplyObject(reply);
redisFree(c);
freeRedisConfig(cfg);
if (abort_test) exit(1);
return NULL;
}
static void freeRedisConfig(redisConfig *cfg) {
@@ -513,44 +521,39 @@ static void readHandler(aeEventLoop *el, int fd, void *privdata, int mask) {
exit(1);
}
redisReply *r = reply;
int is_err = (r->type == REDIS_REPLY_ERROR);
if (is_err && config.showerrors) {
/* TODO: static lasterr_time not thread-safe */
static time_t lasterr_time = 0;
time_t now = time(NULL);
if (lasterr_time != now) {
lasterr_time = now;
if (c->cluster_node) {
printf("Error from server %s:%d: %s\n",
if (r->type == REDIS_REPLY_ERROR) {
/* Try to update slots configuration if reply error is
* MOVED/ASK/CLUSTERDOWN and the key(s) used by the command
* contain(s) the slot hash tag.
* If the error is not topology-update related then we
* immediately exit to avoid false results. */
if (c->cluster_node && c->staglen) {
int fetch_slots = 0, do_wait = 0;
if (!strncmp(r->str,"MOVED",5) || !strncmp(r->str,"ASK",3))
fetch_slots = 1;
else if (!strncmp(r->str,"CLUSTERDOWN",11)) {
/* Usually the cluster is able to recover itself after
* a CLUSTERDOWN error, so try to sleep one second
* before requesting the new configuration. */
fetch_slots = 1;
do_wait = 1;
printf("Error from server %s:%d: %s.\n",
c->cluster_node->ip,
c->cluster_node->port,
r->str);
}
if (do_wait) sleep(1);
if (fetch_slots && !fetchClusterSlotsConfiguration(c))
exit(1);
} else {
if (c->cluster_node) {
printf("Error from server %s:%d: %s\n",
c->cluster_node->ip,
c->cluster_node->port,
r->str);
} else printf("Error from server: %s\n", r->str);
}
}
/* Try to update slots configuration if reply error is
* MOVED/ASK/CLUSTERDOWN and the key(s) used by the command
* contain(s) the slot hash tag. */
if (is_err && c->cluster_node && c->staglen) {
int fetch_slots = 0, do_wait = 0;
if (!strncmp(r->str,"MOVED",5) || !strncmp(r->str,"ASK",3))
fetch_slots = 1;
else if (!strncmp(r->str,"CLUSTERDOWN",11)) {
/* Usually the cluster is able to recover itself after
* a CLUSTERDOWN error, so try to sleep one second
* before requesting the new configuration. */
fetch_slots = 1;
do_wait = 1;
printf("Error from server %s:%d: %s\n",
c->cluster_node->ip,
c->cluster_node->port,
r->str);
}
if (do_wait) sleep(1);
if (fetch_slots && !fetchClusterSlotsConfiguration(c))
exit(1);
}
}
freeReplyObject(reply);
@@ -1293,8 +1296,7 @@ static int fetchClusterSlotsConfiguration(client c) {
atomicGetIncr(config.is_fetching_slots, is_fetching_slots, 1);
if (is_fetching_slots) return -1; //TODO: use other codes || errno ?
atomicSet(config.is_fetching_slots, 1);
if (config.showerrors)
printf("Cluster slots configuration changed, fetching new one...\n");
printf("WARNING: Cluster slots configuration changed, fetching new one...\n");
const char *errmsg = "Failed to update cluster slots configuration";
static dictType dtype = {
dictSdsHash, /* hash function */
@@ -1470,7 +1472,8 @@ int parseOptions(int argc, const char **argv) {
} else if (!strcmp(argv[i],"-I")) {
config.idlemode = 1;
} else if (!strcmp(argv[i],"-e")) {
config.showerrors = 1;
printf("WARNING: -e option has been deprecated. "
"We now immediatly exit on error to avoid false results.\n");
} else if (!strcmp(argv[i],"-t")) {
if (lastarg) goto invalid;
/* We get the list of tests to run as a string in the form
@@ -1573,8 +1576,6 @@ usage:
" is executed. Default tests use this to hit random keys in the\n"
" specified range.\n"
" -P <numreq> Pipeline <numreq> requests. Default 1 (no pipeline).\n"
" -e If server replies with errors, show them on stdout.\n"
" (no more than 1 error per second is displayed)\n"
" -q Quiet. Just show query/sec values\n"
" --precision Number of decimal places to display in latency output (default 0)\n"
" --csv Output in CSV format\n"
@@ -1699,7 +1700,6 @@ int main(int argc, const char **argv) {
config.keepalive = 1;
config.datasize = 3;
config.pipeline = 1;
config.showerrors = 0;
config.randomkeys = 0;
config.randomkeys_keyspacelen = 0;
config.quiet = 0;
@@ -1782,8 +1782,9 @@ int main(int argc, const char **argv) {
} else {
config.redis_config =
getRedisConfig(config.hostip, config.hostport, config.hostsocket);
if (config.redis_config == NULL)
if (config.redis_config == NULL) {
fprintf(stderr, "WARN: could not fetch server CONFIG\n");
}
}
if (config.num_threads > 0) {
pthread_mutex_init(&(config.liveclients_mutex), NULL);
@@ -1946,8 +1947,8 @@ int main(int argc, const char **argv) {
}
if (test_is_selected("lrange") || test_is_selected("lrange_500")) {
len = redisFormatCommand(&cmd,"LRANGE mylist%s 0 449",tag);
benchmark("LRANGE_500 (first 450 elements)",cmd,len);
len = redisFormatCommand(&cmd,"LRANGE mylist%s 0 499",tag);
benchmark("LRANGE_500 (first 500 elements)",cmd,len);
free(cmd);
}
@@ -1974,6 +1975,7 @@ int main(int argc, const char **argv) {
if (!config.csv) printf("\n");
} while(config.loop);
zfree(data);
if (config.redis_config != NULL) freeRedisConfig(config.redis_config);
return 0;
+4 -2
View File
@@ -39,12 +39,14 @@
static char error[1044];
static off_t epos;
static long long line = 1;
int consumeNewline(char *buf) {
if (strncmp(buf,"\r\n",2) != 0) {
ERROR("Expected \\r\\n, got: %02x%02x",buf[0],buf[1]);
return 0;
}
line += 1;
return 1;
}
@@ -201,8 +203,8 @@ int redis_check_aof_main(int argc, char **argv) {
off_t pos = process(fp);
off_t diff = size-pos;
printf("AOF analyzed: size=%lld, ok_up_to=%lld, diff=%lld\n",
(long long) size, (long long) pos, (long long) diff);
printf("AOF analyzed: size=%lld, ok_up_to=%lld, ok_up_to_line=%lld, diff=%lld\n",
(long long) size, (long long) pos, line, (long long) diff);
if (diff > 0) {
if (fix) {
char buf[2];
+1 -1
View File
@@ -250,7 +250,7 @@ int redis_check_rdb(char *rdbfilename, FILE *fp) {
rdbstate.doing = RDB_CHECK_DOING_READ_LEN;
if ((dbid = rdbLoadLen(&rdb,NULL)) == RDB_LENERR)
goto eoferr;
rdbCheckInfo("Selecting DB ID %d", dbid);
rdbCheckInfo("Selecting DB ID %llu", (unsigned long long)dbid);
continue; /* Read type again. */
} else if (type == RDB_OPCODE_RESIZEDB) {
/* RESIZEDB: Hint about the size of the keys in the currently
+20 -14
View File
@@ -663,7 +663,7 @@ static void cliOutputHelp(int argc, char **argv) {
help = entry->org;
if (group == -1) {
/* Compare all arguments */
if (argc == entry->argc) {
if (argc <= entry->argc) {
for (j = 0; j < argc; j++) {
if (strcasecmp(argv[j],entry->argv[j]) != 0) break;
}
@@ -844,7 +844,9 @@ static int cliConnect(int flags) {
cliRefreshPrompt();
}
if (config.hostsocket == NULL) {
/* Do not use hostsocket when we got redirected in cluster mode */
if (config.hostsocket == NULL ||
(config.cluster_mode && config.cluster_reissue_command)) {
context = redisConnect(config.hostip,config.hostport);
} else {
context = redisConnectUnix(config.hostsocket);
@@ -1127,6 +1129,7 @@ static sds cliFormatReplyRaw(redisReply *r) {
case REDIS_REPLY_DOUBLE:
out = sdscatprintf(out,"%s",r->str);
break;
case REDIS_REPLY_SET:
case REDIS_REPLY_ARRAY:
case REDIS_REPLY_PUSH:
for (i = 0; i < r->elements; i++) {
@@ -1185,6 +1188,7 @@ static sds cliFormatReplyCSV(redisReply *r) {
out = sdscat(out,r->integer ? "true" : "false");
break;
case REDIS_REPLY_ARRAY:
case REDIS_REPLY_SET:
case REDIS_REPLY_PUSH:
case REDIS_REPLY_MAP: /* CSV has no map type, just output flat list. */
for (i = 0; i < r->elements; i++) {
@@ -1308,6 +1312,7 @@ static int cliReadReply(int output_raw_strings) {
if (output) {
out = cliFormatReply(reply, config.output, output_raw_strings);
fwrite(out,sdslen(out),1,stdout);
fflush(stdout);
sdsfree(out);
}
freeReplyObject(reply);
@@ -1892,6 +1897,7 @@ static void usage(void) {
" --lru-test <keys> Simulate a cache workload with an 80-20 distribution.\n"
" --replica Simulate a replica showing commands received from the master.\n"
" --rdb <filename> Transfer an RDB dump from remote server to local file.\n"
" Use filename of \"-\" to write to stdout.\n"
" --pipe Transfer raw Redis protocol from stdin to server.\n"
" --pipe-timeout <n> In --pipe mode, abort with error if after sending all data.\n"
" no reply is received within <n> seconds.\n"
@@ -5481,7 +5487,7 @@ static void clusterManagerNodeArrayReset(clusterManagerNodeArray *array) {
static void clusterManagerNodeArrayShift(clusterManagerNodeArray *array,
clusterManagerNode **nodeptr)
{
assert(array->nodes < (array->nodes + array->len));
assert(array->len > 0);
/* If the first node to be shifted is not NULL, decrement count. */
if (*array->nodes != NULL) array->count--;
/* Store the first node to be shifted into 'nodeptr'. */
@@ -5494,7 +5500,7 @@ static void clusterManagerNodeArrayShift(clusterManagerNodeArray *array,
static void clusterManagerNodeArrayAdd(clusterManagerNodeArray *array,
clusterManagerNode *node)
{
assert(array->nodes < (array->nodes + array->len));
assert(array->len > 0);
assert(node != NULL);
assert(array->count < array->len);
array->nodes[array->count++] = node;
@@ -6534,9 +6540,9 @@ static int clusterManagerCommandImport(int argc, char **argv) {
}
if (config.cluster_manager_command.flags & CLUSTER_MANAGER_CMD_FLAG_COPY)
strcat(cmdfmt, " %s");
cmdfmt = sdscat(cmdfmt," COPY");
if (config.cluster_manager_command.flags & CLUSTER_MANAGER_CMD_FLAG_REPLACE)
strcat(cmdfmt, " %s");
cmdfmt = sdscat(cmdfmt," REPLACE");
/* Use SCAN to iterate over the keys, migrating to the
* right node as needed. */
@@ -6568,8 +6574,7 @@ static int clusterManagerCommandImport(int argc, char **argv) {
printf("Migrating %s to %s:%d: ", key, target->ip, target->port);
redisReply *r = reconnectingRedisCommand(src_ctx, cmdfmt,
target->ip, target->port,
key, 0, timeout,
"COPY", "REPLACE");
key, 0, timeout);
if (!r || r->type == REDIS_REPLY_ERROR) {
if (r && r->str) {
clusterManagerLogErr("Source %s:%d replied with "
@@ -6871,7 +6876,7 @@ void showLatencyDistSamples(struct distsamples *samples, long long tot) {
printf("\033[38;5;0m"); /* Set foreground color to black. */
for (j = 0; ; j++) {
int coloridx =
ceil((float) samples[j].count / tot * (spectrum_palette_size-1));
ceil((double) samples[j].count / tot * (spectrum_palette_size-1));
int color = spectrum_palette[coloridx];
printf("\033[48;5;%dm%c", (int)color, samples[j].character);
samples[j].count = 0;
@@ -6985,7 +6990,7 @@ static void latencyDistMode(void) {
#define RDB_EOF_MARK_SIZE 40
void sendReplconf(const char* arg1, const char* arg2) {
printf("sending REPLCONF %s %s\n", arg1, arg2);
fprintf(stderr, "sending REPLCONF %s %s\n", arg1, arg2);
redisReply *reply = redisCommand(context, "REPLCONF %s %s", arg1, arg2);
/* Handle any error conditions */
@@ -7045,7 +7050,7 @@ unsigned long long sendSync(redisContext *c, char *out_eof) {
}
*p = '\0';
if (buf[0] == '-') {
printf("SYNC with master failed: %s\n", buf);
fprintf(stderr, "SYNC with master failed: %s\n", buf);
exit(1);
}
if (strncmp(buf+1,"EOF:",4) == 0 && strlen(buf+5) >= RDB_EOF_MARK_SIZE) {
@@ -7150,8 +7155,9 @@ static void getRDB(clusterManagerNode *node) {
payload, filename);
}
int write_to_stdout = !strcmp(filename,"-");
/* Write to file. */
if (!strcmp(filename,"-")) {
if (write_to_stdout) {
fd = STDOUT_FILENO;
} else {
fd = open(filename, O_CREAT|O_WRONLY, 0644);
@@ -7193,7 +7199,7 @@ static void getRDB(clusterManagerNode *node) {
}
if (usemark) {
payload = ULLONG_MAX - payload - RDB_EOF_MARK_SIZE;
if (ftruncate(fd, payload) == -1)
if (!write_to_stdout && ftruncate(fd, payload) == -1)
fprintf(stderr,"ftruncate failed: %s.\n", strerror(errno));
fprintf(stderr,"Transfer finished with success after %llu bytes\n", payload);
} else {
@@ -7202,7 +7208,7 @@ static void getRDB(clusterManagerNode *node) {
redisFree(s); /* Close the connection ASAP as fsync() may take time. */
if (node)
node->context = NULL;
if (fsync(fd) == -1) {
if (!write_to_stdout && fsync(fd) == -1) {
fprintf(stderr,"Fail to fsync '%s': %s\n", filename, strerror(errno));
exit(1);
}
+2
View File
@@ -836,6 +836,7 @@ REDISMODULE_API int (*RedisModule_AuthenticateClientWithUser)(RedisModuleCtx *ct
REDISMODULE_API int (*RedisModule_DeauthenticateAndCloseClient)(RedisModuleCtx *ctx, uint64_t client_id) REDISMODULE_ATTR;
REDISMODULE_API RedisModuleString * (*RedisModule_GetClientCertificate)(RedisModuleCtx *ctx, uint64_t id) REDISMODULE_ATTR;
REDISMODULE_API int *(*RedisModule_GetCommandKeys)(RedisModuleCtx *ctx, RedisModuleString **argv, int argc, int *num_keys) REDISMODULE_ATTR;
REDISMODULE_API const char *(*RedisModule_GetCurrentCommandName)(RedisModuleCtx *ctx) REDISMODULE_ATTR;
REDISMODULE_API int (*RedisModule_RegisterDefragFunc)(RedisModuleCtx *ctx, RedisModuleDefragFunc func) REDISMODULE_ATTR;
REDISMODULE_API void *(*RedisModule_DefragAlloc)(RedisModuleDefragCtx *ctx, void *ptr) REDISMODULE_ATTR;
REDISMODULE_API RedisModuleString *(*RedisModule_DefragRedisModuleString)(RedisModuleDefragCtx *ctx, RedisModuleString *str) REDISMODULE_ATTR;
@@ -1108,6 +1109,7 @@ static int RedisModule_Init(RedisModuleCtx *ctx, const char *name, int ver, int
REDISMODULE_GET_API(AuthenticateClientWithUser);
REDISMODULE_GET_API(GetClientCertificate);
REDISMODULE_GET_API(GetCommandKeys);
REDISMODULE_GET_API(GetCurrentCommandName);
REDISMODULE_GET_API(RegisterDefragFunc);
REDISMODULE_GET_API(DefragAlloc);
REDISMODULE_GET_API(DefragRedisModuleString);
+1
View File
@@ -377,6 +377,7 @@ void replicationFeedSlavesFromMasterStream(list *slaves, char *buf, size_t bufle
}
void replicationFeedMonitors(client *c, list *monitors, int dictid, robj **argv, int argc) {
if (!(listLength(server.monitors) && !server.loading)) return;
listNode *ln;
listIter li;
int j;
+13 -4
View File
@@ -31,6 +31,7 @@
#include "sha1.h"
#include "rand.h"
#include "cluster.h"
#include "monotonic.h"
#include <lua.h>
#include <lauxlib.h>
@@ -1427,7 +1428,7 @@ sds luaCreateFunction(client *c, lua_State *lua, robj *body) {
/* This is the Lua script "count" hook that we use to detect scripts timeout. */
void luaMaskCountHook(lua_State *lua, lua_Debug *ar) {
long long elapsed = mstime() - server.lua_time_start;
long long elapsed = elapsedMs(server.lua_time_start);
UNUSED(ar);
UNUSED(lua);
@@ -1578,7 +1579,8 @@ void evalGenericCommand(client *c, int evalsha) {
server.in_eval = 1;
server.lua_caller = c;
server.lua_cur_script = funcname + 2;
server.lua_time_start = mstime();
server.lua_time_start = getMonotonicUs();
server.lua_time_snapshot = mstime();
server.lua_kill = 0;
if (server.lua_time_limit > 0 && ldb.active == 0) {
lua_sethook(lua,luaMaskCountHook,LUA_MASKCOUNT,100000);
@@ -1688,6 +1690,9 @@ void evalGenericCommand(client *c, int evalsha) {
}
void evalCommand(client *c) {
/* Explicitly feed monitor here so that lua commands appear after their
* script command. */
replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
if (!(c->flags & CLIENT_LUA_DEBUG))
evalGenericCommand(c,0);
else
@@ -1695,6 +1700,9 @@ void evalCommand(client *c) {
}
void evalShaCommand(client *c) {
/* Explicitly feed monitor here so that lua commands appear after their
* script command. */
replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
if (sdslen(c->argv[1]->ptr) != 40) {
/* We know that a match is not possible if the provided SHA is
* not the right length. So we return an error ASAP, this way
@@ -2729,7 +2737,7 @@ void luaLdbLineHook(lua_State *lua, lua_Debug *ar) {
/* Check if a timeout occurred. */
if (ar->event == LUA_HOOKCOUNT && ldb.step == 0 && bp == 0) {
mstime_t elapsed = mstime() - server.lua_time_start;
mstime_t elapsed = elapsedMs(server.lua_time_start);
mstime_t timelimit = server.lua_time_limit ?
server.lua_time_limit : 5000;
if (elapsed >= timelimit) {
@@ -2759,6 +2767,7 @@ void luaLdbLineHook(lua_State *lua, lua_Debug *ar) {
lua_pushstring(lua, "timeout during Lua debugging with client closing connection");
lua_error(lua);
}
server.lua_time_start = mstime();
server.lua_time_start = getMonotonicUs();
server.lua_time_snapshot = mstime();
}
}
+37 -20
View File
@@ -759,6 +759,21 @@ sds sdstrim(sds s, const char *cset) {
return s;
}
/* Changes the input string to be a subset of the original.
* It does not release the free space in the string, so a call to
* sdsRemoveFreeSpace may be wise after. */
void sdssubstr(sds s, size_t start, size_t len) {
/* Clamp out of range input */
size_t oldlen = sdslen(s);
if (start >= oldlen) start = len = 0;
if (len > oldlen-start) len = oldlen-start;
/* Move the data */
if (len) memmove(s, s+start, len);
s[len] = 0;
sdssetlen(s,len);
}
/* Turn the string into a smaller (or equal) string containing only the
* substring specified by the 'start' and 'end' indexes.
*
@@ -770,6 +785,11 @@ sds sdstrim(sds s, const char *cset) {
*
* The string is modified in-place.
*
* NOTE: this function can be misleading and can have unexpected behaviour,
* specifically when you want the length of the new string to be 0.
* Having start==end will result in a string with one character.
* please consider using sdssubstr instead.
*
* Example:
*
* s = sdsnew("Hello World");
@@ -777,28 +797,13 @@ sds sdstrim(sds s, const char *cset) {
*/
void sdsrange(sds s, ssize_t start, ssize_t end) {
size_t newlen, len = sdslen(s);
if (len == 0) return;
if (start < 0) {
start = len+start;
if (start < 0) start = 0;
}
if (end < 0) {
end = len+end;
if (end < 0) end = 0;
}
if (start < 0)
start = len + start;
if (end < 0)
end = len + end;
newlen = (start > end) ? 0 : (end-start)+1;
if (newlen != 0) {
if (start >= (ssize_t)len) {
newlen = 0;
} else if (end >= (ssize_t)len) {
end = len-1;
newlen = (start > end) ? 0 : (end-start)+1;
}
}
if (start && newlen) memmove(s, s+start, newlen);
s[newlen] = 0;
sdssetlen(s,newlen);
sdssubstr(s, start, newlen);
}
/* Apply tolower() to every character of the sds string 's'. */
@@ -1353,6 +1358,18 @@ int sdsTest(int argc, char **argv, int accurate) {
test_cond("sdsrange(...,100,100)",
sdslen(y) == 0 && memcmp(y,"\0",1) == 0);
sdsfree(y);
y = sdsdup(x);
sdsrange(y,4,6);
test_cond("sdsrange(...,4,6)",
sdslen(y) == 0 && memcmp(y,"\0",1) == 0);
sdsfree(y);
y = sdsdup(x);
sdsrange(y,3,6);
test_cond("sdsrange(...,3,6)",
sdslen(y) == 1 && memcmp(y,"o\0",2) == 0);
sdsfree(y);
sdsfree(x);
x = sdsnew("foo");
+1
View File
@@ -238,6 +238,7 @@ sds sdscatprintf(sds s, const char *fmt, ...);
sds sdscatfmt(sds s, char const *fmt, ...);
sds sdstrim(sds s, const char *cset);
void sdssubstr(sds s, size_t start, size_t len);
void sdsrange(sds s, ssize_t start, ssize_t end);
void sdsupdatelen(sds s);
void sdsclear(sds s);
+10 -8
View File
@@ -3175,11 +3175,13 @@ void sentinelConfigSetCommand(client *c) {
sentinel.announce_port = numval;
} else if (!strcasecmp(o->ptr, "sentinel-user")) {
sdsfree(sentinel.sentinel_auth_user);
sentinel.sentinel_auth_user = sdsnew(val->ptr);
sentinel.sentinel_auth_user = sdslen(val->ptr) == 0 ?
NULL : sdsdup(val->ptr);
drop_conns = 1;
} else if (!strcasecmp(o->ptr, "sentinel-pass")) {
sdsfree(sentinel.sentinel_auth_pass);
sentinel.sentinel_auth_pass = sdsnew(val->ptr);
sentinel.sentinel_auth_pass = sdslen(val->ptr) == 0 ?
NULL : sdsdup(val->ptr);
drop_conns = 1;
} else {
addReplyErrorFormat(c, "Invalid argument '%s' to SENTINEL CONFIG SET",
@@ -4119,16 +4121,16 @@ void sentinelSetCommand(client *c) {
int numargs = j-old_j+1;
switch(numargs) {
case 2:
sentinelEvent(LL_WARNING,"+set",ri,"%@ %s %s",c->argv[old_j]->ptr,
c->argv[old_j+1]->ptr);
sentinelEvent(LL_WARNING,"+set",ri,"%@ %s %s",(char*)c->argv[old_j]->ptr,
(char*)c->argv[old_j+1]->ptr);
break;
case 3:
sentinelEvent(LL_WARNING,"+set",ri,"%@ %s %s %s",c->argv[old_j]->ptr,
c->argv[old_j+1]->ptr,
c->argv[old_j+2]->ptr);
sentinelEvent(LL_WARNING,"+set",ri,"%@ %s %s %s",(char*)c->argv[old_j]->ptr,
(char*)c->argv[old_j+1]->ptr,
(char*)c->argv[old_j+2]->ptr);
break;
default:
sentinelEvent(LL_WARNING,"+set",ri,"%@ %s",c->argv[old_j]->ptr);
sentinelEvent(LL_WARNING,"+set",ri,"%@ %s",(char*)c->argv[old_j]->ptr);
break;
}
}
+39 -26
View File
@@ -706,7 +706,7 @@ struct redisCommand redisCommandTable[] = {
0,NULL,0,0,0,0,0,0},
{"auth",authCommand,-2,
"no-auth no-script ok-loading ok-stale fast no-monitor no-slowlog @connection",
"no-auth no-script ok-loading ok-stale fast @connection",
0,NULL,0,0,0,0,0,0},
/* We don't allow PING during loading since in Redis PING is used as
@@ -749,7 +749,7 @@ struct redisCommand redisCommandTable[] = {
0,NULL,0,0,0,0,0,0},
{"exec",execCommand,1,
"no-script no-monitor no-slowlog ok-loading ok-stale @transaction",
"no-script no-slowlog ok-loading ok-stale @transaction",
0,NULL,0,0,0,0,0,0},
{"discard",discardCommand,1,
@@ -901,17 +901,21 @@ struct redisCommand redisCommandTable[] = {
0,NULL,0,0,0,0,0,0},
{"hello",helloCommand,-1,
"no-auth no-script fast no-monitor ok-loading ok-stale @connection",
"no-auth no-script fast ok-loading ok-stale @connection",
0,NULL,0,0,0,0,0,0},
/* EVAL can modify the dataset, however it is not flagged as a write
* command since we do the check while running commands from Lua. */
* command since we do the check while running commands from Lua.
*
* EVAL and EVALSHA also feed monitors before the commands are executed,
* as opposed to after.
*/
{"eval",evalCommand,-3,
"no-script may-replicate @scripting",
"no-script no-monitor may-replicate @scripting",
0,evalGetKeys,0,0,0,0,0,0},
{"evalsha",evalShaCommand,-3,
"no-script may-replicate @scripting",
"no-script no-monitor may-replicate @scripting",
0,evalGetKeys,0,0,0,0,0,0},
{"slowlog",slowlogCommand,-2,
@@ -1839,6 +1843,7 @@ void clientsCron(void) {
if (clientsCronResizeQueryBuffer(c)) continue;
if (clientsCronTrackExpansiveClients(c, curr_peak_mem_usage_slot)) continue;
if (clientsCronTrackClientsMemUsage(c)) continue;
if (closeClientOnOutputBufferLimitReached(c, 0)) continue;
}
}
@@ -2603,6 +2608,7 @@ void createSharedObjects(void) {
shared.getack = createStringObject("GETACK",6);
shared.special_asterick = createStringObject("*",1);
shared.special_equals = createStringObject("=",1);
shared.redacted = makeObjectShared(createStringObject("(redacted)",10));
for (j = 0; j < OBJ_SHARED_INTEGERS; j++) {
shared.integers[j] =
@@ -3624,12 +3630,6 @@ void preventCommandPropagation(client *c) {
c->flags |= CLIENT_PREVENT_PROP;
}
/* Avoid logging any information about this client's arguments
* since they contain sensitive information. */
void preventCommandLogging(client *c) {
c->flags |= CLIENT_PREVENT_LOGGING;
}
/* AOF specific version of preventCommandPropagation(). */
void preventCommandAOF(client *c) {
c->flags |= CLIENT_PREVENT_AOF_PROP;
@@ -3643,7 +3643,7 @@ void preventCommandReplication(client *c) {
/* Log the last command a client executed into the slowlog. */
void slowlogPushCurrentCommand(client *c, struct redisCommand *cmd, ustime_t duration) {
/* Some commands may contain sensitive data that should not be available in the slowlog. */
if ((c->flags & CLIENT_PREVENT_LOGGING) || (cmd->flags & CMD_SKIP_SLOWLOG))
if (cmd->flags & CMD_SKIP_SLOWLOG)
return;
/* If command argument vector was rewritten, use the original
@@ -3697,17 +3697,6 @@ void call(client *c, int flags) {
struct redisCommand *real_cmd = c->cmd;
static long long prev_err_count;
server.fixed_time_expire++;
/* Send the command to clients in MONITOR mode if applicable.
* Administrative commands are considered too dangerous to be shown. */
if (listLength(server.monitors) &&
!server.loading &&
!(c->cmd->flags & (CMD_SKIP_MONITOR|CMD_ADMIN)))
{
replicationFeedMonitors(c,server.monitors,c->db->id,c->argv,c->argc);
}
/* Initialization: clear the flags that must be set by the command on
* demand, and initialize the array for additional commands propagation. */
c->flags &= ~(CLIENT_FORCE_AOF|CLIENT_FORCE_REPL|CLIENT_PREVENT_PROP);
@@ -3717,7 +3706,13 @@ void call(client *c, int flags) {
/* Call the command. */
dirty = server.dirty;
prev_err_count = server.stat_total_error_replies;
updateCachedTime(0);
/* Update cache time, in case we have nested calls we want to
* update only on the first call*/
if (server.fixed_time_expire++ == 0) {
updateCachedTime(0);
}
elapsedStart(&call_timer);
c->cmd->proc(c);
const long duration = elapsedUs(call_timer);
@@ -3773,6 +3768,14 @@ void call(client *c, int flags) {
if ((flags & CMD_CALL_SLOWLOG) && !(c->flags & CLIENT_BLOCKED))
slowlogPushCurrentCommand(c, real_cmd, duration);
/* Send the command to clients in MONITOR mode if applicable.
* Administrative commands are considered too dangerous to be shown. */
if (!(c->cmd->flags & (CMD_SKIP_MONITOR|CMD_ADMIN))) {
robj **argv = c->original_argv ? c->original_argv : c->argv;
int argc = c->original_argv ? c->original_argc : c->argc;
replicationFeedMonitors(c,server.monitors,c->db->id,argv,argc);
}
/* Clear the original argv.
* If the client is blocked we will handle slowlog when it is unblocked. */
if (!(c->flags & CLIENT_BLOCKED))
@@ -3985,6 +3988,8 @@ int processCommand(client *c) {
return C_OK;
}
int is_read_command = (c->cmd->flags & CMD_READONLY) ||
(c->cmd->proc == execCommand && (c->mstate.cmd_flags & CMD_READONLY));
int is_write_command = (c->cmd->flags & CMD_WRITE) ||
(c->cmd->proc == execCommand && (c->mstate.cmd_flags & CMD_WRITE));
int is_denyoom_command = (c->cmd->flags & CMD_DENYOOM) ||
@@ -4194,7 +4199,7 @@ int processCommand(client *c) {
c->cmd->proc != discardCommand &&
c->cmd->proc != watchCommand &&
c->cmd->proc != unwatchCommand &&
c->cmd->proc != resetCommand &&
c->cmd->proc != resetCommand &&
!(c->cmd->proc == shutdownCommand &&
c->argc == 2 &&
tolower(((char*)c->argv[1]->ptr)[0]) == 'n') &&
@@ -4206,6 +4211,14 @@ int processCommand(client *c) {
return C_OK;
}
/* Prevent a replica from sending commands that access the keyspace.
* The main objective here is to prevent abuse of client pause check
* from which replicas are exempt. */
if ((c->flags & CLIENT_SLAVE) && (is_may_replicate_command || is_write_command || is_read_command)) {
rejectCommandFormat(c, "Replica can't interract with the keyspace");
return C_OK;
}
/* If the server is paused, block the client until
* the pause has ended. Replicas are never paused. */
if (!(c->flags & CLIENT_SLAVE) &&
+11 -6
View File
@@ -279,7 +279,6 @@ extern int configOOMScoreAdjValuesDefaults[CONFIG_OOM_COUNT];
and AOF client */
#define CLIENT_REPL_RDBONLY (1ULL<<42) /* This client is a replica that only wants
RDB without replication buffer. */
#define CLIENT_PREVENT_LOGGING (1ULL<<43) /* Prevent logging of command to slowlog */
/* Client block type (btype field in client structure)
* if CLIENT_BLOCKED flag is set. */
@@ -986,7 +985,7 @@ struct sharedObjectsStruct {
*script, *replconf, *eval, *persist, *set, *pexpireat, *pexpire,
*time, *pxat, *px, *retrycount, *force, *justid,
*lastid, *ping, *setid, *keepttl, *load, *createconsumer,
*getack, *special_asterick, *special_equals, *default_username,
*getack, *special_asterick, *special_equals, *default_username, *redacted,
*select[PROTO_SHARED_SELECT_CMDS],
*integers[OBJ_SHARED_INTEGERS],
*mbulkhdr[OBJ_SHARED_BULKHDR_LEN], /* "*<value>\r\n" */
@@ -1571,7 +1570,8 @@ struct redisServer {
dict *lua_scripts; /* A dictionary of SHA1 -> Lua scripts */
unsigned long long lua_scripts_mem; /* Cached scripts' memory + oh */
mstime_t lua_time_limit; /* Script timeout in milliseconds */
mstime_t lua_time_start; /* Start time of script, milliseconds time */
monotime lua_time_start; /* monotonic timer to detect timed-out script */
mstime_t lua_time_snapshot; /* Snapshot of mstime when script is started */
int lua_write_dirty; /* True if a write command was called during the
execution of the current script. */
int lua_random_dirty; /* True if a random command was called during the
@@ -1779,6 +1779,7 @@ void moduleFireServerEvent(uint64_t eid, int subid, void *data);
void processModuleLoadingProgressEvent(int is_aof);
int moduleTryServeClientBlockedOnKey(client *c, robj *key);
void moduleUnblockClient(client *c);
int moduleBlockedClientMayTimeout(client *c);
int moduleClientIsBlockedOnKeys(client *c);
void moduleNotifyUserChanged(client *c);
void moduleNotifyKeyUnlink(robj *key, robj *val);
@@ -1794,7 +1795,7 @@ void getRandomHexChars(char *p, size_t len);
void getRandomBytes(unsigned char *p, size_t len);
uint64_t crc64(uint64_t crc, const unsigned char *s, uint64_t l);
void exitFromChild(int retcode);
size_t redisPopcount(void *s, long count);
long long redisPopcount(void *s, long count);
int redisSetProcTitle(char *title);
int validateProcTitleTemplate(const char *template);
int redisCommunicateSystemd(const char *sd_notify_msg);
@@ -1840,6 +1841,7 @@ void addReplyErrorSds(client *c, sds err);
void addReplyError(client *c, const char *err);
void addReplyStatus(client *c, const char *status);
void addReplyDouble(client *c, double d);
void addReplyBigNum(client *c, const char* num, size_t len);
void addReplyHumanLongDouble(client *c, long double d);
void addReplyLongLong(client *c, long long ll);
void addReplyArrayLen(client *c, long length);
@@ -1864,9 +1866,10 @@ sds getAllClientsInfoString(int type);
void rewriteClientCommandVector(client *c, int argc, ...);
void rewriteClientCommandArgument(client *c, int i, robj *newval);
void replaceClientCommandVector(client *c, int argc, robj **argv);
void redactClientCommandArgument(client *c, int argc);
unsigned long getClientOutputBufferMemoryUsage(client *c);
int freeClientsInAsyncFreeQueue(void);
void asyncCloseClientOnOutputBufferLimitReached(client *c);
int closeClientOnOutputBufferLimitReached(client *c, int async);
int getClientType(client *c);
int getClientTypeByName(char *name);
char *getClientTypeName(int class);
@@ -1911,6 +1914,7 @@ void disableTracking(client *c);
void trackingRememberKeys(client *c);
void trackingInvalidateKey(client *c, robj *keyobj);
void trackingInvalidateKeysOnFlush(int async);
void freeTrackingRadixTree(rax *rt);
void freeTrackingRadixTreeAsync(rax *rt);
void trackingLimitUsedSlots(void);
uint64_t trackingGetTotalItems(void);
@@ -1943,6 +1947,7 @@ void initClientMultiState(client *c);
void freeClientMultiState(client *c);
void queueMultiCommand(client *c);
void touchWatchedKey(redisDb *db, robj *key);
int isWatchedKeyExpired(client *c);
void touchAllWatchedKeysInDb(redisDb *emptied, redisDb *replaced_with);
void discardTransaction(client *c);
void flagTransaction(client *c);
@@ -2211,7 +2216,6 @@ void redisOpArrayInit(redisOpArray *oa);
void redisOpArrayFree(redisOpArray *oa);
void forceCommandPropagation(client *c, int flags);
void preventCommandPropagation(client *c);
void preventCommandLogging(client *c);
void preventCommandAOF(client *c);
void preventCommandReplication(client *c);
void slowlogPushCurrentCommand(client *c, struct redisCommand *cmd, ustime_t duration);
@@ -2317,6 +2321,7 @@ void initConfigValues();
/* db.c -- Keyspace access API */
int removeExpire(redisDb *db, robj *key);
void propagateExpire(redisDb *db, robj *key, int lazy);
int keyIsExpired(redisDb *db, robj *key);
int expireIfNeeded(redisDb *db, robj *key);
long long getExpire(redisDb *db, robj *key);
void setExpire(client *c, redisDb *db, robj *key, long long when);
+2 -2
View File
@@ -986,7 +986,7 @@ void hrandfieldWithCountCommand(client *c, long l, int withvalues) {
int uniq = 1;
robj *hash;
if ((hash = lookupKeyReadOrReply(c,c->argv[1],shared.null[c->resp]))
if ((hash = lookupKeyReadOrReply(c,c->argv[1],shared.emptyarray))
== NULL || checkType(c,hash,OBJ_HASH)) return;
size = hashTypeLength(hash);
@@ -1175,7 +1175,7 @@ void hrandfieldWithCountCommand(client *c, long l, int withvalues) {
}
}
/* HRANDFIELD [<count> WITHVALUES] */
/* HRANDFIELD key [<count> [WITHVALUES]] */
void hrandfieldCommand(client *c) {
long l;
int withvalues = 0;
+30 -13
View File
@@ -392,12 +392,12 @@ void smoveCommand(client *c) {
}
signalModifiedKey(c,c->db,c->argv[1]);
signalModifiedKey(c,c->db,c->argv[2]);
server.dirty++;
/* An extra key has changed when ele was successfully added to dstset */
if (setTypeAdd(dstset,ele->ptr)) {
server.dirty++;
signalModifiedKey(c,c->db,c->argv[2]);
notifyKeyspaceEvent(NOTIFY_SET,"sadd",c->argv[2],c->db->id);
}
addReply(c,shared.cone);
@@ -858,24 +858,17 @@ void sinterGenericCommand(client *c, robj **setkeys,
int64_t intobj;
void *replylen = NULL;
unsigned long j, cardinality = 0;
int encoding;
int encoding, empty = 0;
for (j = 0; j < setnum; j++) {
robj *setobj = dstkey ?
lookupKeyWrite(c->db,setkeys[j]) :
lookupKeyRead(c->db,setkeys[j]);
if (!setobj) {
zfree(sets);
if (dstkey) {
if (dbDelete(c->db,dstkey)) {
signalModifiedKey(c,c->db,dstkey);
server.dirty++;
}
addReply(c,shared.czero);
} else {
addReply(c,shared.emptyset[c->resp]);
}
return;
/* A NULL is considered an empty set */
empty += 1;
sets[j] = NULL;
continue;
}
if (checkType(c,setobj,OBJ_SET)) {
zfree(sets);
@@ -883,6 +876,24 @@ void sinterGenericCommand(client *c, robj **setkeys,
}
sets[j] = setobj;
}
/* Set intersection with an empty set always results in an empty set.
* Return ASAP if there is an empty set. */
if (empty > 0) {
zfree(sets);
if (dstkey) {
if (dbDelete(c->db,dstkey)) {
signalModifiedKey(c,c->db,dstkey);
notifyKeyspaceEvent(NOTIFY_GENERIC,"del",dstkey,c->db->id);
server.dirty++;
}
addReply(c,shared.czero);
} else {
addReply(c,shared.emptyset[c->resp]);
}
return;
}
/* Sort sets from the smallest to largest, this will improve our
* algorithm's performance */
qsort(sets,setnum,sizeof(robj*),qsortCompareSetsByCardinality);
@@ -976,10 +987,12 @@ void sinterGenericCommand(client *c, robj **setkeys,
zfree(sets);
}
/* SINTER key [key ...] */
void sinterCommand(client *c) {
sinterGenericCommand(c,c->argv+1,c->argc-1,NULL);
}
/* SINTERSTORE destination key [key ...] */
void sinterstoreCommand(client *c) {
sinterGenericCommand(c,c->argv+2,c->argc-2,c->argv[1]);
}
@@ -1149,18 +1162,22 @@ void sunionDiffGenericCommand(client *c, robj **setkeys, int setnum,
zfree(sets);
}
/* SUNION key [key ...] */
void sunionCommand(client *c) {
sunionDiffGenericCommand(c,c->argv+1,c->argc-1,NULL,SET_OP_UNION);
}
/* SUNIONSTORE destination key [key ...] */
void sunionstoreCommand(client *c) {
sunionDiffGenericCommand(c,c->argv+2,c->argc-2,c->argv[1],SET_OP_UNION);
}
/* SDIFF key [key ...] */
void sdiffCommand(client *c) {
sunionDiffGenericCommand(c,c->argv+1,c->argc-1,NULL,SET_OP_DIFF);
}
/* SDIFFSTORE destination key [key ...] */
void sdiffstoreCommand(client *c) {
sunionDiffGenericCommand(c,c->argv+2,c->argc-2,c->argv[1],SET_OP_DIFF);
}
+5 -5
View File
@@ -702,16 +702,16 @@ int64_t streamTrim(stream *s, streamAddTrimArgs *args) {
int64_t deleted = 0;
while (raxNext(&ri)) {
/* Check if we exceeded the amount of work we could do */
if (limit && deleted >= limit)
break;
if (trim_strategy == TRIM_STRATEGY_MAXLEN && s->length <= maxlen)
break;
unsigned char *lp = ri.data, *p = lpFirst(lp);
int64_t entries = lpGetInteger(p);
/* Check if we exceeded the amount of work we could do */
if (limit && (deleted + entries) > limit)
break;
/* Check if we can remove the whole node. */
int remove_node;
streamID master_id = {0}; /* For MINID */
@@ -3233,7 +3233,7 @@ void xtrimCommand(client *c) {
/* Argument parsing. */
streamAddTrimArgs parsed_args;
if (streamParseAddOrTrimArgsOrReply(c, &parsed_args, 1) < 0)
if (streamParseAddOrTrimArgsOrReply(c, &parsed_args, 0) < 0)
return; /* streamParseAddOrTrimArgsOrReply already replied. */
/* If the key does not exist, we are ok returning zero, that is, the
+17 -1
View File
@@ -797,6 +797,12 @@ void stralgoLCS(client *c) {
goto cleanup;
}
/* Detect string truncation or later overflows. */
if (sdslen(a) >= UINT32_MAX-1 || sdslen(b) >= UINT32_MAX-1) {
addReplyError(c, "String too long for LCS");
goto cleanup;
}
/* Compute the LCS using the vanilla dynamic programming technique of
* building a table of LCS(x,y) substrings. */
uint32_t alen = sdslen(a);
@@ -805,9 +811,19 @@ void stralgoLCS(client *c) {
/* Setup an uint32_t array to store at LCS[i,j] the length of the
* LCS A0..i-1, B0..j-1. Note that we have a linear array here, so
* we index it as LCS[j+(blen+1)*j] */
uint32_t *lcs = zmalloc((alen+1)*(blen+1)*sizeof(uint32_t));
#define LCS(A,B) lcs[(B)+((A)*(blen+1))]
/* Try to allocate the LCS table, and abort on overflow or insufficient memory. */
unsigned long long lcssize = (unsigned long long)(alen+1)*(blen+1); /* Can't overflow due to the size limits above. */
unsigned long long lcsalloc = lcssize * sizeof(uint32_t);
uint32_t *lcs = NULL;
if (lcsalloc < SIZE_MAX && lcsalloc / lcssize == sizeof(uint32_t))
lcs = ztrymalloc(lcsalloc);
if (!lcs) {
addReplyError(c, "Insufficient memory");
goto cleanup;
}
/* Start building the LCS table. */
for (uint32_t i = 0; i <= alen; i++) {
for (uint32_t j = 0; j <= blen; j++) {
+24 -8
View File
@@ -3663,7 +3663,12 @@ void zrangeGenericCommand(zrange_result_handler *handler, int argc_start, int st
lookupKeyWrite(c->db,key) :
lookupKeyRead(c->db,key);
if (zobj == NULL) {
addReply(c,shared.emptyarray);
if (store) {
handler->beginResultEmission(handler);
handler->finalizeResultEmission(handler, 0);
} else {
addReply(c, shared.emptyarray);
}
goto cleanup;
}
@@ -3820,11 +3825,16 @@ void genericZpopCommand(client *c, robj **keyv, int keyc, int where, int emitkey
}
void *arraylen_ptr = addReplyDeferredLen(c);
long arraylen = 0;
long result_count = 0;
/* We emit the key only for the blocking variant. */
if (emitkey) addReplyBulk(c,key);
/* Respond with a single (flat) array in RESP2 or if countarg is not
* provided (returning a single element). In RESP3, when countarg is
* provided, use nested array. */
int use_nested_array = c->resp > 2 && countarg != NULL;
/* Remove the element. */
do {
if (zobj->encoding == OBJ_ENCODING_ZIPLIST) {
@@ -3867,16 +3877,19 @@ void genericZpopCommand(client *c, robj **keyv, int keyc, int where, int emitkey
serverAssertWithInfo(c,zobj,zsetDel(zobj,ele));
server.dirty++;
if (arraylen == 0) { /* Do this only for the first iteration. */
if (result_count == 0) { /* Do this only for the first iteration. */
char *events[2] = {"zpopmin","zpopmax"};
notifyKeyspaceEvent(NOTIFY_ZSET,events[where],key,c->db->id);
signalModifiedKey(c,c->db,key);
}
if (use_nested_array) {
addReplyArrayLen(c,2);
}
addReplyBulkCBuffer(c,ele,sdslen(ele));
addReplyDouble(c,score);
sdsfree(ele);
arraylen += 2;
++result_count;
/* Remove the key, if indeed needed. */
if (zsetLength(zobj) == 0) {
@@ -3886,7 +3899,10 @@ void genericZpopCommand(client *c, robj **keyv, int keyc, int where, int emitkey
}
} while(--count);
setDeferredArrayLen(c,arraylen_ptr,arraylen + (emitkey != 0));
if (!use_nested_array) {
result_count *= 2;
}
setDeferredArrayLen(c,arraylen_ptr,result_count + (emitkey != 0));
}
/* ZPOPMIN key [<count>] */
@@ -3987,7 +4003,7 @@ void zrandmemberWithCountCommand(client *c, long l, int withscores) {
int uniq = 1;
robj *zsetobj;
if ((zsetobj = lookupKeyReadOrReply(c, c->argv[1], shared.null[c->resp]))
if ((zsetobj = lookupKeyReadOrReply(c, c->argv[1], shared.emptyarray))
== NULL || checkType(c, zsetobj, OBJ_ZSET)) return;
size = zsetLength(zsetobj);
@@ -4023,7 +4039,7 @@ void zrandmemberWithCountCommand(client *c, long l, int withscores) {
addReplyArrayLen(c,2);
addReplyBulkCBuffer(c, key, sdslen(key));
if (withscores)
addReplyDouble(c, dictGetDoubleVal(de));
addReplyDouble(c, *(double*)dictGetVal(de));
}
} else if (zsetobj->encoding == OBJ_ENCODING_ZIPLIST) {
ziplistEntry *keys, *vals = NULL;
@@ -4172,7 +4188,7 @@ void zrandmemberWithCountCommand(client *c, long l, int withscores) {
}
}
/* ZRANDMEMBER [<count> WITHSCORES] */
/* ZRANDMEMBER key [<count> [WITHSCORES]] */
void zrandmemberCommand(client *c) {
long l;
int withscores = 0;
+2
View File
@@ -146,6 +146,8 @@ void tlsInit(void) {
*/
#if OPENSSL_VERSION_NUMBER < 0x10100000L
OPENSSL_config(NULL);
#elif OPENSSL_VERSION_NUMBER < 0x10101000L
OPENSSL_init_crypto(OPENSSL_INIT_LOAD_CONFIG, NULL);
#else
OPENSSL_init_crypto(OPENSSL_INIT_LOAD_CONFIG|OPENSSL_INIT_ATFORK, NULL);
#endif
+1 -1
View File
@@ -326,7 +326,7 @@ void trackingRememberKeyToBroadcast(client *c, char *keyname, size_t keylen) {
* tree. This way we know who was the client that did the last
* change to the key, and can avoid sending the notification in the
* case the client is in NOLOOP mode. */
raxTryInsert(bs->keys,(unsigned char*)keyname,keylen,c,NULL);
raxInsert(bs->keys,(unsigned char*)keyname,keylen,c,NULL);
}
raxStop(&ri);
}
+2 -2
View File
@@ -1,2 +1,2 @@
#define REDIS_VERSION "255.255.255"
#define REDIS_VERSION_NUM 0x00ffffff
#define REDIS_VERSION "6.2.5"
#define REDIS_VERSION_NUM 0x00060205
+1 -1
View File
@@ -263,7 +263,7 @@
* to stay there to signal that a full scan is needed to get the number of
* items inside the ziplist. */
#define ZIPLIST_INCR_LENGTH(zl,incr) { \
if (ZIPLIST_LENGTH(zl) < UINT16_MAX) \
if (intrev16ifbe(ZIPLIST_LENGTH(zl)) < UINT16_MAX) \
ZIPLIST_LENGTH(zl) = intrev16ifbe(intrev16ifbe(ZIPLIST_LENGTH(zl))+incr); \
}
+1
View File
@@ -12,6 +12,7 @@ set ::tlsdir "tests/tls"
# blocking.
proc bg_block_op {host port db ops tls} {
set r [redis $host $port 0 $tls]
$r client setname LOAD_HANDLER
$r select $db
for {set j 0} {$j < $ops} {incr j} {
+1
View File
@@ -5,6 +5,7 @@ set ::tlsdir "tests/tls"
proc bg_complex_data {host port db ops tls} {
set r [redis $host $port 0 $tls]
$r client setname LOAD_HANDLER
$r select $db
createComplexDataset $r $ops
}
+1
View File
@@ -5,6 +5,7 @@ set ::tlsdir "tests/tls"
proc gen_write_load {host port seconds tls} {
set start_time [clock seconds]
set r [redis $host $port 1 $tls]
$r client setname LOAD_HANDLER
$r select 9
while 1 {
$r set [expr rand()] [expr rand()]
+12
View File
@@ -158,6 +158,18 @@ tags {"aof"} {
assert_match "*not valid*" $result
}
test "Short read: Utility should show the abnormal line num in AOF" {
create_aof {
append_to_aof [formatCommand set foo hello]
append_to_aof "!!!"
}
catch {
exec src/redis-check-aof $aof_path
} result
assert_match "*ok_up_to_line=8*" $result
}
test "Short read: Utility should be able to fix the AOF" {
set result [exec src/redis-check-aof --fix $aof_path << "y\n"]
assert_match "*Successfully truncated AOF*" $result
+4 -11
View File
@@ -33,14 +33,9 @@ start_server {tags {"repl"}} {
stop_bg_block_op $load_handle0
stop_bg_block_op $load_handle1
stop_bg_block_op $load_handle2
set retry 10
while {$retry && ([$master debug digest] ne [$slave debug digest])}\
{
after 1000
incr retry -1
}
if {[$master debug digest] ne [$slave debug digest]} {
wait_for_condition 100 100 {
[$master debug digest] == [$slave debug digest]
} else {
set csv1 [csvdump r]
set csv2 [csvdump {r -1}]
set fd [open /tmp/repldump1.txt w]
@@ -49,10 +44,8 @@ start_server {tags {"repl"}} {
set fd [open /tmp/repldump2.txt w]
puts -nonewline $fd $csv2
close $fd
puts "Master - Replica inconsistency"
puts "Run diff -u against /tmp/repldump*.txt for more info"
fail "Master - Replica inconsistency, Run diff -u against /tmp/repldump*.txt for more info"
}
assert_equal [r debug digest] [r -1 debug digest]
}
}
}
+4
View File
@@ -83,7 +83,11 @@ start_server {} {
} else {
fail "Failover from node 0 to node 1 did not finish"
}
# stop the write load and make sure no more commands processed
stop_write_load $load_handler
wait_load_handlers_disconnected
$node_2 replicaof $node_1_host $node_1_port
wait_for_sync $node_0
wait_for_sync $node_2
+22 -6
View File
@@ -283,9 +283,9 @@ start_server {tags {"cli"}} {
assert_equal {key:2} [run_cli --scan --quoted-pattern {"*:\x32"}]
}
test "Connecting as a replica" {
proc test_redis_cli_repl {} {
set fd [open_cli "--replica"]
wait_for_condition 500 500 {
wait_for_condition 500 100 {
[string match {*slave0:*state=online*} [r info]]
} else {
fail "redis-cli --replica did not connect"
@@ -294,14 +294,30 @@ start_server {tags {"cli"}} {
for {set i 0} {$i < 100} {incr i} {
r set test-key test-value-$i
}
r client kill type slave
catch {
assert_match {*SET*key-a*} [read_cli $fd]
wait_for_condition 500 100 {
[string match {*test-value-99*} [read_cli $fd]]
} else {
fail "redis-cli --replica didn't read commands"
}
close_cli $fd
fconfigure $fd -blocking true
r client kill type slave
catch { close_cli $fd } err
assert_match {*Server closed the connection*} $err
}
test "Connecting as a replica" {
# Disk-based master
assert_match "OK" [r config set repl-diskless-sync no]
test_redis_cli_repl
# Disk-less master
assert_match "OK" [r config set repl-diskless-sync yes]
assert_match "OK" [r config set repl-diskless-sync-delay 0]
test_redis_cli_repl
} {}
test "Piping raw protocol" {
set cmds [tmpfile "cli_cmds"]
set cmds_fd [open $cmds "w"]
+5 -12
View File
@@ -21,15 +21,9 @@ start_server {tags {"repl network"}} {
stop_bg_complex_data $load_handle0
stop_bg_complex_data $load_handle1
stop_bg_complex_data $load_handle2
set retry 10
while {$retry && ([$master debug digest] ne [$slave debug digest])}\
{
after 1000
incr retry -1
}
assert {[$master dbsize] > 0}
if {[$master debug digest] ne [$slave debug digest]} {
wait_for_condition 100 100 {
[$master debug digest] == [$slave debug digest]
} else {
set csv1 [csvdump r]
set csv2 [csvdump {r -1}]
set fd [open /tmp/repldump1.txt w]
@@ -38,10 +32,9 @@ start_server {tags {"repl network"}} {
set fd [open /tmp/repldump2.txt w]
puts -nonewline $fd $csv2
close $fd
puts "Master - Replica inconsistency"
puts "Run diff -u against /tmp/repldump*.txt for more info"
fail "Master - Replica inconsistency, Run diff -u against /tmp/repldump*.txt for more info"
}
assert_equal [r debug digest] [r -1 debug digest]
assert {[$master dbsize] > 0}
}
}
}
+5 -12
View File
@@ -97,15 +97,9 @@ proc test_psync {descr duration backlog_size backlog_ttl delay cond mdl sdl reco
fail "Slave still not connected after some time"
}
set retry 10
while {$retry && ([$master debug digest] ne [$slave debug digest])}\
{
after 1000
incr retry -1
}
assert {[$master dbsize] > 0}
if {[$master debug digest] ne [$slave debug digest]} {
wait_for_condition 100 100 {
[$master debug digest] == [$slave debug digest]
} else {
set csv1 [csvdump r]
set csv2 [csvdump {r -1}]
set fd [open /tmp/repldump1.txt w]
@@ -114,10 +108,9 @@ proc test_psync {descr duration backlog_size backlog_ttl delay cond mdl sdl reco
set fd [open /tmp/repldump2.txt w]
puts -nonewline $fd $csv2
close $fd
puts "Master - Replica inconsistency"
puts "Run diff -u against /tmp/repldump*.txt for more info"
fail "Master - Replica inconsistency, Run diff -u against /tmp/repldump*.txt for more info"
}
assert_equal [r debug digest] [r -1 debug digest]
assert {[$master dbsize] > 0}
eval $cond
}
}
+46 -10
View File
@@ -316,15 +316,12 @@ foreach mdl {no yes} {
stop_write_load $load_handle3
stop_write_load $load_handle4
# Make sure that slaves and master have same
# number of keys
wait_for_condition 500 100 {
[$master dbsize] == [[lindex $slaves 0] dbsize] &&
[$master dbsize] == [[lindex $slaves 1] dbsize] &&
[$master dbsize] == [[lindex $slaves 2] dbsize]
} else {
fail "Different number of keys between master and replica after too long time."
}
# Make sure no more commands processed
wait_load_handlers_disconnected
wait_for_ofs_sync $master [lindex $slaves 0]
wait_for_ofs_sync $master [lindex $slaves 1]
wait_for_ofs_sync $master [lindex $slaves 2]
# Check digests
set digest [$master debug digest]
@@ -774,6 +771,45 @@ test "diskless replication child being killed is collected" {
}
}
test "diskless replication read pipe cleanup" {
# In diskless replication, we create a read pipe for the RDB, between the child and the parent.
# When we close this pipe (fd), the read handler also needs to be removed from the event loop (if it still registered).
# Otherwise, next time we will use the same fd, the registration will be fail (panic), because
# we will use EPOLL_CTL_MOD (the fd still register in the event loop), on fd that already removed from epoll_ctl
start_server {tags {"repl"}} {
set master [srv 0 client]
set master_host [srv 0 host]
set master_port [srv 0 port]
set master_pid [srv 0 pid]
$master config set repl-diskless-sync yes
$master config set repl-diskless-sync-delay 0
# put enough data in the db, and slowdown the save, to keep the parent busy at the read process
$master config set rdb-key-save-delay 100000
$master debug populate 20000 test 10000
$master config set rdbcompression no
start_server {} {
set replica [srv 0 client]
set loglines [count_log_lines 0]
$replica config set repl-diskless-load swapdb
$replica replicaof $master_host $master_port
# wait for the replicas to start reading the rdb
wait_for_log_messages 0 {"*Loading DB in memory*"} $loglines 800 10
set loglines [count_log_lines 0]
# send FLUSHALL so the RDB child will be killed
$master flushall
# wait for another RDB child process to be started
wait_for_log_messages -1 {"*Background RDB transfer started by pid*"} $loglines 800 10
# make sure master is alive
$master ping
}
}
}
test {replicaof right after disconnection} {
# this is a rare race condition that was reproduced sporadically by the psync2 unit.
# see details in #7205
@@ -864,7 +900,7 @@ test {Kill rdb child process if its dumping RDB is not useful} {
# Slave2 disconnect with master
$slave2 slaveof no one
# Should kill child
wait_for_condition 20 10 {
wait_for_condition 100 10 {
[s 0 rdb_bgsave_in_progress] eq 0
} else {
fail "can't kill rdb child"
+1
View File
@@ -18,6 +18,7 @@ endif
TEST_MODULES = \
commandfilter.so \
basics.so \
testrdb.so \
fork.so \
infotest.so \
@@ -31,7 +31,7 @@
*/
#define REDISMODULE_EXPERIMENTAL_API
#include "../redismodule.h"
#include "redismodule.h"
#include <string.h>
/* --------------------------------- Helpers -------------------------------- */
@@ -152,11 +152,64 @@ int TestUnlink(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
return failTest(ctx, "Could not verify key to be unlinked");
}
return RedisModule_ReplyWithSimpleString(ctx, "OK");
}
/* TEST.STRING.TRUNCATE -- Test truncating an existing string object. */
int TestStringTruncate(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
RedisModule_AutoMemory(ctx);
REDISMODULE_NOT_USED(argv);
REDISMODULE_NOT_USED(argc);
RedisModule_Call(ctx, "SET", "cc", "foo", "abcde");
RedisModuleKey *k = RedisModule_OpenKey(ctx, RedisModule_CreateStringPrintf(ctx, "foo"), REDISMODULE_READ | REDISMODULE_WRITE);
if (!k) return failTest(ctx, "Could not create key");
size_t len = 0;
char* s;
/* expand from 5 to 8 and check null pad */
if (REDISMODULE_ERR == RedisModule_StringTruncate(k, 8)) {
return failTest(ctx, "Could not truncate string value (8)");
}
s = RedisModule_StringDMA(k, &len, REDISMODULE_READ);
if (!s) {
return failTest(ctx, "Failed to read truncated string (8)");
} else if (len != 8) {
return failTest(ctx, "Failed to expand string value (8)");
} else if (0 != strncmp(s, "abcde\0\0\0", 8)) {
return failTest(ctx, "Failed to null pad string value (8)");
}
/* shrink from 8 to 4 */
if (REDISMODULE_ERR == RedisModule_StringTruncate(k, 4)) {
return failTest(ctx, "Could not truncate string value (4)");
}
s = RedisModule_StringDMA(k, &len, REDISMODULE_READ);
if (!s) {
return failTest(ctx, "Failed to read truncated string (4)");
} else if (len != 4) {
return failTest(ctx, "Failed to shrink string value (4)");
} else if (0 != strncmp(s, "abcd", 4)) {
return failTest(ctx, "Failed to truncate string value (4)");
}
/* shrink to 0 */
if (REDISMODULE_ERR == RedisModule_StringTruncate(k, 0)) {
return failTest(ctx, "Could not truncate string value (0)");
}
s = RedisModule_StringDMA(k, &len, REDISMODULE_READ);
if (!s) {
return failTest(ctx, "Failed to read truncated string (0)");
} else if (len != 0) {
return failTest(ctx, "Failed to shrink string value to (0)");
}
return RedisModule_ReplyWithSimpleString(ctx, "OK");
}
int NotifyCallback(RedisModuleCtx *ctx, int type, const char *event,
RedisModuleString *key) {
RedisModule_AutoMemory(ctx);
/* Increment a counter on the notifications: for each key notified we
* increment a counter */
RedisModule_Log(ctx, "notice", "Got event type %d, event %s, key %s", type,
@@ -168,6 +221,7 @@ int NotifyCallback(RedisModuleCtx *ctx, int type, const char *event,
/* TEST.NOTIFICATIONS -- Test Keyspace Notifications. */
int TestNotifications(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
RedisModule_AutoMemory(ctx);
REDISMODULE_NOT_USED(argv);
REDISMODULE_NOT_USED(argc);
@@ -279,6 +333,9 @@ int TestCtxFlags(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
flags = RedisModule_GetContextFlags(ctx);
if (!(flags & REDISMODULE_CTX_FLAGS_AOF)) FAIL("AOF Flag not set after config set");
/* Disable RDB saving and test the flag. */
RedisModule_Call(ctx, "config", "ccc", "set", "save", "");
flags = RedisModule_GetContextFlags(ctx);
if (flags & REDISMODULE_CTX_FLAGS_RDB) FAIL("RDB Flag was set");
/* Enable RDB to test RDB flags */
RedisModule_Call(ctx, "config", "ccc", "set", "save", "900 1");
@@ -290,8 +347,12 @@ int TestCtxFlags(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
if (flags & REDISMODULE_CTX_FLAGS_READONLY) FAIL("Read-only flag was set");
if (flags & REDISMODULE_CTX_FLAGS_CLUSTER) FAIL("Cluster flag was set");
/* Disable maxmemory and test the flag. (it is implicitly set in 32bit builds. */
RedisModule_Call(ctx, "config", "ccc", "set", "maxmemory", "0");
flags = RedisModule_GetContextFlags(ctx);
if (flags & REDISMODULE_CTX_FLAGS_MAXMEMORY) FAIL("Maxmemory flag was set");
/* Enable maxmemory and test the flag. */
RedisModule_Call(ctx, "config", "ccc", "set", "maxmemory", "100000000");
flags = RedisModule_GetContextFlags(ctx);
if (!(flags & REDISMODULE_CTX_FLAGS_MAXMEMORY))
@@ -324,7 +385,11 @@ end:
int TestAssertStringReply(RedisModuleCtx *ctx, RedisModuleCallReply *reply, char *str, size_t len) {
RedisModuleString *mystr, *expected;
if (RedisModule_CallReplyType(reply) != REDISMODULE_REPLY_STRING) {
if (RedisModule_CallReplyType(reply) == REDISMODULE_REPLY_ERROR) {
RedisModule_Log(ctx,"warning","Test error reply: %s",
RedisModule_CallReplyStringPtr(reply, NULL));
return 0;
} else if (RedisModule_CallReplyType(reply) != REDISMODULE_REPLY_STRING) {
RedisModule_Log(ctx,"warning","Unexpected reply type %d",
RedisModule_CallReplyType(reply));
return 0;
@@ -345,7 +410,11 @@ int TestAssertStringReply(RedisModuleCtx *ctx, RedisModuleCallReply *reply, char
/* Return 1 if the reply matches the specified integer, otherwise log errors
* in the server log and return 0. */
int TestAssertIntegerReply(RedisModuleCtx *ctx, RedisModuleCallReply *reply, long long expected) {
if (RedisModule_CallReplyType(reply) != REDISMODULE_REPLY_INTEGER) {
if (RedisModule_CallReplyType(reply) == REDISMODULE_REPLY_ERROR) {
RedisModule_Log(ctx,"warning","Test error reply: %s",
RedisModule_CallReplyStringPtr(reply, NULL));
return 0;
} else if (RedisModule_CallReplyType(reply) != REDISMODULE_REPLY_INTEGER) {
RedisModule_Log(ctx,"warning","Unexpected reply type %d",
RedisModule_CallReplyType(reply));
return 0;
@@ -366,8 +435,11 @@ int TestAssertIntegerReply(RedisModuleCtx *ctx, RedisModuleCallReply *reply, lon
reply = RedisModule_Call(ctx,name,__VA_ARGS__); \
} while (0)
/* TEST.IT -- Run all the tests. */
int TestIt(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
/* TEST.BASICS -- Run all the tests.
* Note: it is useful to run these tests from the module rather than TCL
* since it's easier to check the reply types like that (make a distinction
* between 0 and "0", etc. */
int TestBasics(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
REDISMODULE_NOT_USED(argv);
REDISMODULE_NOT_USED(argc);
@@ -390,6 +462,9 @@ int TestIt(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
T("test.string.append","");
if (!TestAssertStringReply(ctx,reply,"foobar",6)) goto fail;
T("test.string.truncate","");
if (!TestAssertStringReply(ctx,reply,"OK",2)) goto fail;
T("test.unlink","");
if (!TestAssertStringReply(ctx,reply,"OK",2)) goto fail;
@@ -407,7 +482,7 @@ int TestIt(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
fail:
RedisModule_ReplyWithSimpleString(ctx,
"SOME TEST NOT PASSED! Check server logs");
"SOME TEST DID NOT PASS! Check server logs");
return REDISMODULE_OK;
}
@@ -430,6 +505,10 @@ int RedisModule_OnLoad(RedisModuleCtx *ctx, RedisModuleString **argv, int argc)
TestStringAppendAM,"write deny-oom",1,1,1) == REDISMODULE_ERR)
return REDISMODULE_ERR;
if (RedisModule_CreateCommand(ctx,"test.string.truncate",
TestStringTruncate,"write deny-oom",1,1,1) == REDISMODULE_ERR)
return REDISMODULE_ERR;
if (RedisModule_CreateCommand(ctx,"test.string.printf",
TestStringPrintf,"write deny-oom",1,1,1) == REDISMODULE_ERR)
return REDISMODULE_ERR;
@@ -442,8 +521,8 @@ int RedisModule_OnLoad(RedisModuleCtx *ctx, RedisModuleString **argv, int argc)
TestUnlink,"write deny-oom",1,1,1) == REDISMODULE_ERR)
return REDISMODULE_ERR;
if (RedisModule_CreateCommand(ctx,"test.it",
TestIt,"readonly",1,1,1) == REDISMODULE_ERR)
if (RedisModule_CreateCommand(ctx,"test.basics",
TestBasics,"readonly",1,1,1) == REDISMODULE_ERR)
return REDISMODULE_ERR;
RedisModule_SubscribeToKeyspaceEvents(ctx,
+72
View File
@@ -195,6 +195,66 @@ int HelloDoubleBlock_RedisCommand(RedisModuleCtx *ctx, RedisModuleString **argv,
return REDISMODULE_OK;
}
RedisModuleBlockedClient *blocked_client = NULL;
/* BLOCK.BLOCK [TIMEOUT] -- Blocks the current client until released
* or TIMEOUT seconds. If TIMEOUT is zero, no timeout function is
* registered.
*/
int Block_RedisCommand(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
if (RedisModule_IsBlockedReplyRequest(ctx)) {
RedisModuleString *r = RedisModule_GetBlockedClientPrivateData(ctx);
return RedisModule_ReplyWithString(ctx, r);
} else if (RedisModule_IsBlockedTimeoutRequest(ctx)) {
RedisModule_UnblockClient(blocked_client, NULL); /* Must be called to avoid leaks. */
blocked_client = NULL;
return RedisModule_ReplyWithSimpleString(ctx, "Timed out");
}
if (argc != 2) return RedisModule_WrongArity(ctx);
long long timeout;
if (RedisModule_StringToLongLong(argv[1], &timeout) != REDISMODULE_OK) {
return RedisModule_ReplyWithError(ctx, "ERR invalid timeout");
}
if (blocked_client) {
return RedisModule_ReplyWithError(ctx, "ERR another client already blocked");
}
/* Block client. We use this function as both a reply and optional timeout
* callback and differentiate the different code flows above.
*/
blocked_client = RedisModule_BlockClient(ctx, Block_RedisCommand,
timeout > 0 ? Block_RedisCommand : NULL, NULL, timeout);
return REDISMODULE_OK;
}
/* BLOCK.IS_BLOCKED -- Returns 1 if we have a blocked client, or 0 otherwise.
*/
int IsBlocked_RedisCommand(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
UNUSED(argv);
UNUSED(argc);
RedisModule_ReplyWithLongLong(ctx, blocked_client ? 1 : 0);
return REDISMODULE_OK;
}
/* BLOCK.RELEASE [reply] -- Releases the blocked client and produce the specified reply.
*/
int Release_RedisCommand(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
if (argc != 2) return RedisModule_WrongArity(ctx);
if (!blocked_client) {
return RedisModule_ReplyWithError(ctx, "ERR No blocked client");
}
RedisModuleString *replystr = argv[1];
RedisModule_RetainString(ctx, replystr);
int err = RedisModule_UnblockClient(blocked_client, replystr);
blocked_client = NULL;
RedisModule_ReplyWithSimpleString(ctx, "OK");
return REDISMODULE_OK;
}
int RedisModule_OnLoad(RedisModuleCtx *ctx, RedisModuleString **argv, int argc) {
UNUSED(argv);
@@ -215,5 +275,17 @@ int RedisModule_OnLoad(RedisModuleCtx *ctx, RedisModuleString **argv, int argc)
HelloBlockNoTracking_RedisCommand,"",0,0,0) == REDISMODULE_ERR)
return REDISMODULE_ERR;
if (RedisModule_CreateCommand(ctx, "block.block",
Block_RedisCommand, "", 0, 0, 0) == REDISMODULE_ERR)
return REDISMODULE_ERR;
if (RedisModule_CreateCommand(ctx,"block.is_blocked",
IsBlocked_RedisCommand,"",0,0,0) == REDISMODULE_ERR)
return REDISMODULE_ERR;
if (RedisModule_CreateCommand(ctx,"block.release",
Release_RedisCommand,"",0,0,0) == REDISMODULE_ERR)
return REDISMODULE_ERR;
return REDISMODULE_OK;
}
+46 -19
View File
@@ -34,13 +34,14 @@ array set ::redis::fd {}
array set ::redis::addr {}
array set ::redis::blocking {}
array set ::redis::deferred {}
array set ::redis::readraw {}
array set ::redis::reconnect {}
array set ::redis::tls {}
array set ::redis::callback {}
array set ::redis::state {} ;# State in non-blocking reply reading
array set ::redis::statestack {} ;# Stack of states, for nested mbulks
proc redis {{server 127.0.0.1} {port 6379} {defer 0} {tls 0} {tlsoptions {}}} {
proc redis {{server 127.0.0.1} {port 6379} {defer 0} {tls 0} {tlsoptions {}} {readraw 0}} {
if {$tls} {
package require tls
::tls::init \
@@ -58,6 +59,7 @@ proc redis {{server 127.0.0.1} {port 6379} {defer 0} {tls 0} {tlsoptions {}}} {
set ::redis::addr($id) [list $server $port]
set ::redis::blocking($id) 1
set ::redis::deferred($id) $defer
set ::redis::readraw($id) $readraw
set ::redis::reconnect($id) 0
set ::redis::tls($id) $tls
::redis::redis_reset_state $id
@@ -158,6 +160,7 @@ proc ::redis::__method__close {id fd} {
catch {unset ::redis::addr($id)}
catch {unset ::redis::blocking($id)}
catch {unset ::redis::deferred($id)}
catch {unset ::redis::readraw($id)}
catch {unset ::redis::reconnect($id)}
catch {unset ::redis::tls($id)}
catch {unset ::redis::state($id)}
@@ -174,6 +177,10 @@ proc ::redis::__method__deferred {id fd val} {
set ::redis::deferred($id) $val
}
proc ::redis::__method__readraw {id fd val} {
set ::redis::readraw($id) $val
}
proc ::redis::redis_write {fd buf} {
puts -nonewline $fd $buf
}
@@ -240,26 +247,46 @@ proc ::redis::redis_read_null fd {
return {}
}
proc ::redis::redis_read_bool fd {
set v [redis_read_line $fd]
if {$v == "t"} {return 1}
if {$v == "f"} {return 0}
return -code error "Bad protocol, '$v' as bool type"
}
proc ::redis::redis_read_reply {id fd} {
set type [read $fd 1]
switch -exact -- $type {
_ {redis_read_null $fd}
: -
+ {redis_read_line $fd}
, {expr {double([redis_read_line $fd])}}
- {return -code error [redis_read_line $fd]}
$ {redis_bulk_read $fd}
> -
~ -
* {redis_multi_bulk_read $id $fd}
% {redis_read_map $id $fd}
default {
if {$type eq {}} {
catch {close $fd}
set ::redis::fd($id) {}
return -code error "I/O error reading reply"
if {$::redis::readraw($id)} {
return [redis_read_line $fd]
}
while {1} {
set type [read $fd 1]
switch -exact -- $type {
_ {return [redis_read_null $fd]}
: -
( -
+ {return [redis_read_line $fd]}
, {return [expr {double([redis_read_line $fd])}]}
# {return [redis_read_bool $fd]}
- {return -code error [redis_read_line $fd]}
$ {return [redis_bulk_read $fd]}
> -
~ -
* {return [redis_multi_bulk_read $id $fd]}
% {return [redis_read_map $id $fd]}
| {
# ignore attributes for now (nowhere to store them)
redis_read_map $id $fd
continue
}
default {
if {$type eq {}} {
catch {close $fd}
set ::redis::fd($id) {}
return -code error "I/O error reading reply"
}
return -code error "Bad protocol, '$type' as reply type byte"
}
return -code error "Bad protocol, '$type' as reply type byte"
}
}
}
+8
View File
@@ -504,6 +504,14 @@ proc stop_write_load {handle} {
catch {exec /bin/kill -9 $handle}
}
proc wait_load_handlers_disconnected {{level 0}} {
wait_for_condition 50 100 {
![string match {*name=LOAD_HANDLER*} [r $level client list]]
} else {
fail "load_handler(s) still connected after too long time."
}
}
proc K { x y } { set x }
# Shuffle a list with Fisher-Yates algorithm.
+2 -9
View File
@@ -41,15 +41,8 @@ start_server {tags {"aofrw"}} {
stop_write_load $load_handle3
stop_write_load $load_handle4
# Make sure that we remain the only connected client.
# This step is needed to make sure there are no pending writes
# that will be processed between the two "debug digest" calls.
wait_for_condition 50 100 {
[llength [split [string trim [r client list]] "\n"]] == 1
} else {
puts [r client list]
fail "Clients generating loads are not disconnecting"
}
# Make sure no more commands processed, before taking debug digest
wait_load_handlers_disconnected
# Get the data set digest
set d1 [r debug digest]
+28
View File
@@ -349,3 +349,31 @@ start_server {tags {"bitops"}} {
}
}
}
start_server {tags {"bitops large-memory"}} {
test "BIT pos larger than UINT_MAX" {
set bytes [expr (1 << 29) + 1]
set bitpos [expr (1 << 32)]
set oldval [lindex [r config get proto-max-bulk-len] 1]
r config set proto-max-bulk-len $bytes
r setbit mykey $bitpos 1
assert_equal $bytes [r strlen mykey]
assert_equal 1 [r getbit mykey $bitpos]
assert_equal [list 128 128 -1] [r bitfield mykey get u8 $bitpos set u8 $bitpos 255 get i8 $bitpos]
assert_equal $bitpos [r bitpos mykey 1]
assert_equal $bitpos [r bitpos mykey 1 [expr $bytes - 1]]
if {$::accurate} {
# set all bits to 1
set mega [expr (1 << 23)]
set part [string repeat "\xFF" $mega]
for {set i 0} {$i < 64} {incr i} {
r setrange mykey [expr $i * $mega] $part
}
r setrange mykey [expr $bytes - 1] "\xFF"
assert_equal [expr $bitpos + 8] [r bitcount mykey]
assert_equal -1 [r bitpos mykey 0 0 [expr $bytes - 1]]
}
r config set proto-max-bulk-len $oldval
r del mykey
} {1}
}
+46
View File
@@ -31,6 +31,52 @@ start_server {tags {"introspection"}} {
assert_match {*lua*"set"*"foo"*"bar"*} [$rd read]
}
test {MONITOR supports redacting command arguments} {
set rd [redis_deferring_client]
$rd monitor
$rd read ; # Discard the OK
r migrate [srv 0 host] [srv 0 port] key 9 5000
r migrate [srv 0 host] [srv 0 port] key 9 5000 AUTH user
r migrate [srv 0 host] [srv 0 port] key 9 5000 AUTH2 user password
catch {r auth not-real} _
catch {r auth not-real not-a-password} _
catch {r hello 2 AUTH not-real not-a-password} _
assert_match {*"key"*"9"*"5000"*} [$rd read]
assert_match {*"key"*"9"*"5000"*"(redacted)"*} [$rd read]
assert_match {*"key"*"9"*"5000"*"(redacted)"*"(redacted)"*} [$rd read]
assert_match {*"auth"*"(redacted)"*} [$rd read]
assert_match {*"auth"*"(redacted)"*"(redacted)"*} [$rd read]
assert_match {*"hello"*"2"*"AUTH"*"(redacted)"*"(redacted)"*} [$rd read]
$rd close
}
test {MONITOR correctly handles multi-exec cases} {
set rd [redis_deferring_client]
$rd monitor
$rd read ; # Discard the OK
# Make sure multi-exec statements are ordered
# correctly
r multi
r set foo bar
r exec
assert_match {*"multi"*} [$rd read]
assert_match {*"set"*"foo"*"bar"*} [$rd read]
assert_match {*"exec"*} [$rd read]
# Make sure we close multi statements on errors
r multi
catch {r syntax error} _
catch {r exec} _
assert_match {*"multi"*} [$rd read]
assert_match {*"exec"*} [$rd read]
$rd close
}
test {CLIENT GETNAME should return NIL if name is not assigned} {
r client getname
} {}
+12
View File
@@ -0,0 +1,12 @@
set testmodule [file normalize tests/modules/basics.so]
start_server {tags {"modules"}} {
r module load $testmodule
test {test module api basics} {
r test.basics
} {ALL TESTS PASSED}
r module unload test
}
@@ -85,4 +85,33 @@ start_server {tags {"modules"}} {
assert_equal [r slowlog len] 0
}
}
test "client unblock works only for modules with timeout support" {
set rd [redis_deferring_client]
$rd client id
set id [$rd read]
# Block with a timeout function - may unblock
$rd block.block 20000
wait_for_condition 50 100 {
[r block.is_blocked] == 1
} else {
fail "Module did not block"
}
assert_equal 1 [r client unblock $id]
assert_match {*Timed out*} [$rd read]
# Block without a timeout function - cannot unblock
$rd block.block 0
wait_for_condition 50 100 {
[r block.is_blocked] == 1
} else {
fail "Module did not block"
}
assert_equal 0 [r client unblock $id]
assert_equal "OK" [r block.release foobar]
assert_equal "foobar" [$rd read]
}
}
+16
View File
@@ -121,6 +121,22 @@ start_server {tags {"multi"}} {
r exec
} {}
test {EXEC fail on lazy expired WATCHed key} {
r flushall
r debug set-active-expire 0
r del key
r set key 1 px 2
r watch key
after 100
r multi
r incr key
assert_equal [r exec] {}
r debug set-active-expire 1
} {OK} {needs:debug}
test {After successful EXEC key is no longer watched} {
r set x 30
r watch x
+2 -2
View File
@@ -25,7 +25,7 @@ test {CONFIG SET port number} {
test {CONFIG SET bind address} {
start_server {} {
# non-valid address
catch {r CONFIG SET bind "some.wrong.bind.address"} e
catch {r CONFIG SET bind "999.999.999.999"} e
assert_match {*Failed to bind to specified addresses*} $e
# make sure server still bound to the previous address
@@ -33,4 +33,4 @@ test {CONFIG SET bind address} {
$rd PING
$rd close
}
}
}
+59 -52
View File
@@ -18,65 +18,72 @@ start_server {tags {"obuf-limits"}} {
assert {$omem >= 70000 && $omem < 200000}
$rd1 close
}
test {Client output buffer soft limit is not enforced if time is not overreached} {
r config set client-output-buffer-limit {pubsub 0 100000 10}
set rd1 [redis_deferring_client]
$rd1 subscribe foo
set reply [$rd1 read]
assert {$reply eq "subscribe foo 1"}
set omem 0
set start_time 0
set time_elapsed 0
while 1 {
if {$start_time != 0} {
# Slow down loop when omen has reached the limit.
after 10
}
r publish foo [string repeat "x" 1000]
set clients [split [r client list] "\r\n"]
set c [split [lindex $clients 1] " "]
if {![regexp {omem=([0-9]+)} $c - omem]} break
if {$omem > 100000} {
if {$start_time == 0} {set start_time [clock seconds]}
set time_elapsed [expr {[clock seconds]-$start_time}]
if {$time_elapsed >= 5} break
}
foreach {soft_limit_time wait_for_timeout} {3 yes
4 no } {
if $wait_for_timeout {
set test_name "Client output buffer soft limit is enforced if time is overreached"
} else {
set test_name "Client output buffer soft limit is not enforced too early and is enforced when no traffic"
}
assert {$omem >= 100000 && $time_elapsed >= 5 && $time_elapsed <= 10}
$rd1 close
}
test {Client output buffer soft limit is enforced if time is overreached} {
r config set client-output-buffer-limit {pubsub 0 100000 3}
set rd1 [redis_deferring_client]
test $test_name {
r config set client-output-buffer-limit "pubsub 0 100000 $soft_limit_time"
set soft_limit_time [expr $soft_limit_time*1000]
set rd1 [redis_deferring_client]
$rd1 subscribe foo
set reply [$rd1 read]
assert {$reply eq "subscribe foo 1"}
$rd1 client setname test_client
set reply [$rd1 read]
assert {$reply eq "OK"}
set omem 0
set start_time 0
set time_elapsed 0
while 1 {
if {$start_time != 0} {
# Slow down loop when omen has reached the limit.
after 10
$rd1 subscribe foo
set reply [$rd1 read]
assert {$reply eq "subscribe foo 1"}
set omem 0
set start_time 0
set time_elapsed 0
set last_under_limit_time [clock milliseconds]
while 1 {
r publish foo [string repeat "x" 1000]
set clients [split [r client list] "\r\n"]
set c [lsearch -inline $clients *name=test_client*]
if {$start_time != 0} {
set time_elapsed [expr {[clock milliseconds]-$start_time}]
# Make sure test isn't taking too long
assert {$time_elapsed <= [expr $soft_limit_time+3000]}
}
if {$wait_for_timeout && $c == ""} {
# Make sure we're disconnected when we reach the soft limit
assert {$omem >= 100000 && $time_elapsed >= $soft_limit_time}
break
} else {
assert {[regexp {omem=([0-9]+)} $c - omem]}
}
if {$omem > 100000} {
if {$start_time == 0} {set start_time $last_under_limit_time}
if {!$wait_for_timeout && $time_elapsed >= [expr $soft_limit_time-1000]} break
# Slow down loop when omem has reached the limit.
after 10
} else {
# if the OS socket buffers swallowed what we previously filled, reset the start timer.
set start_time 0
set last_under_limit_time [clock milliseconds]
}
}
r publish foo [string repeat "x" 1000]
set clients [split [r client list] "\r\n"]
set c [split [lindex $clients 1] " "]
if {![regexp {omem=([0-9]+)} $c - omem]} break
if {$omem > 100000} {
if {$start_time == 0} {set start_time [clock seconds]}
set time_elapsed [expr {[clock seconds]-$start_time}]
if {$time_elapsed >= 10} break
if {!$wait_for_timeout} {
# After we completely stopped the traffic, wait for soft limit to time out
set timeout [expr {$soft_limit_time+1500 - ([clock milliseconds]-$start_time)}]
wait_for_condition [expr $timeout/10] 10 {
[lsearch [split [r client list] "\r\n"] *name=test_client*] == -1
} else {
fail "Soft limit timed out but client still connected"
}
}
$rd1 close
}
assert {$omem >= 100000 && $time_elapsed < 6}
$rd1 close
}
test {No response for single command if client output buffer hard limit is enforced} {
-3
View File
@@ -266,9 +266,6 @@ start_server {overrides {save ""} tags {"other"}} {
assert_equal [$rd read] "OK"
$rd reset
# skip reset ouptut
$rd read
assert_equal [$rd read] "RESET"
assert_no_match {*flags=O*} [r client list]
+88
View File
@@ -102,6 +102,94 @@ start_server {tags {"protocol network"}} {
} {*Protocol error*}
}
unset c
# recover the broken connection
reconnect
r ping
# raw RESP response tests
r readraw 1
test "raw protocol response" {
r srandmember nonexisting_key
} {*-1}
r deferred 1
test "raw protocol response - deferred" {
r srandmember nonexisting_key
r read
} {*-1}
test "raw protocol response - multiline" {
r sadd ss a
assert_equal [r read] {:1}
r srandmember ss 100
assert_equal [r read] {*1}
assert_equal [r read] {$1}
assert_equal [r read] {a}
}
# restore connection settings
r readraw 0
r deferred 0
# check the connection still works
assert_equal [r ping] {PONG}
test {RESP3 attributes} {
r hello 3
set res [r debug protocol attrib]
# currently the parser in redis.tcl ignores the attributes
# restore state
r hello 2
set _ $res
} {Some real reply following the attribute}
test {RESP3 attributes readraw} {
r hello 3
r readraw 1
r deferred 1
r debug protocol attrib
assert_equal [r read] {|1}
assert_equal [r read] {$14}
assert_equal [r read] {key-popularity}
assert_equal [r read] {*2}
assert_equal [r read] {$7}
assert_equal [r read] {key:123}
assert_equal [r read] {:90}
assert_equal [r read] {$39}
assert_equal [r read] {Some real reply following the attribute}
# restore state
r readraw 0
r deferred 0
r hello 2
set _ {}
} {}
test {RESP3 attributes on RESP2} {
r hello 2
set res [r debug protocol attrib]
set _ $res
} {Some real reply following the attribute}
test "test big number parsing" {
r hello 3
r debug protocol bignum
} {1234567999999999999999999999999999999}
test "test bool parsing" {
r hello 3
assert_equal [r debug protocol true] 1
assert_equal [r debug protocol false] 0
r hello 2
assert_equal [r debug protocol true] 1
assert_equal [r debug protocol false] 0
set _ {}
} {}
}
start_server {tags {"regression"}} {
+21 -4
View File
@@ -45,18 +45,35 @@ start_server {tags {"slowlog"} overrides {slowlog-log-slower-than 1000000}} {
r config set slowlog-log-slower-than 0
r slowlog reset
r config set masterauth ""
r acl setuser slowlog-test-user
r acl setuser slowlog-test-user +get +set
r config set slowlog-log-slower-than 0
r config set slowlog-log-slower-than 10000
set slowlog_resp [r slowlog get]
# Make sure normal configs work, but the two sensitive
# commands are omitted
assert_equal 2 [llength $slowlog_resp]
assert_equal {slowlog reset} [lindex [lindex [r slowlog get] 1] 3]
# commands are omitted or redacted
assert_equal 4 [llength $slowlog_resp]
assert_equal {slowlog reset} [lindex [lindex [r slowlog get] 3] 3]
assert_equal {config set masterauth (redacted)} [lindex [lindex [r slowlog get] 2] 3]
assert_equal {acl setuser (redacted) (redacted) (redacted)} [lindex [lindex [r slowlog get] 1] 3]
assert_equal {config set slowlog-log-slower-than 0} [lindex [lindex [r slowlog get] 0] 3]
}
test {SLOWLOG - Some commands can redact sensitive fields} {
r config set slowlog-log-slower-than 0
r slowlog reset
r migrate [srv 0 host] [srv 0 port] key 9 5000
r migrate [srv 0 host] [srv 0 port] key 9 5000 AUTH user
r migrate [srv 0 host] [srv 0 port] key 9 5000 AUTH2 user password
r config set slowlog-log-slower-than 10000
# Make sure all 3 commands were logged, but the sensitive fields are omitted
assert_equal 4 [llength [r slowlog get]]
assert_match {* key 9 5000} [lindex [lindex [r slowlog get] 2] 3]
assert_match {* key 9 5000 AUTH (redacted)} [lindex [lindex [r slowlog get] 1] 3]
assert_match {* key 9 5000 AUTH2 (redacted) (redacted)} [lindex [lindex [r slowlog get] 0] 3]
}
test {SLOWLOG - Rewritten commands are logged as their original command} {
r config set slowlog-log-slower-than 0
+27
View File
@@ -132,6 +132,22 @@ start_server {tags {"tracking network"}} {
assert {$keys eq {mykey}}
}
test {Tracking gets notification of lazy expired keys} {
r CLIENT TRACKING off
r CLIENT TRACKING on BCAST REDIRECT $redir_id NOLOOP
# Use multi-exec to expose a race where the key gets an two invalidations
# in the same event loop, once by the client so filtered by NOLOOP, and
# the second one by the lazy expire
r MULTI
r SET mykey{t} myval px 1
r SET mykeyotherkey{t} myval ; # We should not get it
r DEBUG SLEEP 0.1
r GET mykey{t}
r EXEC
set keys [lsort [lindex [$rd_redirection read] 2]]
assert {$keys eq {mykey{t}}}
} {} {needs:debug}
test {HELLO 3 reply is correct} {
set reply [r HELLO 3]
assert_equal [dict get $reply proto] 3
@@ -395,6 +411,17 @@ start_server {tags {"tracking network"}} {
assert {[lindex msg 2] eq {} }
}
test {Test ASYNC flushall} {
clean_all
r CLIENT TRACKING on REDIRECT $redir_id
r GET key1
r GET key2
assert_equal [s 0 tracking_total_keys] 2
$rd_sg FLUSHALL ASYNC
assert_equal [s 0 tracking_total_keys] 0
assert_equal [lindex [$rd_redirection read] 2] {}
}
# Keys are defined to be evicted 100 at a time by default.
# If after eviction the number of keys still surpasses the limit
# defined in tracking-table-max-keys, we increases eviction
+13
View File
@@ -72,6 +72,19 @@ start_server {tags {"hash"}} {
r hrandfield nonexisting_key 100
} {}
# Make sure we can distinguish between an empty array and a null response
r readraw 1
test "HRANDFIELD count of 0 is handled correctly - emptyarray" {
r hrandfield myhash 0
} {*0}
test "HRANDFIELD with <count> against non existing key - emptyarray" {
r hrandfield nonexisting_key 100
} {*0}
r readraw 0
foreach {type contents} "
hashtable {{a 1} {b 2} {c 3} {d 4} {e 5} {6 f} {7 g} {8 h} {9 i} {[randstring 70 90 alpha] 10}}
ziplist {{a 1} {b 2} {c 3} {d 4} {e 5} {6 f} {7 g} {8 h} {9 i} {10 j}} " {
+225 -9
View File
@@ -276,14 +276,86 @@ start_server {
}
}
test "SINTER against non-set should throw error" {
r set key1 x
assert_error "WRONGTYPE*" {r sinter key1 noset}
test "SDIFF against non-set should throw error" {
# with an empty set
r set key1{t} x
assert_error "WRONGTYPE*" {r sdiff key1{t} noset{t}}
# different order
assert_error "WRONGTYPE*" {r sdiff noset{t} key1{t}}
# with a legal set
r del set1{t}
r sadd set1{t} a b c
assert_error "WRONGTYPE*" {r sdiff key1{t} set1{t}}
# different order
assert_error "WRONGTYPE*" {r sdiff set1{t} key1{t}}
}
test "SUNION against non-set should throw error" {
r set key1 x
assert_error "WRONGTYPE*" {r sunion key1 noset}
test "SDIFF should handle non existing key as empty" {
r del set1{t} set2{t} set3{t}
r sadd set1{t} a b c
r sadd set2{t} b c d
assert_equal {a} [lsort [r sdiff set1{t} set2{t} set3{t}]]
assert_equal {} [lsort [r sdiff set3{t} set2{t} set1{t}]]
}
test "SDIFFSTORE against non-set should throw error" {
r del set1{t} set2{t} set3{t} key1{t}
r set key1{t} x
# with en empty dstkey
assert_error "WRONGTYPE*" {r SDIFFSTORE set3{t} key1{t} noset{t}}
assert_equal 0 [r exists set3{t}]
assert_error "WRONGTYPE*" {r SDIFFSTORE set3{t} noset{t} key1{t}}
assert_equal 0 [r exists set3{t}]
# with a legal dstkey
r sadd set1{t} a b c
r sadd set2{t} b c d
r sadd set3{t} e
assert_error "WRONGTYPE*" {r SDIFFSTORE set3{t} key1{t} set1{t} noset{t}}
assert_equal 1 [r exists set3{t}]
assert_equal {e} [lsort [r smembers set3{t}]]
assert_error "WRONGTYPE*" {r SDIFFSTORE set3{t} set1{t} key1{t} set2{t}}
assert_equal 1 [r exists set3{t}]
assert_equal {e} [lsort [r smembers set3{t}]]
}
test "SDIFFSTORE should handle non existing key as empty" {
r del set1{t} set2{t} set3{t}
r set setres{t} xxx
assert_equal 0 [r sdiffstore setres{t} foo111{t} bar222{t}]
assert_equal 0 [r exists setres{t}]
# with a legal dstkey, should delete dstkey
r sadd set3{t} a b c
assert_equal 0 [r sdiffstore set3{t} set1{t} set2{t}]
assert_equal 0 [r exists set3{t}]
r sadd set1{t} a b c
assert_equal 3 [r sdiffstore set3{t} set1{t} set2{t}]
assert_equal 1 [r exists set3{t}]
assert_equal {a b c} [lsort [r smembers set3{t}]]
# with a legal dstkey and empty set2, should delete the dstkey
r sadd set3{t} a b c
assert_equal 0 [r sdiffstore set3{t} set2{t} set1{t}]
assert_equal 0 [r exists set3{t}]
}
test "SINTER against non-set should throw error" {
r set key1{t} x
assert_error "WRONGTYPE*" {r sinter key1{t} noset{t}}
# different order
assert_error "WRONGTYPE*" {r sinter noset{t} key1{t}}
r sadd set1{t} a b c
assert_error "WRONGTYPE*" {r sinter key1{t} set1{t}}
# different order
assert_error "WRONGTYPE*" {r sinter set1{t} key1{t}}
}
test "SINTER should handle non existing key as empty" {
@@ -303,10 +375,115 @@ start_server {
lsort [r sinter set1 set2]
} {1 2 3}
test "SINTERSTORE against non-set should throw error" {
r del set1{t} set2{t} set3{t} key1{t}
r set key1{t} x
# with en empty dstkey
assert_error "WRONGTYPE*" {r sinterstore set3{t} key1{t} noset{t}}
assert_equal 0 [r exists set3{t}]
assert_error "WRONGTYPE*" {r sinterstore set3{t} noset{t} key1{t}}
assert_equal 0 [r exists set3{t}]
# with a legal dstkey
r sadd set1{t} a b c
r sadd set2{t} b c d
r sadd set3{t} e
assert_error "WRONGTYPE*" {r sinterstore set3{t} key1{t} set2{t} noset{t}}
assert_equal 1 [r exists set3{t}]
assert_equal {e} [lsort [r smembers set3{t}]]
assert_error "WRONGTYPE*" {r sinterstore set3{t} noset{t} key1{t} set2{t}}
assert_equal 1 [r exists set3{t}]
assert_equal {e} [lsort [r smembers set3{t}]]
}
test "SINTERSTORE against non existing keys should delete dstkey" {
r set setres xxx
assert_equal 0 [r sinterstore setres foo111 bar222]
assert_equal 0 [r exists setres]
r del set1{t} set2{t} set3{t}
r set setres{t} xxx
assert_equal 0 [r sinterstore setres{t} foo111{t} bar222{t}]
assert_equal 0 [r exists setres{t}]
# with a legal dstkey
r sadd set3{t} a b c
assert_equal 0 [r sinterstore set3{t} set1{t} set2{t}]
assert_equal 0 [r exists set3{t}]
r sadd set1{t} a b c
assert_equal 0 [r sinterstore set3{t} set1{t} set2{t}]
assert_equal 0 [r exists set3{t}]
assert_equal 0 [r sinterstore set3{t} set2{t} set1{t}]
assert_equal 0 [r exists set3{t}]
}
test "SUNION against non-set should throw error" {
r set key1{t} x
assert_error "WRONGTYPE*" {r sunion key1{t} noset{t}}
# different order
assert_error "WRONGTYPE*" {r sunion noset{t} key1{t}}
r del set1{t}
r sadd set1{t} a b c
assert_error "WRONGTYPE*" {r sunion key1{t} set1{t}}
# different order
assert_error "WRONGTYPE*" {r sunion set1{t} key1{t}}
}
test "SUNION should handle non existing key as empty" {
r del set1{t} set2{t} set3{t}
r sadd set1{t} a b c
r sadd set2{t} b c d
assert_equal {a b c d} [lsort [r sunion set1{t} set2{t} set3{t}]]
}
test "SUNIONSTORE against non-set should throw error" {
r del set1{t} set2{t} set3{t} key1{t}
r set key1{t} x
# with en empty dstkey
assert_error "WRONGTYPE*" {r sunionstore set3{t} key1{t} noset{t}}
assert_equal 0 [r exists set3{t}]
assert_error "WRONGTYPE*" {r sunionstore set3{t} noset{t} key1{t}}
assert_equal 0 [r exists set3{t}]
# with a legal dstkey
r sadd set1{t} a b c
r sadd set2{t} b c d
r sadd set3{t} e
assert_error "WRONGTYPE*" {r sunionstore set3{t} key1{t} key2{t} noset{t}}
assert_equal 1 [r exists set3{t}]
assert_equal {e} [lsort [r smembers set3{t}]]
assert_error "WRONGTYPE*" {r sunionstore set3{t} noset{t} key1{t} key2{t}}
assert_equal 1 [r exists set3{t}]
assert_equal {e} [lsort [r smembers set3{t}]]
}
test "SUNIONSTORE should handle non existing key as empty" {
r del set1{t} set2{t} set3{t}
r set setres{t} xxx
assert_equal 0 [r sunionstore setres{t} foo111{t} bar222{t}]
assert_equal 0 [r exists setres{t}]
# set1 set2 both empty, should delete the dstkey
r sadd set3{t} a b c
assert_equal 0 [r sunionstore set3{t} set1{t} set2{t}]
assert_equal 0 [r exists set3{t}]
r sadd set1{t} a b c
r sadd set3{t} e f
assert_equal 3 [r sunionstore set3{t} set1{t} set2{t}]
assert_equal 1 [r exists set3{t}]
assert_equal {a b c} [lsort [r smembers set3{t}]]
r sadd set3{t} d
assert_equal 3 [r sunionstore set3{t} set2{t} set1{t}]
assert_equal 1 [r exists set3{t}]
assert_equal {a b c} [lsort [r smembers set3{t}]]
}
test "SUNIONSTORE against non existing keys should delete dstkey" {
@@ -403,10 +580,27 @@ start_server {
assert {[lsort $union] eq [lsort $content]}
}
test "SRANDMEMBER count of 0 is handled correctly" {
r srandmember myset 0
} {}
test "SRANDMEMBER with <count> against non existing key" {
r srandmember nonexisting_key 100
} {}
# Make sure we can distinguish between an empty array and a null response
r readraw 1
test "SRANDMEMBER count of 0 is handled correctly - emptyarray" {
r srandmember myset 0
} {*0}
test "SRANDMEMBER with <count> against non existing key - emptyarray" {
r srandmember nonexisting_key 100
} {*0}
r readraw 0
foreach {type contents} {
hashtable {
1 5 10 50 125 50000 33959417 4775547 65434162
@@ -632,6 +826,28 @@ start_server {
lsort [r smembers set]
} {a b c}
test "SMOVE only notify dstset when the addition is successful" {
r del srcset{t}
r del dstset{t}
r sadd srcset{t} a b
r sadd dstset{t} a
r watch dstset{t}
r multi
r sadd dstset{t} c
set r2 [redis_client]
$r2 smove srcset{t} dstset{t} a
# The dstset is actually unchanged, multi should success
r exec
set res [r scard dstset{t}]
assert_equal $res 2
$r2 close
}
tags {slow} {
test {intsets implementation stress testing} {
for {set j 0} {$j < 20} {incr j} {
+19
View File
@@ -199,6 +199,15 @@ start_server {
assert {[r EXISTS otherstream] == 0}
}
test {XADD with LIMIT delete entries no more than limit} {
r del yourstream
for {set j 0} {$j < 3} {incr j} {
r XADD yourstream * xitem v
}
r XADD yourstream MAXLEN ~ 0 limit 1 * xitem v
assert {[r XLEN yourstream] == 4}
}
test {XRANGE COUNT works as expected} {
assert {[llength [r xrange mystream - + COUNT 10]] == 10}
}
@@ -525,6 +534,16 @@ start_server {
}
assert_error ERR* {r XTRIM mystream MAXLEN 1 LIMIT 30}
}
test {XTRIM with LIMIT delete entries no more than limit} {
r del mystream
r config set stream-node-max-entries 2
for {set j 0} {$j < 3} {incr j} {
r XADD mystream * xitem v
}
assert {[r XTRIM mystream MAXLEN ~ 0 LIMIT 1] == 0}
assert {[r XTRIM mystream MAXLEN ~ 0 LIMIT 2] == 2}
}
}
start_server {tags {"stream"} overrides {appendonly yes}} {
+91 -13
View File
@@ -960,6 +960,39 @@ start_server {tags {"zset"}} {
assert_equal 1 [r zcard z2]
}
test "Basic ZPOP - $encoding RESP3" {
r hello 3
r del z1
create_zset z1 {0 a 1 b 2 c 3 d}
assert_equal {a 0.0} [r zpopmin z1]
assert_equal {d 3.0} [r zpopmax z1]
r hello 2
}
test "ZPOP with count - $encoding RESP3" {
r hello 3
r del z1
create_zset z1 {0 a 1 b 2 c 3 d}
assert_equal {{a 0.0} {b 1.0}} [r zpopmin z1 2]
assert_equal {{d 3.0} {c 2.0}} [r zpopmax z1 2]
r hello 2
}
test "BZPOP - $encoding RESP3" {
r hello 3
set rd [redis_deferring_client]
create_zset zset {0 a 1 b 2 c}
$rd bzpopmin zset 5
assert_equal {zset a 0} [$rd read]
$rd bzpopmin zset 5
assert_equal {zset b 1} [$rd read]
$rd bzpopmax zset 5
assert_equal {zset c 2} [$rd read]
assert_equal 0 [r exists zset]
r hello 2
}
r config set zset-max-ziplist-entries $original_max_entries
r config set zset-max-ziplist-value $original_max_value
}
@@ -1568,6 +1601,19 @@ start_server {tags {"zset"}} {
r zrange z1 5 0 BYSCORE REV LIMIT 0 2 WITHSCORES
} {d 4 c 3}
test {ZRANGESTORE - src key missing} {
set res [r zrangestore z2{t} missing{t} 0 -1]
assert_equal $res 0
r exists z2{t}
} {0}
test {ZRANGESTORE - src key wrong type} {
r zadd z2{t} 1 a
r set foo{t} bar
assert_error "*WRONGTYPE*" {r zrangestore z2{t} foo{t} 0 -1}
r zrange z2{t} 0 -1
} {a}
test {ZRANGESTORE - empty range} {
set res [r zrangestore z2 z1 5 6]
assert_equal $res 0
@@ -1616,6 +1662,20 @@ start_server {tags {"zset"}} {
return $res
}
# Check whether the zset members belong to the zset
proc check_member {mydict res} {
foreach ele $res {
assert {[dict exists $mydict $ele]}
}
}
# Check whether the zset members and score belong to the zset
proc check_member_and_score {mydict res} {
foreach {key val} $res {
assert_equal $val [dict get $mydict $key]
}
}
foreach {type contents} "ziplist {1 a 2 b 3 c} skiplist {1 a 2 b 3 [randstring 70 90 alpha]}" {
set original_max_value [lindex [r config get zset-max-ziplist-value] 1]
r config set zset-max-ziplist-value 10
@@ -1654,6 +1714,19 @@ start_server {tags {"zset"}} {
r zrandmember nonexisting_key 100
} {}
# Make sure we can distinguish between an empty array and a null response
r readraw 1
test "ZRANDMEMBER count of 0 is handled correctly - emptyarray" {
r zrandmember myzset 0
} {*0}
test "ZRANDMEMBER with <count> against non existing key - emptyarray" {
r zrandmember nonexisting_key 100
} {*0}
r readraw 0
foreach {type contents} "
skiplist {1 a 2 b 3 c 4 d 5 e 6 f 7 g 7 h 9 i 10 [randstring 70 90 alpha]}
ziplist {1 a 2 b 3 c 4 d 5 e 6 f 7 g 7 h 9 i 10 j} " {
@@ -1676,25 +1749,29 @@ start_server {tags {"zset"}} {
# PATH 1: Use negative count.
# 1) Check that it returns repeated elements with and without values.
# 2) Check that all the elements actually belong to the original zset.
set res [r zrandmember myzset -20]
assert_equal [llength $res] 20
check_member $mydict $res
set res [r zrandmember myzset -1001]
assert_equal [llength $res] 1001
check_member $mydict $res
# again with WITHSCORES
set res [r zrandmember myzset -20 withscores]
assert_equal [llength $res] 40
check_member_and_score $mydict $res
set res [r zrandmember myzset -1001 withscores]
assert_equal [llength $res] 2002
check_member_and_score $mydict $res
# Test random uniform distribution
# df = 9, 40 means 0.00001 probability
set res [r zrandmember myzset -1000]
assert_lessthan [chi_square_value $res] 40
# 2) Check that all the elements actually belong to the original zset.
foreach {key val} $res {
assert {[dict exists $mydict $key]}
}
check_member $mydict $res
# 3) Check that eventually all the elements are returned.
# Use both WITHSCORES and without
@@ -1710,7 +1787,7 @@ start_server {tags {"zset"}} {
} else {
set res [r zrandmember myzset -3]
foreach key $res {
dict append auxset $key $val
dict append auxset $key
}
}
if {[lsort [dict keys $mydict]] eq
@@ -1726,11 +1803,13 @@ start_server {tags {"zset"}} {
set res [r zrandmember myzset $size]
assert_equal [llength $res] 10
assert_equal [lsort $res] [lsort [dict keys $mydict]]
check_member $mydict $res
# again with WITHSCORES
set res [r zrandmember myzset $size withscores]
assert_equal [llength $res] 20
assert_equal [lsort $res] [lsort $mydict]
check_member_and_score $mydict $res
}
# PATH 3: Ask almost as elements as there are in the set.
@@ -1742,18 +1821,17 @@ start_server {tags {"zset"}} {
#
# We can test both the code paths just changing the size but
# using the same code.
foreach size {8 2} {
foreach size {1 2 8} {
# 1) Check that all the elements actually belong to the
# original set.
set res [r zrandmember myzset $size]
assert_equal [llength $res] $size
check_member $mydict $res
# again with WITHSCORES
set res [r zrandmember myzset $size withscores]
assert_equal [llength $res] [expr {$size * 2}]
# 1) Check that all the elements actually belong to the
# original set.
foreach ele [dict keys $res] {
assert {[dict exists $mydict $ele]}
}
check_member_and_score $mydict $res
# 2) Check that eventually all the elements are returned.
# Use both WITHSCORES and without