Commit Graph

6580 Commits

Author SHA1 Message Date
antirez
fa707ca154 Streams: more advanced XADD and XRANGE tests. 2017-12-01 10:24:24 +01:00
antirez
7a41b402c1 Streams: basic XADD tests. 2017-12-01 10:24:24 +01:00
antirez
26d4f8e3ec Streams: AOF rewriting + minor iterator improvements. 2017-12-01 10:24:24 +01:00
antirez
01ea018c40 Streams: export iteration API. 2017-12-01 10:24:24 +01:00
antirez
9ed40f0fc3 Streams: implement streamReplyWithRange() in terms of the iterator. 2017-12-01 10:24:24 +01:00
antirez
a58733cacf Streams: stream iteration refactoring, WIP 2. 2017-12-01 10:24:24 +01:00
antirez
b1ec333633 Streams: stream iteration refactoring, WIP 1. 2017-12-01 10:24:24 +01:00
antirez
1a603e1a87 Streams: fix bug in XREAD last received ID processing. 2017-12-01 10:24:24 +01:00
antirez
94af55c5ea Streams: fix memory leak in freeStream(). 2017-12-01 10:24:24 +01:00
antirez
3a0b78bc52 Streams: rewrite XADD ID argument for AOF/slaves. 2017-12-01 10:24:24 +01:00
antirez
19b06935d5 Streams: fix XADD API and keyspace notifications.
XADD was suboptimal in the first incarnation of the command, not being
able to accept an ID (very useufl for replication), nor options for
having capped streams.

The keyspace notification for streams was not implemented.
2017-12-01 10:24:24 +01:00
antirez
db89f7474d Streams: When XREAD blocks without COUNT, set a default one.
A client may lose a lot of time between invocations of blocking XREAD,
for example because it is processing the messages or for any other
cause. When it returns back, it may provide a low enough message ID that
the server will block to send an unreasonable number of messages in a
single call. For this reason we set a COUNT when the client is blocked
with XREAD calls, even if no COUNT is given. This is arbitrarily set to
1000 because it's enough to avoid slowing down the reception of many
messages, but low enough to avoid to block.
2017-12-01 10:24:24 +01:00
antirez
c128190026 Streams: fix handleClientsBlockedOnKeys() access to invalid ID. 2017-12-01 10:24:24 +01:00
antirez
6468cb2e82 Streams: fix XREAD ready-key signaling.
With lists we need to signal only on key creation, but streams can
provide data to clients listening at every new item added.
To make this slightly more efficient we now track different classes of
blocked clients to avoid signaling keys when there is nobody listening.
A typical case is when the stream is used as a time series DB and
accessed only by range with XRANGE.
2017-12-01 10:24:24 +01:00
antirez
b5be5093fe Streams: fix XREAD timeout handling, zero is valid. 2017-12-01 10:24:24 +01:00
antirez
2cacdcd6f8 Streams: XREAD related code to serve blocked clients. 2017-12-01 10:24:24 +01:00
antirez
0adb43b68f Streams: XREAD ability to block fixed. 2017-12-01 10:24:24 +01:00
antirez
6a1c92d52d Streams: synchronous xread fixes and improvements. 2017-12-01 10:24:24 +01:00
antirez
a7d898334a Streams: XREAD get-key method fixed. 2017-12-01 10:24:24 +01:00
antirez
110041825c Streams: XREAD get-keys method. 2017-12-01 10:24:24 +01:00
antirez
fa61720d30 Streams: XREAD, first draft. Handling of blocked clients still missing. 2017-12-01 10:24:24 +01:00
antirez
e65b4825f0 Streams: XREAD arguments parsing. 2017-12-01 10:24:24 +01:00
antirez
4086dff477 Streams: augment client.bpop with XREAD specific fields. 2017-12-01 10:24:24 +01:00
antirez
f80dfbf464 Streams: more internal preparation for blocking XREAD. 2017-12-01 10:24:24 +01:00
antirez
4a377cecd8 Streams: initial work to use blocking lists logic for streams XREAD. 2017-12-01 10:24:24 +01:00
antirez
439120c620 Streams: implement stream object release. 2017-12-01 10:24:24 +01:00
antirez
ec9bbe96bf Streams: XLEN command. 2017-12-01 10:24:24 +01:00
antirez
98d184db12 Streams: Save stream->length in RDB. 2017-12-01 10:24:24 +01:00
antirez
cd18f06e9c Streams: change listpack allocator to zmalloc. 2017-12-01 10:24:24 +01:00
antirez
edd70c1993 Streams: RDB loading. RDB saving modified.
After a few attempts it looked quite saner to just add the last item ID
at the end of the serialized listpacks, instead of scanning the last
listpack loaded from head to tail just to fetch it. It's a disk space VS
CPU-and-simplicity tradeoff basically.
2017-12-01 10:24:24 +01:00
antirez
485014cc74 Streams: RDB saving. 2017-12-01 10:24:24 +01:00
antirez
100d43c1ac Streams: assign value of 6 to OBJ_STREAM + some refactoring. 2017-12-01 10:24:24 +01:00
antirez
79866a6361 Streams: 12 commits squashed into the initial Streams implementation. 2017-12-01 10:24:24 +01:00
antirez
045d65c3af PSYNC2: Fix off by one buffer size in luaCreateFunction(). 2017-11-30 18:38:29 +01:00
antirez
452ad2e928 PSYNC2: just store script bodies into RDB.
Related to #4483. As suggested by @soloestoy, we can retrieve the SHA1
from the body. Given that in the new implementation using AUX fields we
ended copying around a lot to create new objects and strings, extremize
such concept and trade CPU for space inside the RDB file.
2017-11-30 18:38:26 +01:00
antirez
28dfdca733 PSYNC2: luaCreateFunction() should handle NULL client parameter.
See #4483. This is needed because luaCreateFunction() is now called
from RDB loading code outside a client context.
2017-11-30 18:37:52 +01:00
antirez
f11a7585a8 PSYNC2: Save Lua scripts state into RDB file.
This is currently needed in order to fix #4483, but this can be
useful in other contexts, so maybe later we may want to remove the
conditionals and always save/load scripts.

Note that we are using the "lua" AUX field here, in order to guarantee
backward compatibility of the RDB file. The unknown AUX fields must be
discarded by past versions of Redis.
2017-11-30 18:37:52 +01:00
antirez
6fb04d4637 Regression test: Slave restart with EVALSHA in backlog issue #4483. 2017-11-30 18:37:10 +01:00
antirez
3b9be93fda Prevent corruption of server.executable after DEBUG RESTART.
Doing the following ended with a broken server.executable:

1. Start Redis with src/redis-server
2. Send CONFIG SET DIR /tmp/
3. Send DEBUG RESTART

At this point we called execve with an argv[0] that is no longer related
to the new path. So after the restart the absolute path of the
executable is recomputed in the wrong way. With this fix we pass the
absolute path already computed as argv[0].
2017-11-30 18:30:06 +01:00
antirez
d8f8701032 Be more verbose when DEBUG RESTART fails. 2017-11-30 18:08:21 +01:00
zhaozhao.zz
43be967690 networking: optimize unlinkClient() in freeClient() 2017-11-30 18:11:05 +08:00
zhaozhao.zz
1b5f56d042 aof: cast sdslen to ssize_t 2017-11-30 10:27:12 +08:00
zhaozhao.zz
2d73cf2367 aof: fix the short write 2017-11-30 10:22:12 +08:00
Itamar Haber
0752a834f9 Check arity in SLOWLOG before accessing arg 2017-11-30 00:30:30 +02:00
antirez
2785d6caa0 Merge branch 'lfu-fixes' into unstable 2017-11-29 17:16:13 +01:00
Itamar Haber
51d42c88a6 Merge branch 'unstable' into help_subcommands 2017-11-28 21:28:40 +02:00
Itamar Haber
59d52f7fab Standardizes the 'help' subcommand
This adds a new `addReplyHelp` helper that's used by commands
when returning a help text. The following commands have been
touched: DEBUG, OBJECT, COMMAND, PUBSUB, SCRIPT and SLOWLOG.

WIP

Fix entry command table entry for OBJECT for HELP option.

After #4472 the command may have just 2 arguments.

Improve OBJECT HELP descriptions.

See #4472.

WIP 2

WIP 3
2017-11-28 21:15:45 +02:00
Salvatore Sanfilippo
565e139a56
Merge pull request #4200 from jeesyn/fix_typo
fix a typo
2017-11-28 18:44:11 +01:00
Salvatore Sanfilippo
b998e2035c
Merge pull request #4165 from zyegfryed/patch-1
Fix some typos
2017-11-28 18:43:45 +01:00
Salvatore Sanfilippo
923502a70b
Merge pull request #4166 from charpty/wip-redisclic-typo
redis-cli.c typo: helpe -> helper.
2017-11-28 18:41:51 +01:00