Lines Matching refs:to
7 LOW: No need to upgrade unless there are new features you want to use.
25 as expected, but worth to fix ASAP anyway.
34 Fix a bug to delay bgsave while AOF rewrite in progress for replication
38 Update linenoise to fix insecure redis-cli history file creation.
51 1. There was a bug in the List type implementation, able to cause the
52 crash of the server under certain (non trivial to replicate) circumstances
54 that was able to easily trigger the bug was added to the test suite.
60 This is not really a bug fix, but may allow to more easily detect that
61 a slave is able to fail over its master reducing certain delays.
64 process could spawn at the same time. This is known to trash disk I/O,
65 AOF performances, and to ultimately create latency in the Redis server.
66 Normally Redis avoids to have the two writing children at the same time, but
68 fix with me) where the double-fork could happen. In order to fix this bug
69 non trivial changes to the replication code were operated, however it was
70 important to back port this fix into 3.2.2 because the bug could lead to
75 in the points reported by GEORADIUS. The errors were found to be related
76 to three different bugs (one of these was a bug in the test itself).
77 It's not a critical bug: the effect is to, sometimes, don't report objects
87 and `ROLE` command as having arbitrary IP and port. This allows to have
92 Redis uses in order to load the dataset in memory, so a given version
93 of Redis is always able to check the RDB it produced... without another
94 external check tool which is supposed to be taken in sync with the
96 is able to spot more complex bugs, since it really loads the dataset
99 About the redis-check-dump utility, now it is also able to show certain
110 The ability to check how much memory was used by the instance that produced
114 modified in order to be a bit less verbose compared to the last releases.
120 Ability of slave to announce arbitrary ip/port to master.
124 redis-benchmark: new option to show server errors on stdout.
180 Make tcp-keepalive default to 300 in internal conf.
196 Added a trivial program to randomly corrupt RDB files in /utils.
235 Upgrade urgency HIGH: Critical fix to Redis Sentinel, due to 3.2.0 regression
236 compared to 3.0.
239 maturity to Redis 3.2. The list of commits following this note will tell
244 refactoring of Redis Sentinel, in order to implement connection sharing
245 to make Sentinel able to scale better (few Sentinels to monitor many
247 commands in the Redis link. This in turn resulted into an inability to talk
249 of Redis Sentinel to reconfigure back the old master, after a failover,
251 to the inability to talk with the old master at all.
259 5. Jemalloc now is really configured to save you memory, for a problem a
264 LRU evictions purposes. A new TOUCH command was introduced *just* to
267 7. A bug was fixed in redis-cli, that connected to the instance running on the
278 release, but nothing critical AFAIK, and I wanted to release this one ASAP,
300 This bug most experienced effect was an inability of Redis to
301 reconfigure back old masters to slaves after they are reachable again
302 after a failover. This was due to failing to reset the count of the
308 Many thanks to people reporting the issue, and especially to
317 | redis-cli: really connect to the right server.
324 the arguments, defaulting to instance 6379. After the connection is
328 whatever port you specify is ignored and 6379 is connected to instead.
356 code in to `rdb.c` appears to be capable as always to restore data from
358 current version in order to correctly restore data, is while loading the
361 For the above reasons, this commit enables RESTORE to accept older
388 | Minor aesthetic fixes to PR #3264.
393 Comment format fixed + local var modified from camel case to underscore
407 optimization: lookup key only once, and grow at once to the max need
425 | fix check when can't send the command to the promoted slave
525 I'm the author of this line but I can't see a good reason for it to
532 computed (which is invalid) only then it was trimmed to 25 instead of
548 apparently possible to write it in a more correct way.
571 Thanks to @oranagra for noticing it was missing.
607 | fix comment "b>a" to "a > b"
642 The existing `R` flag appears to be sufficient and there's no apparent reason why the command shoul…
662 | Code to access object string bytes repeated 3x refactored into 1 function.
712 | Actually use --with-lg-quantum=3 to build jemalloc.
732 Use the COMMAND output to fill with partial information the built-in
733 help. This makes redis-cli able to at least complete commands that are
734 exported by the Redis server it is connected to, but were not available
755 Upgrade urgency HIGH: 3.2-RC users should upgrade ASAP to this more
770 could cause data written via scripts to be lost during reshardings.
781 * redis-cli show hits about the commands arguments to the right.
783 * GEORADIUS got a STORE / STOREDIST option to store the result into a target
784 key (as as orted set) instead of reporting it to the user.
787 the past a slave could migrate only to a master that used to have slaves
851 | Fix a possible race condition of sdown detection if the connection to master/slave/sentinel decam…
884 | Cluster test 12: reshard back just a few slots to speedup the test.
925 Fix issue in case the redirect address is in ipv6 format. Parse from behind to extract last part of…
935 Thanks to @tushar2708 for the PR. I applied a slightly different fix.
936 Thanks to @cespare for reporting.
945 | Quick fix to avoid false positive in replica migration test.
971 I've renamed maxmemoryToString to evictPolicyToString since that is
972 more accurate (and easier to mentally connect with the correct data), as
973 well as updated the function to user server.maxmemory_policy rather than
1028 Thanks to @oranagra for the idea of allowing CONFIG GET during loading.
1040 Thanks to @oranagra for the hint about misplaced 'r' flags.
1120 | additional fix to issue #2948
1135 Thanks to @tidwall for reporting.
1148 1/5 of all the cluster slots from master 0 to the other 4 masters and
1163 to other nodes, the master remains empty and the slaves migrate away
1164 to other masters with non-zero nodes. However the old master now empty,
1166 no way to tell it had slaves in the past.
1171 slaves, they are automatically considered to be targets for replicas
1228 As a side effect, cat commands.txt | redis-cli now is able to handle
1241 The old code mis-calculated the amount of time to wait till next event.
1247 | ae.c: comment to explain why we have a useless maxId check.
1257 | Fix ae.c to avoid timers infinite loop.
1263 test case that was used to verify the fix.
1266 a timer event (that is the time after it want to run again) is added
1267 to the event *start time*. So if the event takes, in order to run, more
1268 than the time it says it want to be scheduled again for running, an
1355 We want to report the original command in the stats, for example GEOADD,
1398 The new bitfield command is an extension to the Redis bit operations,
1402 (up to 64 bit signed integers and 63 bit unsigned integers).
1416 command with sub-commands similar to what Yoav envisioned, using short
1474 | Fix GEORADIUS STORE/DIST refcount after backport to 3.2.
1499 Related to issue #3019.
1530 | cmsgpack: pass correct osize values to lua allocator, update correct buf free space in cmsgpack
1577 to reference a node in an unique way, so they were relying on CLUSTER
1593 (Salvatore Sanfilippo, thanks to Paweł Kowalczykf or the original
1596 to avoid initiating inwanted handshakes.
1602 * [FIX] Many fixes to MIGRATE multiple keys implementation. The command
1603 could handle errors in a faulty way leading to crashes or other
1610 * [FIX] Fix redis-trib rebalance when nodes need to be left empty because
1615 always be free to move keys between nodes about open slots, in order
1616 to reshard, fix the cluster configuration, and so forth.
1620 * [FIX] Redis-cli implementation of Lua debugging now allows to use the
1621 SCRIPT DEBUG command directly, switching to debugging mode as needed.
1623 * [FIX] Redis-trib is now able to fix more errors. A new CLUSTER subcommand
1624 called BUMPEPOCH was introduced in order to support new modes
1627 harder to run Redis in a configuration that is unsecure because no
1628 firewalling was used in order to protect Redis from external accesses.
1630 * [NEW] Cluster/Sentinel tests now use OSX leak to perform leak detection
1643 to its internals. It's still fresh code compared to 3.0.
1647 * [NEW] Lua scripts "effect replication". Makes possible to write scripts
1650 * [NEW] Lua scripts selective replication. Makes possible to replicate to
1656 Matt Stancliff "imported back" the work to Redis and created the
1668 * [NEW] Modify Jemalloc size classes to make certain Redis objects fit
1671 keys. The slaves are only able to logically consider a key expired
1677 savings and storage space in RDB gains (up to 10x sometimes).
1682 and in order to make the replication of a this class of commands,
1687 * [NEW] Faster RDB loading via the RESIZEDB opcode to avoid useless hash tables
1693 * [NEW] CLIENT KILL TYPE MASTER, to kill (disconnect) masters from slaves.
1695 * [NEW] Jemalloc updated to 4.0.3 (Salvatore Sanfilippo)
1704 All the Redis Cluster changes in 3.2 were backported to 3.0, so there is
1717 * [NEW] Sentinel connection sharing. Makes Sentinels able to scale to
1720 * More things backported to Redis 3.0 in the past, so no longer news of 3.2.
1722 Migrating from 3.0 to 3.2
1726 upgrading your application from 3.0 to 3.2. However this is a list of small
1729 * The default configuration file now binds to 127.0.0.1.
1730 * Slaves try to no longer expose stale data about already expired keys.
1731 * The RDB format changed. Redis 3.2 is still able to read 3.0 (and all the
1740 Salvatore Sanfilippo. Thanks to Redis Labs for making all this possible.
1741 Also many thanks to all the other contributors and the amazing community