1. 29 Apr, 2020 2 commits
    • antirez's avatar
      Comment clearly why we moved some code in #6623. · d2a6c5ff
      antirez authored
      d2a6c5ff
    • srzhao's avatar
      fix pipelined WAIT performance issue. · 4024bc7e
      srzhao authored
      If client gets blocked again in `processUnblockedClients`, redis will not send
      `REPLCONF GETACK *` to slaves untill next eventloop, so the client will be
      blocked for 100ms by default(10hz) if no other file event fired.
      
      move server.get_ack_from_slaves sinppet after `processUnblockedClients`, so
      that both the first WAIT command that puts client in blocked context and the
      following WAIT command processed in processUnblockedClients would trigger
      redis-sever to send `REPLCONF GETACK *`, so that the eventloop would get
      `REPLCONG ACK <reploffset>` from slaves and unblocked ASAP.
      4024bc7e
  2. 28 Apr, 2020 2 commits
    • Guy Benoish's avatar
      Extend XINFO STREAM output · 1e2aee39
      Guy Benoish authored
      Introducing XINFO STREAM <key> FULL
      1e2aee39
    • Oran Agra's avatar
      hickup, re-fix dictEncObjKeyCompare · 9a3dab0a
      Oran Agra authored
      come to think of it, in theory (not in practice), getDecodedObject can
      return the same original object with refcount incremented, so the
      pointer comparision in the previous commit was invalid.
      so now instead of checking the encoding, we explicitly check the
      refcount.
      9a3dab0a
  3. 27 Apr, 2020 4 commits
    • antirez's avatar
      Rework comment in dictEncObjKeyCompare(). · 31781e97
      antirez authored
      31781e97
    • Oran Agra's avatar
      allow dictFind using static robj · 09a5c078
      Oran Agra authored
      since the recent addition of OBJ_STATIC_REFCOUNT and the assertion in
      incrRefCount it is now impossible to use dictFind using a static robj,
      because dictEncObjKeyCompare will call getDecodedObject which tries to
      increment the refcount just in order to decrement it later.
      09a5c078
    • Oran Agra's avatar
      Keep track of meaningful replication offset in replicas too · 4447ddc8
      Oran Agra authored
      Now both master and replicas keep track of the last replication offset
      that contains meaningful data (ignoring the tailing pings), and both
      trim that tail from the replication backlog, and the offset with which
      they try to use for psync.
      
      the implication is that if someone missed some pings, or even have
      excessive pings that the promoted replica has, it'll still be able to
      psync (avoid full sync).
      
      the downside (which was already committed) is that replicas running old
      code may fail to psync, since the promoted replica trims pings form it's
      backlog.
      
      This commit adds a test that reproduces several cases of promotions and
      demotions with stale and non-stale pings
      
      Background:
      The mearningful offset on the master was added recently to solve a problem were
      the master is left all alone, injecting PINGs into it's backlog when no one is
      listening and then gets demoted and tries to replicate from a replica that didn't
      have any of the PINGs (or at least not the last ones).
      
      however, consider this case:
      master A has two replicas (B and C) replicating directly from it.
      there's no traffic at all, and also no network issues, just many pings in the
      tail of the backlog. now B gets promoted, A becomes a replica of B, and C
      remains a replica of A. when A gets demoted, it trims the pings from its
      backlog, and successfully replicate from B. however, C is still aware of
      these PINGs, when it'll disconnect and re-connect to A, it'll ask for something
      that's not in the backlog anymore (since A trimmed the tail of it's backlog),
      and be forced to do a full sync (something it didn't have to do before the
      meaningful offset fix).
      
      Besides that, the psync2 test was always failing randomly here and there, it
      turns out the reason were PINGs. Investigating it shows the following scenario:
      
      cycle 1: redis #1 is master, and all the rest are direct replicas of #1
      cycle 2: redis #2 is promoted to master, #1 is a replica of #2 and #3 is replica of #1
      now we see that when #1 is demoted it prints:
      17339:S 21 Apr 2020 11:16:38.523 * Using the meaningful offset 3929963 instead of 3929977 to exclude the final PINGs (14 bytes difference)
      17339:S 21 Apr 2020 11:16:39.391 * Trying a partial resynchronization (request e2b3f8817735fdfe5fa4626766daa938b61419e5:3929964).
      17339:S 21 Apr 2020 11:16:39.392 * Successful partial resynchronization with master.
      and when #3 connects to the demoted #2, #2 says:
      17339:S 21 Apr 2020 11:16:40.084 * Partial resynchronization not accepted: Requested offset for secondary ID was 3929978, but I can reply up to 3929964
      
      so the issue here is that the meaningful offset feature saved the day for the
      demoted master (since it needs to sync from a replica that didn't get the last
      ping), but it didn't help one of the other replicas which did get the last ping.
      4447ddc8
    • antirez's avatar
      Fix STRALGO command flags. · 3497fd00
      antirez authored
      3497fd00
  4. 25 Apr, 2020 3 commits
  5. 24 Apr, 2020 3 commits
    • antirez's avatar
      LCS -> STRALGO LCS. · 8a7f255c
      antirez authored
      STRALGO should be a container for mostly read-only string
      algorithms in Redis. The algorithms should have two main
      characteristics:
      
      1. They should be non trivial to compute, and often not part of
      programming language standard libraries.
      2. They should be fast enough that it is a good idea to have optimized C
      implementations.
      
      Next thing I would love to see? A small strings compression algorithm.
      8a7f255c
    • Oran Agra's avatar
      optimize memory usage of deferred replies · fb732f7a
      Oran Agra authored
      When deffered reply is added the previous reply node cannot be used so
      all the extra space we allocated in it is wasted. in case someone uses
      deffered replies in a loop, each time adding a small reply, each of
      these reply nodes (the small string reply) would have consumed a 16k
      block.
      now when we add anther diferred reply node, we trim the unused portion
      of the previous reply block.
      
      see #7123
      fb732f7a
    • antirez's avatar
      57a0c9c9
  6. 23 Apr, 2020 5 commits
  7. 22 Apr, 2020 4 commits
  8. 21 Apr, 2020 3 commits
  9. 20 Apr, 2020 5 commits
  10. 18 Apr, 2020 1 commit
    • zhenwei pi's avatar
      Threaded IO: set thread name for redis-server · 5010da6a
      zhenwei pi authored
      
      
      Set thread name for each thread of redis-server, this helps us to
      monitor the utilization and optimise the performance.
      
      And suggested-by Salvatore, implement this feature for multi
      platforms. Currently support linux and bsd, ignore other OS.
      
      An exmaple on Linux:
       # top -d 5 -p `pidof redis-server ` -H
      
          PID USER      PR  NI    VIRT    RES    SHR S %CPU %MEM     TIME+ COMMAND
      3682671 root      20   0  227744   8248   3836 R 99.2  0.0   0:19.53 redis-server
      3682677 root      20   0  227744   8248   3836 S 26.4  0.0   0:04.15 io_thd_3
      3682675 root      20   0  227744   8248   3836 S 23.6  0.0   0:03.98 io_thd_1
      3682676 root      20   0  227744   8248   3836 S 23.6  0.0   0:03.97 io_thd_2
      3682672 root      20   0  227744   8248   3836 S  0.2  0.0   0:00.02 bio_close_file
      3682673 root      20   0  227744   8248   3836 S  0.2  0.0   0:00.02 bio_aof_fsync
      3682674 root      20   0  227744   8248   3836 S  0.0  0.0   0:00.00 bio_lazy_free
      3682678 root      20   0  227744   8248   3836 S  0.0  0.0   0:00.00 jemalloc_bg_thd
      3682682 root      20   0  227744   8248   3836 S  0.0  0.0   0:00.00 jemalloc_bg_thd
      3682683 root      20   0  227744   8248   3836 S  0.0  0.0   0:00.00 jemalloc_bg_thd
      3682684 root      20   0  227744   8248   3836 S  0.0  0.0   0:00.00 jemalloc_bg_thd
      3682685 root      20   0  227744   8248   3836 S  0.0  0.0   0:00.00 jemalloc_bg_thd
      3682687 root      20   0  227744   8248   3836 S  0.0  0.0   0:00.00 jemalloc_bg_thd
      
      Another exmaple on FreeBSD-12.1:
        PID USERNAME    PRI NICE   SIZE    RES STATE    C   TIME    WCPU COMMAND
       5212 root        100    0    48M  7280K CPU2     2   0:26  99.52% redis-server{redis-server}
       5212 root         38    0    48M  7280K umtxn    4   0:06  26.94% redis-server{io_thd_3}
       5212 root         36    0    48M  7280K umtxn    6   0:06  26.84% redis-server{io_thd_1}
       5212 root         39    0    48M  7280K umtxn    1   0:06  25.30% redis-server{io_thd_2}
       5212 root         20    0    48M  7280K uwait    3   0:00   0.00% redis-server{redis-server}
       5212 root         21    0    48M  7280K uwait    2   0:00   0.00% redis-server{bio_close_file}
       5212 root         21    0    48M  7280K uwait    3   0:00   0.00% redis-server{bio_aof_fsync}
       5212 root         21    0    48M  7280K uwait    0   0:00   0.00% redis-server{bio_lazy_free}
      Signed-off-by: default avatarzhenwei pi <pizhenwei@bytedance.com>
      5010da6a
  11. 17 Apr, 2020 2 commits
  12. 16 Apr, 2020 3 commits
  13. 15 Apr, 2020 3 commits