1. 08 Oct, 2018 1 commit
    • antirez's avatar
      Transactions: move the QUEUED logic inside call(). (Work in progress) · 71c37605
      antirez authored
      As a side effect this makes things like #5201 fixed automatically.
      The problem with handling it in ProcessQueryBuffer() or similar is that
      a client that is handled just via the call() interface will misbehave,
      and call()-ing MULTI and later some command will see just the other
      command executed like if the client was not in a transaction.
      
      For now this has worked well enough but probably it makes more sense if
      we can handle that case on call().
      
      WARNING: this comment is not enough and will introduce bugs. It's a WIP.
      71c37605
  2. 02 Oct, 2018 1 commit
  3. 30 Sep, 2018 2 commits
    • Bruce Merry's avatar
      Fix incorrect memory usage accounting in zrealloc · 8fd1031b
      Bruce Merry authored
      When HAVE_MALLOC_SIZE is false, each call to zrealloc causes used_memory
      to increase by PREFIX_SIZE more than it should, due to mis-matched
      accounting between the original zmalloc (which includes PREFIX size in
      its increment) and zrealloc (which misses it from its decrement).
      
      I've also supplied a command-line test to easily demonstrate the
      problem. It's not wired into the test framework, because I don't know
      TCL so I'm not sure how to automate it.
      8fd1031b
    • Oran Agra's avatar
      fix #5024 - commandstats for multi-exec were logged as EXEC. · f03aed3c
      Oran Agra authored
      this was broken a while back by ba9154d7
      the purpose of which was to fix commandstats for GEOADD
      f03aed3c
  4. 19 Sep, 2018 2 commits
  5. 12 Sep, 2018 1 commit
  6. 11 Sep, 2018 2 commits
  7. 05 Sep, 2018 3 commits
  8. 31 Aug, 2018 2 commits
    • antirez's avatar
      While the slave is busy, just accumulate master input. · 9ab91b8c
      antirez authored
      Processing command from the master while the slave is in busy state is
      not correct, however we cannot, also, just reply -BUSY to the
      replication stream commands from the master. The correct solution is to
      stop processing data from the master, but just accumulate the stream
      into the buffers and resume the processing later.
      
      Related to #5297.
      9ab91b8c
    • antirez's avatar
      Allow scripts to timeout on slaves as well. · f5b29c64
      antirez authored
      See reasoning in #5297.
      f5b29c64
  9. 29 Aug, 2018 2 commits
  10. 27 Aug, 2018 2 commits
  11. 14 Aug, 2018 1 commit
  12. 13 Aug, 2018 1 commit
  13. 03 Aug, 2018 1 commit
  14. 31 Jul, 2018 4 commits
  15. 30 Jul, 2018 3 commits
  16. 23 Jul, 2018 2 commits
  17. 20 Jul, 2018 2 commits
  18. 19 Jul, 2018 8 commits