diff options
| author | Milind Changire <mchangir@redhat.com> | 2018-04-09 11:26:05 +0530 |
|---|---|---|
| committer | Raghavendra G <rgowdapp@redhat.com> | 2018-04-13 03:23:47 +0000 |
| commit | 31da3f1d8513b10d0449a1855fbad5060bd6191d (patch) | |
| tree | 9a91b13f4843ebe4a0d91bf9c11236ace6fe41d3 /rpc/rpc-transport | |
| parent | 1f0765242a689980265c472646c64473a92d94c0 (diff) | |
rpc: set listen-backlog to high value
Problem:
On node reboot, when glusterd starts volumes rapidly, there's a flood of
connections from the bricks to glusterd and from the self-heal daemons
to the bricks. This causes SYN Flooding and dropped connections when the
listen-backlog is not enough to hold the pending connections to
compensate for the rate at which connections are accepted by the RPC
layer.
Solution:
Increase the listen-backlog value to 1024. This is a partial solution.
Part of the solution is to rearm the listener socket early for quicker
accept() of connections.
See commit 6964640a977cb10c0c95a94e03c229918fa6eca8 (change 19833)
Change-Id: I62283d1f4990dd43839f9a6932cf8a36effd632c
fixes: bz#1564600
Signed-off-by: Milind Changire <mchangir@redhat.com>
Diffstat (limited to 'rpc/rpc-transport')
| -rw-r--r-- | rpc/rpc-transport/socket/src/socket.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/rpc/rpc-transport/socket/src/socket.c b/rpc/rpc-transport/socket/src/socket.c index a000b26e515..34bdbdea9d0 100644 --- a/rpc/rpc-transport/socket/src/socket.c +++ b/rpc/rpc-transport/socket/src/socket.c @@ -4794,7 +4794,7 @@ struct volume_options options[] = { .description = "This option uses the value of backlog argument that " "defines the maximum length to which the queue of " "pending connections for socket fd may grow.", - .default_value = "10", + .default_value = "1024", }, { .key = {"transport.tcp-user-timeout"}, .type = GF_OPTION_TYPE_INT, |