]> git.apps.os.sepia.ceph.com Git - ceph-ci.git/commit
osd/OSD: auto mark heartbeat sessions as stale and tear them down
authorxie xingguo <xie.xingguo@zte.com.cn>
Wed, 26 Jun 2019 06:24:08 +0000 (14:24 +0800)
committerxie xingguo <xie.xingguo@zte.com.cn>
Sat, 29 Jun 2019 00:14:13 +0000 (08:14 +0800)
commit6cc90f363b8096d2d5fad30e57426d0cea9e3478
treefa301ff59e7f4acee612833b7f7cd5a01f2cf11f
parent8e8db703172fc9bccd96b7de344d6a7d761b7862
osd/OSD: auto mark heartbeat sessions as stale and tear them down

The primary benefit is that the OSD doesn't need to keep a flood of
blocked heartbeat messages around in memory.
This prevents OSDs from accumulating heartbeat messages due to a
broken switch and then exhausting the whole node's memory:

Jun 11 04:19:26 host-192-168-9-12 kernel: [409881.137077] Out of memory:
Kill process 1471476 (ceph-osd) score 47 or sacrifice child
Jun 11 04:19:26 host-192-168-9-12 kernel: [409881.146054] Killed process
1471476 (ceph-osd) total-vm:4822548kB, anon-rss:3097860kB,
file-rss:2556kB, shmem-rss:0kB

Fixes: http://tracker.ceph.com/issues/40586
Signed-off-by: xie xingguo <xie.xingguo@zte.com.cn>
src/common/options.cc
src/osd/OSD.cc
src/osd/OSD.h