https://github.com/torvalds/linux
Revision 5bdca4e0768d3e0f4efa43d9a2cc8210aeb91ab9 authored by Sage Weil on 10 July 2012, 18:53:34 UTC, committed by Sage Weil on 18 July 2012, 02:35:59 UTC
In ancient times, the messenger could both initiate and accept connections.
An artifact if that was data structures to store/process an incoming
ceph_msg_connect request and send an outgoing ceph_msg_connect_reply.
Sadly, the negotiation code was referencing those structures and ignoring
important information (like the peer's connect_seq) from the correct ones.

Among other things, this fixes tight reconnect loops where the server sends
RETRY_SESSION and we (the client) retries with the same connect_seq as last
time.  This bug pretty easily triggered by injecting socket failures on the
MDS and running some fs workload like workunits/direct_io/test_sync_io.

Signed-off-by: Sage Weil <sage@inktank.com>
1 parent a018540
History
Tip revision: 5bdca4e0768d3e0f4efa43d9a2cc8210aeb91ab9 authored by Sage Weil on 10 July 2012, 18:53:34 UTC
libceph: fix messenger retry
Tip revision: 5bdca4e
File Mode Size
Documentation
arch
block
crypto
drivers
firmware
fs
include
init
ipc
kernel
lib
mm
net
samples
scripts
security
sound
tools
usr
virt
.gitignore -rw-r--r-- 1014 bytes
.mailmap -rw-r--r-- 4.4 KB
COPYING -rw-r--r-- 18.3 KB
CREDITS -rw-r--r-- 92.7 KB
Kbuild -rw-r--r-- 2.5 KB
Kconfig -rw-r--r-- 252 bytes
MAINTAINERS -rw-r--r-- 208.2 KB
Makefile -rw-r--r-- 46.0 KB
README -rw-r--r-- 17.3 KB
REPORTING-BUGS -rw-r--r-- 3.3 KB

README

back to top