staging: lustre: ptlrpc: prevent request timeout grow due to recovery
authorMikhail Pershin <mike.pershin@intel.com>
Sun, 18 Sep 2016 20:38:53 +0000 (16:38 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 19 Sep 2016 08:08:22 +0000 (10:08 +0200)
commitbe2479818ff227fbb6800935909b65e06159735c
tree5b55e56def08b826022168916e2c0a5bc126cf80
parentbbc222b113fb5bed066a9bcc4ce7c1a92528a47e
staging: lustre: ptlrpc: prevent request timeout grow due to recovery

Patch fixes the issue seen on the client with growing request
timeout which occurred after the server side patch landed for
LU-5079. While commit itself is correct, it reveals another
issue. If request is being processed for a long time on server
then client adaptive timeouts will adapt to that after receiving
reply and new requests will have bigger timeout. Another problem
is that server AT history is corrupted by recovery request
processing time which not pure service time but includes
also waiting time for clients to recover.

Patch prevents the AT stats update from early replies on client and
from recovering requests processing time on server.
The ptlrpc_at_recv_early_reply() still updates the current request
timeout as asked by server, but don't include this into AT stats.
The real reply will bring that data from server after all.

Signed-off-by: Mikhail Pershin <mike.pershin@intel.com>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-6084
Reviewed-on: http://review.whamcloud.com/13520
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Jian Yu <jian.yu@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/staging/lustre/lustre/ptlrpc/client.c
drivers/staging/lustre/lustre/ptlrpc/service.c