lockd: don't clear sm_monitored on nsm_reboot_lookup
authorJeff Layton <jlayton@redhat.com>
Fri, 5 Feb 2010 20:09:22 +0000 (15:09 -0500)
committerJ. Bruce Fields <bfields@citi.umich.edu>
Mon, 8 Feb 2010 21:20:35 +0000 (16:20 -0500)
commit7e469af97eed947ba9204712601281a69ae8eb6c
treef4488e393995ae523ed6dea5370ddad7aa1505b5
parentcdd30fa1664e0245fa64330c7cc2ddab7e47c223
lockd: don't clear sm_monitored on nsm_reboot_lookup

When lockd gets a notify downcall from statd, it'll search its hosts
cache and then clear the sm_monitored bit on the host it finds. The idea
is apparently to make lockd redo a SM_MON on the next lock request.

This is unnecessary and causes the kernel's NSM cache to go out of sync
with statd. statd doesn't stop monitoring a host when it gets a
SM_NOTIFY and there's no guarantee that another lock will occur after
the reclaim and before the unmount. In that event, no SM_UNMON will
occur.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu>
fs/lockd/mon.c