[PATCH] knfsd: Fix bug in recent lockd patches that can cause reclaim to fail
authorNeilBrown <neilb@suse.de>
Tue, 17 Oct 2006 07:10:17 +0000 (00:10 -0700)
committerLinus Torvalds <torvalds@g5.osdl.org>
Tue, 17 Oct 2006 15:18:46 +0000 (08:18 -0700)
When an nfs server shuts down, lockd needs to release all the locks even
though the client still holds them.

It should therefore not 'unmonitor' the clients, so that the files in nfs/sm
will still be there when the nfs server restarts, so that those clients will
be told to reclaim their locks.

However the hosts are fully unmonitored, so statd may well remove the files.

lockd has a test for 'sm_sticky' and avoid the unmonitor call if it is set,
but it is currently not set.

So set it when tearing down lockd.

Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
fs/lockd/svcsubs.c

index 514f5f20701ea3d3cc8ba0410dd8fd3098e71c87..c5f9113cdc70e89d84c4789e8c1271bf1e52a7b6 100644 (file)
@@ -324,7 +324,16 @@ nlmsvc_same_host(struct nlm_host *host, struct nlm_host *other)
 static int
 nlmsvc_is_client(struct nlm_host *host, struct nlm_host *dummy)
 {
-       return host->h_server;
+       if (host->h_server) {
+               /* we are destroying locks even though the client
+                * hasn't asked us too, so don't unmonitor the
+                * client
+                */
+               if (host->h_nsmhandle)
+                       host->h_nsmhandle->sm_sticky = 1;
+               return 1;
+       } else
+               return 0;
 }
 
 /*