[Gluster-devel] Problem with namespace-cache in 2.5

[Anand Avati]

Harris,
  this bug got fixed a while back. Are you still facing the same issue?

thanks,
avati

2007/6/22, Harris Landgarten <harrisl at lhjonline.com>:
>
> I am testing 2.5-patch-194. I have a two brick setup with ns setup as a
> second volume on the second brick.
>
> I am exporting /mnt/export on the bricks and /mnt/namespace on the ns
> volume
>
> All seems ok but the namespace cache is not filling completely.
>
> [root at domU-12-31-35-00-36-41:~] find /mnt/export/ -type d -ls | wc -l
> 3947
> [root at domU-12-31-35-00-36-41:~] find /mnt/namespace/ -type d -ls | wc -l
> 103
>
>
> This is causing read failures on the clients
>
> sudo find /mnt/glusterfs/secondary/ -type d -ls
> 4440068    4 drwxr-xr-x   3 harrisl  harrisl      4096 Jun 22 10:33
> /mnt/glusterfs/secondary/
> 4440069    4 drwxr-xr-x  10 harrisl  harrisl      4096 Jun 22 10:36
> /mnt/glusterfs/secondary/0
> 4440160    4 drwxr-xr-x   3 harrisl  harrisl      4096 Jun 22 10:36
> /mnt/glusterfs/secondary/0/5
> 4440161    4 drwxr-xr-x  11 harrisl  harrisl      4096 Jun 22 10:36
> /mnt/glusterfs/secondary/0/5/msg
> 4440177   92 drwxr-xr-x   2 harrisl  harrisl     90112 Jun 22 10:37
> /mnt/glusterfs/secondary/0/5/msg/0
> 4440176  100 drwxr-xr-x   2 harrisl  harrisl     98304 Jun 22 10:41
> /mnt/glusterfs/secondary/0/5/msg/1
> 4440175  100 drwxr-xr-x   2 harrisl  harrisl     98304 Jun 22 10:41
> /mnt/glusterfs/secondary/0/5/msg/2
> 4440174   96 drwxr-xr-x   2 harrisl  harrisl     94208 Jun 22 10:41
> /mnt/glusterfs/secondary/0/5/msg/3
> hangs
>
> 2007-06-22 11:17:07 D [client-protocol.c:4499:notify] client-ns: got
> GF_EVENT_CHILD_UP
> 2007-06-22 11:17:07 D [client-protocol.c:4263:client_protocol_handshake_reply]
> client1: reply frame has callid: 424242
> 2007-06-22 11:17:07 D [client-protocol.c:4297:client_protocol_handshake_reply]
> client1: SETVOLUME on remote-host succeeded
> 2007-06-22 11:17:07 D [client-protocol.c:4263:client_protocol_handshake_reply]
> client2: reply frame has callid: 424242
> 2007-06-22 11:17:07 D [client-protocol.c:4297:client_protocol_handshake_reply]
> client2: SETVOLUME on remote-host succeeded
> 2007-06-22 11:17:07 D [client-protocol.c:4263:client_protocol_handshake_reply]
> client-ns: reply frame has callid: 424242
> 2007-06-22 11:17:07 D [client-protocol.c:4297:client_protocol_handshake_reply]
> client-ns: SETVOLUME on remote-host succeeded
> 2007-06-22 11:17:07 D [fuse-bridge.c:403:fuse_getattr] glusterfs-fuse:
> GETATTR 4440161 (/secondary/0/5/msg)
> 2007-06-22 11:17:07 D [fuse-bridge.c:316:fuse_lookup] glusterfs-fuse:
> LOOKUP 4440161/3 (/secondary/0/5/msg/3)
> 2007-06-22 11:17:07 D [inode.c:478:__create_inode] client-ns/inode: create
> inode(4440174)
> 2007-06-22 11:17:07 D [inode.c:278:__active_inode] client-ns/inode:
> activating inode(4440174), lru=0/1000
> 2007-06-22 11:17:07 D [inode.c:478:__create_inode] bricks/inode: create
> inode(4440174)
> 2007-06-22 11:17:07 D [inode.c:278:__active_inode] bricks/inode:
> activating inode(4440174), lru=0/100
> 2007-06-22 11:17:07 D [inode.c:478:__create_inode] client2/inode: create
> inode(5521417)
> 2007-06-22 11:17:07 D [inode.c:278:__active_inode] client2/inode:
> activating inode(5521417), lru=0/1000
> 2007-06-22 11:17:07 D [inode.c:478:__create_inode] client1/inode: create
> inode(5046281)
> 2007-06-22 11:17:07 D [inode.c:278:__active_inode] client1/inode:
> activating inode(5046281), lru=0/1000
> 2007-06-22 11:17:07 D [fuse-bridge.c:267:fuse_entry_cbk] glusterfs-fuse:
> ENTRY => 4440174
> 2007-06-22 11:17:07 D [inode.c:478:__create_inode] fuse/inode: create
> inode(4440174)
> 2007-06-22 11:17:07 D [inode.c:278:__active_inode] fuse/inode: activating
> inode(4440174), lru=0/0
> 2007-06-22 11:17:07 W [client-protocol.c:3957:client_protocol_reconnect]
> client2: breaking reconnect chain
> 2007-06-22 11:17:07 W [client-protocol.c:3957:client_protocol_reconnect]
> client-ns: breaking reconnect chain
> 2007-06-22 11:17:07 W [client-protocol.c:3957:client_protocol_reconnect]
> client1: breaking reconnect chain
>
>
> I am also getting this on direct reads on files on another client:
>
>
> 2007-06-22 11:17:07 D [fuse-bridge.c:267:fuse_entry_cbk] glusterfs-fuse:
> ENTRY => 4440161
> 2007-06-22 11:17:07 D [fuse-bridge.c:316:fuse_lookup] glusterfs-fuse:
> LOOKUP 4440161/5 (/secondary/0/5/msg/5)
> 2007-06-22 11:17:07 D [fuse-bridge.c:267:fuse_entry_cbk] glusterfs-fuse:
> ENTRY => 4440161
> 2007-06-22 11:17:07 D [fuse-bridge.c:316:fuse_lookup] glusterfs-fuse:
> LOOKUP 4440161/2 (/secondary/0/5/msg/2)
> 2007-06-22 11:17:07 D [fuse-bridge.c:267:fuse_entry_cbk] glusterfs-fuse:
> ENTRY => 4440161
> 2007-06-22 11:17:07 D [fuse-bridge.c:316:fuse_lookup] glusterfs-fuse:
> LOOKUP 4440161/5 (/secondary/0/5/msg/5)
> 2007-06-22 11:17:07 D [fuse-bridge.c:267:fuse_entry_cbk] glusterfs-fuse:
> ENTRY => 4440161
> 2007-06-22 11:17:07 D [fuse-bridge.c:316:fuse_lookup] glusterfs-fuse:
> LOOKUP 4440161/5 (/secondary/0/5/msg/5)
> 2007-06-22 11:17:07 D [fuse-bridge.c:293:fuse_entry_cbk] glusterfs-fuse:
> ERR => -1 (2)
> 2007-06-22 11:17:49 D [fuse-bridge.c:316:fuse_lookup] glusterfs-fuse:
> LOOKUP 1/secondary (/secondary)
> 2007-06-22 11:18:07 D [fuse-bridge.c:316:fuse_lookup] glusterfs-fuse:
> LOOKUP 1/secondary (/secondary)
>
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>



-- 
Anand V. Avati