Re: tstile lockup

["J. Hannken-Illjes" <hannken%eis.cs.tu-bs.de@localhost>]

On Nov 19, 2012, at 4:53 PM, Edgar Fuß <ef%math.uni-bonn.de@localhost> wrote:

> OK, this is the svn process (directly running on the file server, not 
> operating 
> via NFS) tstile-ing:
> 
> crash> ps | grep "\(vnode\|tstile\)"
> 25051    1 3   0         0   fffffe82ec17d200                svn tstile
> crash> t/a fffffe82ec17d200
> trace: pid 25051 lid 1 at 0xfffffe811e901700
> sleepq_block() at sleepq_block+0xad
> turnstile_block() at turnstile_block+0x2bb
> rw_vector_enter() at rw_vector_enter+0x1eb
> wapbl_begin() at wapbl_begin+0x54
<snip>
> rw_vector_enter() at rw_vector_enter+0x1eb
> genfs_lock() at genfs_lock+0x7b
> VOP_LOCK() at VOP_LOCK+0x32
> vn_lock() at vn_lock+0x7d
> vget() at vget+0xa2
> ufs_ihashget() at ufs_ihashget+0x84
> ffs_vget() at ffs_vget+0xc1
> ufs_fhtovp() at ufs_fhtovp+0x1f
> ffs_fhtovp() at ffs_fhtovp+0x55
> VFS_FHTOVP() at VFS_FHTOVP+0x1c
> nfsrv_fhtovp() at nfsrv_fhtovp+0x9a
> nfsrv_getattr() at nfsrv_getattr+0x13a
> nfssvc_nfsd() at nfssvc_nfsd+0x1d7
> sys_nfssvc() at sys_nfssvc+0x22c
> syscall() at syscall+0xc4
> crash> ps | grep "\(vnode\|tstile\)"
> 442      4 3   5         0   fffffe841628c060              slave tstile
> 
> These are somewhat difficult to catch in the act since you have to copy-paste
> the address from the ps output into the t command.
> 
> Now, does that help to diagnose the problem?

Ok, so you see threads waiting for the wapbl rwlock and
some vnode rwlock.  This is ok as long as it doesn't deadlock.

Do you get a deadlock or will the system come back to work
after some time?

--
J. Hannken-Illjes - hannken%eis.cs.tu-bs.de@localhost - TU Braunschweig 
(Germany)