--cNdxnHkX5QqsyA0e
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Tue, Jan 10, 2006 at 10:58:38AM +0900, YAMAMOTO Takashi wrote:
> > just now when stress testing UDF i got a panic in subr_pool:817 (pool_get). 
> > Inspecting this line its a SCHED_ASSERT_UNLOCKED(). I dont have a trace 
> > since the machine got wacky but in my recollection it was in vnode or buf 
> > shortage and asked for a clean vnode. This vnode was then recycled from FFS 
> > and then it got wrong.
> 
> the panic means that someone is calling pool_get with sched_lock held.
> why do you think it's related to vnode or buf?

Another datapoint, propably related.

When i was stresstesting my UDF again with multiple processes accessing the 
same directorytree simultaniously while transfering heaps of data from it 
and printing heaps of data on the console while in the background a 
compilation job was running, the copier suddenly got stuck on [vnode] and 
all the other UDF accessing processes got stuck on the node translation 
lock i inserted. Thus that can only mean that vget(vp, LK_EXCLUSIVE | 
LK_INTERLOCK) got stuck in this or getnewvnode(). Could this be a sheduler 
induced lock conflict? like recently seen on ftp.n.o ?

Reinoud


--cNdxnHkX5QqsyA0e
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (NetBSD)

iQEVAwUBQ8bVeIKcNwBDyKpoAQJesggAvNdGfG2uctPAhfQivj4jKGJr29eBLouS
FfirBU58D8TD/AyLp9zi7MhvHxw2lS7gPlONabTRssavWJ96oXMFnbtUnbRqdvsz
F8lbiwX3KGd0fRqGv2xYsgFA2vE+9fer3QK6o0xJlNQchqqCCcnKz2+5J+JOGRTw
QkFG2idjfbu3JEZPH9QA861Xw2pPfg3czTxN9LKibTebjRqCA8nt7X8AQfen7V4S
SOz9zQI81VJ7gWdnQyWLjUeiUuPlK1KuExCtpsyrYqddlg973fLkyMt2Vxt85vqi
xXJdvNxwQ78tsjLf4hDgJUmyqhHne40N5yQwvO+HjpvsA5gCZpB0WQ==
=3enb
-----END PGP SIGNATURE-----

--cNdxnHkX5QqsyA0e--