Date: Wed, 18 Apr 2007 04:20:16 GMT From: Rong-En Fan <rafan@FreeBSD.org> To: freebsd-bugs@FreeBSD.org Subject: Re: kern/111766: "panic: ffs_blkfree: freeing free block" during disk activity Message-ID: <200704180420.l3I4KGvB040825@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR kern/111766; it has been noted by GNATS. From: Rong-En Fan <rafan@FreeBSD.org> To: Kris Kennaway <kris@obsecurity.org> Cc: Rong-En Fan <rafan@FreeBSD.org>, FreeBSD-gnats-submit@FreeBSD.org Subject: Re: kern/111766: "panic: ffs_blkfree: freeing free block" during disk activity Date: Wed, 18 Apr 2007 12:14:35 +0800 On Wed, Apr 18, 2007 at 12:06:15AM -0400, Kris Kennaway wrote: > On Wed, Apr 18, 2007 at 12:01:53PM +0800, Rong-En Fan wrote: > > I think I tried once but in vain. Some previous backtrace > > at http://www.rafan.org/FreeBSD/panic/freeing_free_block/ > > I haven't see this since last Oct, but the filesystem does not > > get full recently (we got larger disks). > > Unfortunately I dont think the backtrace is very useful, it's > reporting damage that occured at some earlier time. > > > >From previous experience, usually the panic is triggered on > > the full partition but I saw once or twice that it's triggered > > on another non-full filesystem. Let me clear it a bit. When it's triggered on non-full filesystem, I have another filesystem is full and some clients trying to write to the full one. > I am not entirely sure whether it is really due to a full filesystem, > because I've also come across it in random circumstances. I have also > spent some time stressing full filesystems without causing the panic. > If you can come up with a way to reliably produce it, that would be > very useful (possibly essential). Perhaps in the summer, I can find some free time to do this. Just curious, have you tried exported the filesystem and stress via nfs client?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200704180420.l3I4KGvB040825>