Date: Sun, 20 Nov 2011 08:41:28 -0800 From: Jeremy Chadwick <freebsd@jdc.parodius.com> To: Johannes Totz <johannes@jo-t.de> Cc: freebsd-stable@freebsd.org Subject: Re: panic: make_dev_credv: bad si_name (error=17, si_name=pass2) Message-ID: <20111120164128.GA29952@icarus.home.lan> In-Reply-To: <jaba31$fqn$1@dough.gmane.org> References: <jab6mc$phv$1@dough.gmane.org> <20111120160307.GA20262@icarus.home.lan> <jaba31$fqn$1@dough.gmane.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Nov 20, 2011 at 04:32:33PM +0000, Johannes Totz wrote: > On 20/11/2011 16:03, Jeremy Chadwick wrote: > >On Sun, Nov 20, 2011 at 03:34:36PM +0000, Johannes Totz wrote: > >>(Sent twice, first one bounced...) > >>Just got a panic on 9-stable, running inside VirtualBox, trying to > >>build a release-set. Don't know yet if reproducable, just happened a > >>few minutes ago. > >>The whole core.txt stuff follows below (beware of line-breaks): > >> > >>... > >>Unread portion of the kernel message buffer: > >>(ada1:ahcich0:0:0:0): lost device > >>panic: make_dev_credv: bad si_name (error=17, si_name=pass2) > >>cpuid = 0 > >>KDB: stack backtrace: > >>#0 0xc0a4aff7 at kdb_backtrace+0x47 > >>#1 0xc0a185c7 at panic+0x117 > >>#2 0xc09d05de at make_dev_credv+0x9e > >>#3 0xc09d080a at make_dev+0x4a > >>#4 0xc04b79e0 at passregister+0x230 > >>#5 0xc048ece3 at cam_periph_alloc+0x4e3 > >>#6 0xc04b7525 at passasync+0x85 > >>#7 0xc0490442 at xpt_async_bcast+0x32 > >>#8 0xc0492715 at xpt_async+0x105 > >>#9 0xc04991f3 at probedone+0xc33 > >>#10 0xc04958a1 at camisr_runqueue+0x2e1 > >>#11 0xc04959ff at camisr+0x13f > >>#12 0xc09ed69b at intr_event_execute_handlers+0x13b > >>#13 0xc09eee5a at ithread_loop+0x7a > >>#14 0xc09ea8a7 at fork_exit+0x97 > >>#15 0xc0d32734 at fork_trampoline+0x8 > >>... > >>Uptime: 7m57s > >>(ada1:ahcich0:0:0:0): Synchronize cache failed > > > >According to the above, your ada1 "virtual disk" fell off the bus > >entirely. Relevant storage bits taken from your dmesg: > > > >>atapci0:<Intel PIIX4 UDMA33 controller> port 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xd000-0xd00f at device 1.1 on pci0 > >>ata0:<ATA channel 0> on atapci0 > >>ata1:<ATA channel 1> on atapci0 > >>ahci0:<Intel ICH8M AHCI SATA controller> port 0xd040-0xd047,0xd050-0xd057,0xd060-0xd06f mem 0xf0806000-0xf0807fff irq 5 at device 13.0 on pci0 > >>ahci0: AHCI v1.10 with 1 3Gbps ports, Port Multiplier not supported > >>ahcich0:<AHCI channel> at channel 0 on ahci0 > >>... > >>ada0 at ata0 bus 0 scbus0 target 0 lun 0 > >>ada0:<VBOX HARDDISK 1.0> ATA-6 device > >>ada0: 33.300MB/s transfers (UDMA2, PIO 65536bytes) > >>ada0: 20480MB (41943040 512 byte sectors: 16H 63S/T 16383C) > >>ada0: Previously was known as ad0 > >>ada1 at ahcich0 bus 0 scbus2 target 0 lun 0 > >>ada1:<VBOX HARDDISK 1.0> ATA-6 SATA 2.x device > >>ada1: 300.000MB/s transfers (SATA 2.x, UDMA6, PIO 8192bytes) > >>ada1: Command Queueing enabled > >>ada1: 20480MB (41943040 512 byte sectors: 16H 63S/T 16383C) > >>ada1: Previously was known as ad4 > > > >My recommendation is to remove VirtualBox from the picture entirely and > >instead do whatever you were doing (running zfstest) on bare metal. > > Yeah, would love to. > But http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/155587 prevents > me from doing so. I'm trying to build a live-disc that includes > http://svnweb.freebsd.org/base?view=revision&revision=226617 so I > can recover the pool. Can you not use mfsBSD for this task? http://mfsbsd.vx.sk/ The 9.0-RELEASE image provided by mm@ on his site is dated 2011/10/26, while SVN rev 226617 was committed on 2011/10/21 (5 days prior). You might also want to try a 9.0-RC2 build, which was just announced and put on the mirrors a couple days ago: ftp://ftp.freebsd.org/pub/FreeBSD/releases/i386/i386/ ftp://ftp.freebsd.org/pub/FreeBSD/releases/amd64/amd64/ (Comment in passing: I still have no idea why the architecture directories for 9.0 are "doubled up" like that; it almost looks like the result of someone using rsync with a missing trailing slash on the source directory). I can't explain why your ada1 "virtual disk" would drop off the bus entirely. Possibly too much I/O and stress via VirtualBox causes too long a delay, resulting in ahci.ko or CAM bits dropping the idsk from the bus. kern.cam.ada.default_timeout is, by default, 30 seconds, which is an awful long time, but I don't know if that is the only timeout available (e.g. no idea what ahci.ko uses internally, I'd need to check the code). -- | Jeremy Chadwick jdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, US | | Making life hard for others since 1977. PGP 4BD6C0CB |
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20111120164128.GA29952>