Date: Wed, 10 Jun 2009 19:03:33 +0200 From: Andre Oppermann <andre@freebsd.org> To: Scott Long <scottl@samsco.org> Cc: freebsd-stable@freebsd.org Subject: Re: (probe23:arcmsr0:0:16:0): inquiry data fails comparison at DV1 step Message-ID: <4A2FE765.9000803@freebsd.org> In-Reply-To: <4A2FC858.5050800@samsco.org> References: <4A2FA4EF.5060603@freebsd.org> <4A2FC858.5050800@samsco.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Scott Long wrote: > Andre Oppermann wrote: >> Today I upgraded an AMD64 server from FreeBSD 6.4 to 7-STABLE and got >> a big >> nasty surprise with the Areca RAID controller: >> >> arcmsr0: <Areca SATA Host Adapter RAID Controller (RAID6 capable) >> > mem 0xfdbff000-0xfdb400000 irq16 at device 14.0 on pci18 >> ARECA RAID ADAPTER0: Driver Version 1.20.00.15 2007-10-07 >> ARECA RAID ADAPTER0: FIRMWARE VERSION V1.46 2009-01-06 >> arcmsr0: [ITHREAD] >> ... >> (probe23:arcmsr0:0:16:0): inquiry data fails comparison at DV1 step >> run_interrupt_driven_hooks: still waiting after 60 seconds for xpt_config >> >> From here on it hangs and repeats the last message with some sort of >> exponential backoff. Ctrl-Alt-Del to reboot doesn't work but ScrollLock >> does. Hard reset is required to reboot. >> >> Booting the old 6.4 kernel works and the system comes up again with full >> access to the RAID array. >> >> Any help appreciated. >> > > I'll try to reproduce this locally. There have been sporadic reports of > the "fails comparison" problem, but none as fatal as this. Is it > possible to compile your kernel with CAMDEBUG and > CAM_DEBUG_FLAGS=CAM_DEBUG_INFO? CAMDEBUG doesn't give any output. As it turns out the hang seems to be related to an interrupt routing problem somehow. I did an BIOS upgrade and BIOS settings factory reset. The mainboard is an ASUS M2N32 WS Professional with an Athlon64 X2 4800+. Enabling all on-board devices including ATA and various SATA controllers together with "PnP OS" set to yes fixed the hand and allowed FreeBSD 7 to boot sucessfully. Why 6.4 doesn't stumble I don't know. The "(probe23:arcmsr0:0:16:0): inquiry data fails comparison at DV1 step" line is still there. Fortunately the boot process simply continues from there without further trouble. If there are any other CAM debug options or patches you want me to try I can do that. The system is only in partial production and I'm allowed to reboot it during workhours. -- Andre
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4A2FE765.9000803>