Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 18 May 2009 09:25:13 +0200
From:      Magnus Kling <klingfon@gmail.com>
To:        Alexander Motin <mav@freebsd.org>
Cc:        freebsd-current@freebsd.org
Subject:   Re: Kernel panic when reboot on server with a Promise SX4000 and two ATA disks RAID1.
Message-ID:  <43b1bb350905180025g682d3764qba5a450d85d8f961@mail.gmail.com>
In-Reply-To: <4A10F3E3.40306@FreeBSD.org>
References:  <43b1bb350905150939s5d503f00x27116e7ffe79a37@mail.gmail.com> <4A10F3E3.40306@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
2009/5/18 Alexander Motin <mav@freebsd.org>

> Hi.
>
> Magnus Kling wrote:
>
>> After having some trouble with ACPI kernel in 7.1, regarding booting with
>> a Promise SX4000 card in a RAID1 setup, I tried to upgrade to CURRENT to
>> test the bits that John Baldwin wrote and had commited to head.
>>  Result:
>> Well, it boots ok but on reboot I get a kernel panic after the disks have
>> made the sync.
>>  Attached is a bt and panic message.
>>
>
> According to backtrace, system tried to reference address 0xc while sending
> final flush command to the drive. ATA has no other shutdown actions except
> this, so any contexts and states should not be lost in any case. And as soon
> as your drive was detected, the controller is probably operable. Haven't you
> seen any other ATA error messages during boot or later?
>
> If my i386 kernel is more or less coherent to your's,
> ata_promise_sx4_command+0x39 address where it have crashed means
>        caddr_t window = rman_get_virtual(ctlr->r_res1);
> line, and especially 'ctlr->r_res1' part. So looks like
>        struct ata_pci_controller *ctlr = device_get_softc(gparent);
> returned NULL there. But the same technique used by many other ATA drivers
> without problems, so I am a bit surprised.
>
> Is the problem continuously repeatable? Does the controller operates well
> all other time before shutdown? Are you sure that you have rebuilt your
> kernel correctly?
>
> --
> Alexander Motin
>

Yes it is repeatable. Happens every reboot or shutdown. Controller works ok
all other time. (I have my homedir mounted to the raidarray.)

The source was synced and then I did a normal:
make buildworld
make buildkernel
make installkernel
reboot to singel user mode and mergemaster -p + make installworld +
mergemaster then reboot.

I will try to compile it again...

/Magnus



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?43b1bb350905180025g682d3764qba5a450d85d8f961>