2.5.43 smp bootup crash
2.5.43 smp (with Andrew Morton's md-fix)
didn't survive bootup.
It produced a backtrace so long that most of it
scrolled off the screen, before stating that
it didn't sync in an interrupt handler.
The machine was dead, even sysrq+B couldn't
reboot it.
The kernel has preempt, and autodetection of
raid-1 and raid-0.
Helge Hafting
Re: 2.5.43 smp bootup crash, more info - probably scsi/raid
Quoted 4 lines · Helge Hafting
It produced a backtrace so long that most of it
scrolled off the screen, before stating that
it didn't sync in an interrupt handler.
Some of the functions in the trace was scsi stuff. I have a tekram scsi controller, driven by "SYM53C8XX Version 2 SCSI support"
The crash happens immediately after initializing the controller and discovering the two disks. This is where autodetection
of RAID usually happens.
So it seems to me that it is either some scsi problem, or a RAID problem.
The problem affects both 2.5.43 and 2.5.43-mm2.
Helge Hafting
Re: 2.5.43 smp bootup crash, more info - probably scsi/raid
In article <[email protected]>,
Helge Hafting <[email protected]> wrote:
| Helge Hafting wrote: | | > It produced a backtrace so long that most of it | > scrolled off the screen, before stating that | > it didn't sync in an interrupt handler. | > | Some of the functions in the trace was scsi stuff. | I have a tekram scsi controller, driven by | "SYM53C8XX Version 2 SCSI support" | | The crash happens immediately after initializing | the controller and discovering the two | disks. This is where autodetection | of RAID usually happens. | So it seems to me that it is either some | scsi problem, or a RAID problem. | | The problem affects both 2.5.43 and 2.5.43-mm2.
Is it an OOPS or just a BUG? I got a BUG until I applied the patch, now my SCSI devices don't actually work, but I don't get an OOPS.
See my other post on the patch, or write my off-list if you can't find it.
Signature · 3 lines
bill davidsen <[email protected]> CTO, TMR Associates, Inc Doing interesting things with little computers since 1979.
Re: 2.5.43 smp bootup crash, more info - probably scsi/raid
Quoted 2 lines · bill davidsen
In article <[email protected]>,
Helge Hafting <[email protected]> wrote:
[...]
Quoted 4 lines
| The problem affects both 2.5.43 and 2.5.43-mm2.
Is it an OOPS or just a BUG? I got a BUG until I applied the patch, now my SCSI devices don't actually work, but I don't get an OOPS.
Hard to say - the first part scrolls off screen immediately, then the machine hangs. All I see is the last part of a backtrace and a machine that don't respond to anything but the reset and power switches.
This is fixed in 2.5.44 though, there were some known issues with the md driver and vfs.
Helge Hafting