A couple of days ago I dropped Debian on an Indy I've had lying around
for a while. I'd done the first stage of the install, rebooted, and
left it sitting at the last half of the Debian installer for a while.
When I looked at it an hour or two later, the kernel had spat out the
following:
Eeek! page_mapcount(page) went negative! (-1)
page->flags = 14
page->count = 0
page->mapping = 00000000
Break instruction in kernel code[#1]:
Cpu 0
$ 0 : 00000000 1000cc00 ffffffff 8820fe38
$ 4 : 8820fe38 8820fe54 00000001 bf0f0000
$ 8 : 00002085 88260000 0084100b 014f0000
$12 : 89187972 ffffffff 00000029 88260000
$16 : 89125460 1000d000 89125460 10034000
$20 : 89d83d30 88832100 ffffffbf 00000000
$24 : 88260000 88110960
$28 : 89d82000 89d83d08 89333200 8805b06c
Hi : 00000000
Lo : 00000027
epc : 880628b8 page_remove_rmap+0xd8/0xe4 Not tainted
ra : 8805b06c unmap_vmas+0x508/0x680
Status: 1000cc03 KERNEL EXL IE
Cause : 00000024
PrId : 00002020
Modules linked in: unix
Process sh (pid: 7040, threadinfo=89d82000, task=8a114ad8)
[ etc .. full dmesg output is at http://plaz.net.nz/indy-dmesg.txt ]
I think the error may have appeared when the /etc/cron.daily/ scripts
started running.
I'm running a hand-compiled 2.6.16.11 (I had to drop a new kernel on
halfway through the installer because the Debian one has software raid
built as modules)
Is the message above common? As I said, the Indy has been sitting
around doing nothing for quite a while, so the hardware may be a little
flaky. I haven't had a chance to test the SIMMs in there, or find
others to swap in to test.
Thanks,
Sam
signature.asc
Description: This is a digitally signed message part
|