* "swap_free: Unused swap offset entry 00000100" but no crash?
@ 2004-07-27 0:21 Marcin Owsiany
2004-07-27 12:30 ` Keith Owens
0 siblings, 1 reply; 6+ messages in thread
From: Marcin Owsiany @ 2004-07-27 0:21 UTC (permalink / raw)
To: linux-kernel
Hi!
The machine is a dual pentium 4, on an Intel motherboard, running kernel
2.4.26 with SMP. This is a mail smarthost. Load average is around 3 for
most of the day, and around 0.2 during night. The box has been running
for a few months wothout problems.
$ cat /proc/swaps
Filename Type Size Used Priority
/dev/rd/c0d0p1 partition 1461872 2556 -1
$
(c0d0 is a Mylex DAC960 array).
Suddenly on the 8th day of its uptime, it started to log the message
cited in subject after EVERY SINGLE /usr/sbin/cron invocation.
There is an almost perfect correlation: the number of
/USR/SBIN/CRON[PID]: (USER) CMD (blah blah)
lines in syslog is EXACTLY the same as the number of
kernel: swap_free: Unused swap offset entry 00000100
lines.
It does not matter what command, and for what user, is run by cron (even
a single "true" in crontab provokes the kernel message).
The message is logged by kernel at the moment that the new cron process
exits (I found that out when some long-running commands were invoked).
This has been happening constantly for 5 days now, there are no other
messages from kernel in syslog. We also have two more similar machines,
and only this one recently started showing that.
The only references to this message I could find with google were also
describing almost immediate crashes, and hardware problems were
suggested. But in my case the system works just fine.
I didn't try rebooting it, swapoff/on-ing or restarting cron yet - maybe
someone has some ideas on what to test to find out what caused this
before I do that?
Also, I would be grateful if someone could explain what is that number in the
message supposed to be? An address?
lspci -v:
00:00.0 Host bridge: Intel Corp.: Unknown device 2570 (rev 02)
Subsystem: ABIT Computer Corp.: Unknown device 100a
Flags: bus master, fast devsel, latency 0
Memory at e0000000 (32-bit, prefetchable) [size=128M]
Capabilities: [e4] #09 [2106]
Capabilities: [a0] AGP version 3.0
00:01.0 PCI bridge: Intel Corp.: Unknown device 2571 (rev 02) (prog-if 00 [Normal decode])
Flags: bus master, 66Mhz, fast devsel, latency 64
Bus: primary=00, secondary=01, subordinate=01, sec-latency=32
Memory behind bridge: f0000000-f1ffffff
Prefetchable memory behind bridge: e8000000-efffffff
00:1e.0 PCI bridge: Intel Corp. 82820 820 (Camino 2) Chipset PCI (rev c2) (prog-if 00 [Normal decode])
Flags: bus master, fast devsel, latency 0
Bus: primary=00, secondary=02, subordinate=03, sec-latency=32
I/O behind bridge: 00009000-00009fff
Memory behind bridge: f2000000-f3ffffff
Prefetchable memory behind bridge: f4000000-f47fffff
00:1f.0 ISA bridge: Intel Corp.: Unknown device 24d0 (rev 02)
Flags: bus master, medium devsel, latency 0
00:1f.1 IDE interface: Intel Corp.: Unknown device 24db (rev 02) (prog-if 8a [Master SecP PriP])
Subsystem: ABIT Computer Corp.: Unknown device 100a
Flags: bus master, medium devsel, latency 0, IRQ 16
I/O ports at <unassigned>
I/O ports at <unassigned>
I/O ports at <unassigned>
I/O ports at <unassigned>
I/O ports at f000 [size=16]
Memory at 40000000 (32-bit, non-prefetchable) [size=1K]
00:1f.3 SMBus: Intel Corp.: Unknown device 24d3 (rev 02)
Subsystem: ABIT Computer Corp.: Unknown device 100a
Flags: medium devsel, IRQ 17
I/O ports at 0500 [size=32]
01:00.0 VGA compatible controller: nVidia Corporation: Unknown device 0181 (rev c1) (prog-if 00 [VGA])
Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 16
Memory at f0000000 (32-bit, non-prefetchable) [size=16M]
Memory at e8000000 (32-bit, prefetchable) [size=128M]
Expansion ROM at <unassigned> [disabled] [size=128K]
Capabilities: [60] Power Management version 2
Capabilities: [44] AGP version 3.0
02:04.0 Ethernet controller: Intel Corp. 82557 [Ethernet Pro 100] (rev 08)
Subsystem: Intel Corp. EtherExpress PRO/100+ Management Adapter
Flags: bus master, medium devsel, latency 32, IRQ 20
Memory at f3100000 (32-bit, non-prefetchable) [size=4K]
I/O ports at 9000 [size=64]
Memory at f3000000 (32-bit, non-prefetchable) [size=1M]
Capabilities: [dc] Power Management version 2
02:06.0 PCI bridge: Intel Corp. 80960RP [i960 RP Microprocessor/Bridge] (rev 03) (prog-if 00 [Normal decode])
Flags: bus master, medium devsel, latency 32
Bus: primary=02, secondary=03, subordinate=03, sec-latency=32
02:06.1 RAID bus controller: Mylex Corporation DAC960PX (rev 03)
Subsystem: Mylex Corporation DAC960PX
Flags: bus master, medium devsel, latency 32, IRQ 22
Memory at f4000000 (32-bit, prefetchable) [size=8M]
Expansion ROM at <unassigned> [disabled] [size=32K]
free:
total used free shared buffers cached
Mem: 904348 507716 396632 0 34896 254208
-/+ buffers/cache: 218612 685736
Swap: 2000052 61860 1938192
Marcin
--
Marcin Owsiany <marcin@owsiany.pl> http://marcin.owsiany.pl/
GnuPG: 1024D/60F41216 FE67 DA2D 0ACA FC5E 3F75 D6F6 3A0D 8AA0 60F4 1216
"Every program in development at MIT expands until it can read mail."
-- Unknown
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: "swap_free: Unused swap offset entry 00000100" but no crash?
2004-07-27 0:21 "swap_free: Unused swap offset entry 00000100" but no crash? Marcin Owsiany
@ 2004-07-27 12:30 ` Keith Owens
2004-07-27 12:53 ` Robin Holt
0 siblings, 1 reply; 6+ messages in thread
From: Keith Owens @ 2004-07-27 12:30 UTC (permalink / raw)
To: Marcin Owsiany; +Cc: linux-kernel
On Tue, 27 Jul 2004 02:21:54 +0200,
Marcin Owsiany <marcin@owsiany.pl> wrote:
> kernel: swap_free: Unused swap offset entry 00000100
>Also, I would be grateful if someone could explain what is that number in the
>message supposed to be? An address?
It is a swap partition number, but I doubt that you have 256 swap
partitions in your system. Single bit set in a word that is meant to
be 0, most likely to be caused by a hardware single bit error. Run
memtest, burn86 or other memory verification checks.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: "swap_free: Unused swap offset entry 00000100" but no crash?
2004-07-27 12:30 ` Keith Owens
@ 2004-07-27 12:53 ` Robin Holt
2004-07-27 18:58 ` Marcelo Tosatti
0 siblings, 1 reply; 6+ messages in thread
From: Robin Holt @ 2004-07-27 12:53 UTC (permalink / raw)
To: Keith Owens; +Cc: Marcin Owsiany, linux-kernel
On Tue, Jul 27, 2004 at 10:30:02PM +1000, Keith Owens wrote:
> On Tue, 27 Jul 2004 02:21:54 +0200,
> Marcin Owsiany <marcin@owsiany.pl> wrote:
> > kernel: swap_free: Unused swap offset entry 00000100
> >Also, I would be grateful if someone could explain what is that number in the
> >message supposed to be? An address?
>
> It is a swap partition number, but I doubt that you have 256 swap
> partitions in your system. Single bit set in a word that is meant to
> be 0, most likely to be caused by a hardware single bit error. Run
> memtest, burn86 or other memory verification checks.
>
I remember a race condition I thought was possible, but couldn't exactly
pin down the exact sequence. Give me a chance to dig through some of
my notes and see what I come across.
I think I could understand this if there two messages with each invocation,
but not with one.
Marcin, you have a process with a Page Table Entry which indicates it is
pointing to a page which has been swapped out to block 0 of swap device
256. This is probably caused by a problem in the kernel. You can certainly
run memtest et al. If you don't find anything, I would assume the problem
is in the kernel.
Most of the code in the area you would be affected by has changed
drastically in the 2.6 kernel.
Good Luck,
Robin Holt
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: "swap_free: Unused swap offset entry 00000100" but no crash?
2004-07-27 12:53 ` Robin Holt
@ 2004-07-27 18:58 ` Marcelo Tosatti
2004-07-27 20:03 ` Robin Holt
2005-04-17 19:53 ` Marcin Owsiany
0 siblings, 2 replies; 6+ messages in thread
From: Marcelo Tosatti @ 2004-07-27 18:58 UTC (permalink / raw)
To: Robin Holt; +Cc: Keith Owens, Marcin Owsiany, linux-kernel
On Tue, Jul 27, 2004 at 07:53:04AM -0500, Robin Holt wrote:
> On Tue, Jul 27, 2004 at 10:30:02PM +1000, Keith Owens wrote:
> > On Tue, 27 Jul 2004 02:21:54 +0200,
> > Marcin Owsiany <marcin@owsiany.pl> wrote:
> > > kernel: swap_free: Unused swap offset entry 00000100
> > >Also, I would be grateful if someone could explain what is that number in the
> > >message supposed to be? An address?
> >
> > It is a swap partition number, but I doubt that you have 256 swap
> > partitions in your system. Single bit set in a word that is meant to
> > be 0, most likely to be caused by a hardware single bit error. Run
> > memtest, burn86 or other memory verification checks.
> >
>
> I remember a race condition I thought was possible, but couldn't exactly
> pin down the exact sequence. Give me a chance to dig through some of
> my notes and see what I come across.
Would love to hear more details about this.
> I think I could understand this if there two messages with each invocation,
> but not with one.
>
> Marcin, you have a process with a Page Table Entry which indicates it is
> pointing to a page which has been swapped out to block 0 of swap device
> 256. This is probably caused by a problem in the kernel. You can certainly
> run memtest et al. If you don't find anything, I would assume the problem
> is in the kernel.
Marcin, please run the memtest86 and report back.
> Most of the code in the area you would be affected by has changed
> drastically in the 2.6 kernel.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: "swap_free: Unused swap offset entry 00000100" but no crash?
2004-07-27 18:58 ` Marcelo Tosatti
@ 2004-07-27 20:03 ` Robin Holt
2005-04-17 19:53 ` Marcin Owsiany
1 sibling, 0 replies; 6+ messages in thread
From: Robin Holt @ 2004-07-27 20:03 UTC (permalink / raw)
To: Marcelo Tosatti; +Cc: Robin Holt, Keith Owens, Marcin Owsiany, linux-kernel
On Tue, Jul 27, 2004 at 03:58:59PM -0300, Marcelo Tosatti wrote:
> On Tue, Jul 27, 2004 at 07:53:04AM -0500, Robin Holt wrote:
> >
> > I remember a race condition I thought was possible, but couldn't exactly
> > pin down the exact sequence. Give me a chance to dig through some of
> > my notes and see what I come across.
>
> Would love to hear more details about this.
I can't find anything about this. I thought it had something to do with
the a race condition around the swap code, but can't recall much more.
Sorry,
Robin
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: "swap_free: Unused swap offset entry 00000100" but no crash?
2004-07-27 18:58 ` Marcelo Tosatti
2004-07-27 20:03 ` Robin Holt
@ 2005-04-17 19:53 ` Marcin Owsiany
1 sibling, 0 replies; 6+ messages in thread
From: Marcin Owsiany @ 2005-04-17 19:53 UTC (permalink / raw)
To: linux-kernel
On Tue, Jul 27, 2004 at 03:58:59PM -0300, Marcelo Tosatti wrote:
> On Tue, Jul 27, 2004 at 07:53:04AM -0500, Robin Holt wrote:
> > Marcin, you have a process with a Page Table Entry which indicates it is
> > pointing to a page which has been swapped out to block 0 of swap device
> > 256. This is probably caused by a problem in the kernel. You can certainly
> > run memtest et al. If you don't find anything, I would assume the problem
> > is in the kernel.
>
> Marcin, please run the memtest86 and report back.
Finally I was able to take the machine down and test its memory. Indeed
one of the modules was faulty (the very first run of memtest revealed
it).
thanks,
Marcin
--
Marcin Owsiany
porridge@expro.pl
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2005-04-17 19:53 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-07-27 0:21 "swap_free: Unused swap offset entry 00000100" but no crash? Marcin Owsiany
2004-07-27 12:30 ` Keith Owens
2004-07-27 12:53 ` Robin Holt
2004-07-27 18:58 ` Marcelo Tosatti
2004-07-27 20:03 ` Robin Holt
2005-04-17 19:53 ` Marcin Owsiany
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®