Bug#638984: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188]
reassign 638984 src:linux thanks On Sat, Dec 03, 2011 at 07:36:45PM +0100, Sandro Tosi wrote: On Sat, Dec 3, 2011 at 02:01, Jonathan Nieder jrnie...@gmail.com wrote: Hi, Sandro Tosi wrote: On Wed, Aug 24, 2011 at 04:32, Ben Hutchings b...@decadent.org.uk wrote: Unfortunately your log shows almost no useful information; the call traces have been lost. The only function that is mentioned in your log is find_get_pages(), which is not present in any of the call traces in the Red Hat bug report. Oh I see - of course it would have been too easy :) It may be that serial logging will be more reliable; could you enable a serial console and try to capture the error log on another system? We've started the process to enable the serial console logging on the involved machines Any luck? Sadly no :( we weren't able to configure a serial console but the frequency of the crashes has slightly reduced, but still present. I'll try to revamp the internal discussion and hopefully bring some result back to Debian. Does this still occur with current kernels, e.g. Wheezy or later? Cheers, Moritz -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130812150053.ga8...@inutil.org
Bug#638984: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188]
On Sat, Dec 3, 2011 at 02:01, Jonathan Nieder jrnie...@gmail.com wrote: Hi, Sandro Tosi wrote: On Wed, Aug 24, 2011 at 04:32, Ben Hutchings b...@decadent.org.uk wrote: Unfortunately your log shows almost no useful information; the call traces have been lost. The only function that is mentioned in your log is find_get_pages(), which is not present in any of the call traces in the Red Hat bug report. Oh I see - of course it would have been too easy :) It may be that serial logging will be more reliable; could you enable a serial console and try to capture the error log on another system? We've started the process to enable the serial console logging on the involved machines Any luck? Sadly no :( we weren't able to configure a serial console but the frequency of the crashes has slightly reduced, but still present. I'll try to revamp the internal discussion and hopefully bring some result back to Debian. -- Sandro Tosi (aka morph, morpheus, matrixhasu) My website: http://matrixhasu.altervista.org/ Me at Debian: http://wiki.debian.org/SandroTosi -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/CAPdtAj3u0u3hqGvtsjgYMH3ynCpSQ_bPFi=x+8pwzsazk0p...@mail.gmail.com
Bug#638984: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188]
Hi, Sandro Tosi wrote: On Wed, Aug 24, 2011 at 04:32, Ben Hutchings b...@decadent.org.uk wrote: Unfortunately your log shows almost no useful information; the call traces have been lost. The only function that is mentioned in your log is find_get_pages(), which is not present in any of the call traces in the Red Hat bug report. Oh I see - of course it would have been too easy :) It may be that serial logging will be more reliable; could you enable a serial console and try to capture the error log on another system? We've started the process to enable the serial console logging on the involved machines Any luck? Thanks, Jonathan -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20111203010131.ga8...@elie.hsd1.il.comcast.net
Bug#638984: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188]
On Thu, 2011-08-25 at 23:24 +0200, Sandro Tosi wrote: [...] It may be that serial logging will be more reliable; could you enable a serial console and try to capture the error log on another system? We've started the process to enable the serial console logging on the involved machines: do you have any tips that can speed up our operations team to enable it (in particular a way that could get us the messages we need)? See http://www.kernel.org/doc/Documentation/serial-console.txt. Ben. signature.asc Description: This is a digitally signed message part
Bug#638984: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188]
Hi Ben, thanks for the quick reply! On Wed, Aug 24, 2011 at 04:32, Ben Hutchings b...@decadent.org.uk wrote: On Tue, 2011-08-23 at 16:17 +0200, Sandro Tosi wrote: Package: linux-2.6 Version: 2.6.32-35 Severity: important Hello, we've had several server-class machine (G6 and G7 HP Proliant blades) affected by the kernel bug mentioned in the subject. On the console we can see several several instances of the message: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188] and then the machine is freezed and need a hard reset. I've cutpasted what's left from the kern.log (remotely logged on another machine) since I don't have access anymore to those files (plus a bit of anonymization about the network :)). It's an high priority for us to get it fixed so we can provide all the help you might need: just ask about it. I've googled a bit about the problem and found these references: http://kerneltrap.org/mailarchive/linux-kernel/2010/11/16/4646082/thread https://bugzilla.redhat.com/show_bug.cgi?id=649694 [...] The error message you are seeing is not at all specific, so we should not assume it has anything to do with the Red Hat bug report. Unfortunately your log shows almost no useful information; the call traces have been lost. The only function that is mentioned in your log is find_get_pages(), which is not present in any of the call traces in the Red Hat bug report. Oh I see - of course it would have been too easy :) It may be that serial logging will be more reliable; could you enable a serial console and try to capture the error log on another system? We've started the process to enable the serial console logging on the involved machines: do you have any tips that can speed up our operations team to enable it (in particular a way that could get us the messages we need)? Thank in advance, -- Sandro Tosi (aka morph, morpheus, matrixhasu) My website: http://matrixhasu.altervista.org/ Me at Debian: http://wiki.debian.org/SandroTosi -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/capdtaj0xjzw__q5q0o4ov+2owxu8t-wnpcpmd0j-q3jpag5...@mail.gmail.com
Bug#638984: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188]
On Tue, 2011-08-23 at 16:17 +0200, Sandro Tosi wrote: Package: linux-2.6 Version: 2.6.32-35 Severity: important Hello, we've had several server-class machine (G6 and G7 HP Proliant blades) affected by the kernel bug mentioned in the subject. On the console we can see several several instances of the message: BUG: soft lockup - CPU#1 stuck for 61s! [kswapd0:188] and then the machine is freezed and need a hard reset. I've cutpasted what's left from the kern.log (remotely logged on another machine) since I don't have access anymore to those files (plus a bit of anonymization about the network :)). It's an high priority for us to get it fixed so we can provide all the help you might need: just ask about it. I've googled a bit about the problem and found these references: http://kerneltrap.org/mailarchive/linux-kernel/2010/11/16/4646082/thread https://bugzilla.redhat.com/show_bug.cgi?id=649694 [...] The error message you are seeing is not at all specific, so we should not assume it has anything to do with the Red Hat bug report. Unfortunately your log shows almost no useful information; the call traces have been lost. The only function that is mentioned in your log is find_get_pages(), which is not present in any of the call traces in the Red Hat bug report. It may be that serial logging will be more reliable; could you enable a serial console and try to capture the error log on another system? Ben. signature.asc Description: This is a digitally signed message part