Bug#691427: journal commit I/O error on brand-new Thinkpad T430s ext4 on lvm on SSD
Just for the record, I believe this is a firmware bug in the Intel SSD 520 180GB disks used by Lenovo, causing the disks to lock up under heavy load. Perhaps this bug should be closed, or tagged as wontfix, or something else? -- Happy hacking Petter Reinholdtsen -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/2fl38a8jvwt@diskless.uio.no
Bug#691427: journal commit I/O error on brand-new Thinkpad T430s ext4 on lvm on SSD
I just discovered that this bug seem to be reported to the kernel develoers as URL: https://bugzilla.kernel.org/show_bug.cgi?id=51861 . According to that report, the problem went away on its own. While according to URL: http://forums.lenovo.com/t5/T400-T500-and-newer-T-series/T430s-Intel-SSD-520-180GB-issue/td-p/888083/page/2 , replacing the motherboard can help. Not quite sure what to believe. -- Happy hacking Petter Reinholdtsen -- 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/20130704081949.ge26...@diskless.uio.no
Bug#691427: journal commit I/O error on brand-new Thinkpad T430s ext4 on lvm on SSD
Hello Jonathan, Thanks for your interest in my issue. I've run the following script to catch the dmesg messages : #!/bin/sh while /bin/true; do dmesg ~/dmesg_logs/dmesg_$(date +%T).log; sync; sleep 1; done You'll find attached to this email the content of the last file written when the bug occurs. If I can do anything else to help you... Dorian On Fri, Oct 26, 2012 at 1:44 PM, Didier 'OdyX' Raboud o...@debian.orgwrote: Hi Jonathan, Le jeudi, 25 octobre 2012 22.25:03, Jonathan Nieder a écrit : Didier Raboud wrote: Tags: upstream What upstream version did you test? Well we only tried packaged kernels, all wheezy, unstable and experimental exhibit this behaviour. We started to compile an upstream kernel but didn't try yet. I tagged this bug as upstream, considering it's probably not a bug of the packaging. Can you get full dmesg output when in this state? A serial console[1] or netconsole[2] might be helpful in obtaining that, or if that's not possible, a photograph of messages on the screen will do in a pinch. I'll let Dorian send that to the bug, we'll see what that gives. Cheers, OdyX dmesg_10:32:10.log.bz2 Description: BZip2 compressed data
Bug#691427: journal commit I/O error on brand-new Thinkpad T430s ext4 on lvm on SSD
Hi Jonathan, Le jeudi, 25 octobre 2012 22.25:03, Jonathan Nieder a écrit : Didier Raboud wrote: Tags: upstream What upstream version did you test? Well we only tried packaged kernels, all wheezy, unstable and experimental exhibit this behaviour. We started to compile an upstream kernel but didn't try yet. I tagged this bug as upstream, considering it's probably not a bug of the packaging. Can you get full dmesg output when in this state? A serial console[1] or netconsole[2] might be helpful in obtaining that, or if that's not possible, a photograph of messages on the screen will do in a pinch. I'll let Dorian send that to the bug, we'll see what that gives. Cheers, OdyX -- 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/201210261344.35883.o...@debian.org
Bug#691427: journal commit I/O error on brand-new Thinkpad T430s ext4 on lvm on SSD
Source: linux Version: 3.2.32-1 Severity: important Tags: upstream Hi dear linux maintainers, I have setup a Debian Wheezy on a brand new Lenovo Thinkpad T430s with a SSDSC2BW180A3L SSD drive. After the install, the user installed his usual softwares (solr, apache, nodejs, …) and started work. Under some load (importing thousands of records in MySQL via a hand-called PHP script), the system reliably stops working fine enough because of journal commit I/O error leading to a readonly filesystem. Then Gnome becomes (obviously) unuseable and the only way out is a hard-reboot. The disk looks fine (and is brand new), according to the BIOS disk report and SMART tests, so the usual suspiscion that it might be broken is improbable. We have reproduced this bug under 3.2.0-4 (3.2.23-1), 3.2.0-5 (3.2.32-1) and 3.5-trunk (3.5.5-1~experimental.1). The setup is an ext4 / (and /home) on LVM, on the mentionned SSD. What are the things we could do to help resolve this bug? Are there logs that you might want to get? Please advise! Cheers, OdyX, on behalf of Dorian Villet, owner of the (unuseable) T430s. -- 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/20121025145608.8828.41282.reportbug@gyllingar
Bug#691427: journal commit I/O error on brand-new Thinkpad T430s ext4 on lvm on SSD
found 691427 linux/3.2.23-1 , linux/3.5.5-1~experimental.1 quit Hi, Didier Raboud wrote: Tags: upstream What upstream version did you test? [...] Under some load (importing thousands of records in MySQL via a hand-called PHP script), the system reliably stops working fine enough because of journal commit I/O error leading to a readonly filesystem. Then Gnome becomes (obviously) unuseable and the only way out is a hard-reboot. Can you get full dmesg output when in this state? A serial console[1] or netconsole[2] might be helpful in obtaining that, or if that's not possible, a photograph of messages on the screen will do in a pinch. Thanks for reporting, and hope that helps, Jonathan [1] if you are lucky enough to have a docking station with serial port: http://www.kernel.org/doc/Documentation/serial-console.txt [2] http://www.kernel.org/doc/Documentation/networking/netconsole.txt -- 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/20121025202503.GE30334@elie.Belkin
