Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
Here one example for SLE_12 x86_64:
[ 205.213153] BUG: unable to handle kernel paging request at ffffffffa01fae60 [ 205.214124] IP: [<ffffffffa01fae60>] 0xffffffffa01fae60 [ 205.214914] PGD 1a0d067 PUD 1a0e063 PMD 1f99fd067 PTE 0 [ 205.215738] Oops: 0010 [#1] SMP [ 205.216239] Modules linked in: sd_mod raid1 raid0 md_mod cramfs ext4 crc16 mbcache jbd2 sr_mod cdrom ata_generic virtio_blk ata_piix ahci libahci libata scsi_mod virtio_pci virtio_ring virtio binfmt_misc loop autofs4 [last unloaded: scsi_debug] [ 205.217089] Supported: Yes [ 205.217089] CPU: 2 PID: 24652 Comm: systemd-udevd Not tainted 3.12.28-4-default #1 [ 205.217089] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5.1-0-g8936dbb-20150318_003745-cloud117 04/01/2014 [ 205.217089] task: ffff8800bad38040 ti: ffff8800bad32000 task.ti: ffff8800bad32000 [ 205.217089] RIP: 0010:[<ffffffffa01fae60>] [<ffffffffa01fae60>] 0xffffffffa01fae60 [ 205.224409] RSP: 0018:ffff8800bad33d50 EFLAGS: 00010286 [ 205.225280] RAX: ffffffffa01fae60 RBX: ffff8801f9b39028 RCX: 0000000000000286 [ 205.226668] RDX: ffff8801f9b39260 RSI: 0000000000000286 RDI: ffff8801f9b39018 [ 205.227561] RBP: ffff8801f9b39018 R08: ffff8800bad32000 R09: 0000000000000000 [ 205.227561] R10: ffffffff81f7b558 R11: 0000000000000000 R12: ffff8801a8f56180 [ 205.227561] R13: ffff8801f9439bd8 R14: 000000000000005d R15: 0000000000000000 [ 205.227561] FS: 00007f434c12c880(0000) GS:ffff880204f00000(0000) knlGS:0000000000000000 [ 205.227561] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 205.227561] CR2: ffffffffa01fae60 CR3: 00000000bad72000 CR4: 00000000000407e0 [ 205.232887] Stack: [ 205.232887] ffffffff81393a7d ffff8801f9b39028 ffffffff81ae4d40 ffff8801f8c63a80 [ 205.232887] ffffffff8129a352 ffff8800bb46c800 ffff8801f9b39018 ffff8801a8f56300 [ 205.232887] ffffffffa003be21 ffff8800bb46c9a0 ffff8800bb46c990 ffff8801a8f56300 [ 205.232887] Call Trace: [ 205.232887] Inexact backtrace: [ 205.232887] [ 205.232887] [<ffffffff81393a7d>] ? device_release+0x2d/0xa0 [ 205.232887] [<ffffffff8129a352>] ? kobject_cleanup+0x72/0x1a0 [ 205.232887] [<ffffffffa003be21>] ? scsi_host_dev_release+0xc1/0xe0 [scsi_mod] [ 205.232887] [<ffffffff81393a7d>] ? device_release+0x2d/0xa0 [ 205.232887] [<ffffffff8129a352>] ? kobject_cleanup+0x72/0x1a0 [ 205.232887] [<ffffffff81393a7d>] ? device_release+0x2d/0xa0 [ 205.232887] [<ffffffff8129a352>] ? kobject_cleanup+0x72/0x1a0 [ 205.232887] [<ffffffff8106db07>] ? execute_in_process_context+0x67/0x70 [ 205.232887] [<ffffffff81393a7d>] ? device_release+0x2d/0xa0 [ 205.232887] [<ffffffff8129a352>] ? kobject_cleanup+0x72/0x1a0 [ 205.232887] [<ffffffffa021077b>] ? scsi_disk_put+0x2b/0x40 [sd_mod] [ 205.232887] [<ffffffff811d55bd>] ? __blkdev_put+0x15d/0x1a0 [ 205.232887] [<ffffffff811d6011>] ? blkdev_close+0x21/0x30 [ 205.248034] [<ffffffff811a0382>] ? __fput+0xc2/0x240 [ 205.248034] [<ffffffff81073947>] ? task_work_run+0x97/0xd0 [ 205.248034] [<ffffffff81002929>] ? do_notify_resume+0x69/0xa0 [ 205.248034] [<ffffffff81519603>] ? int_signal+0x12/0x17 [ 205.248034] Code: Bad RIP value. [ 205.248034] RIP [<ffffffffa01fae60>] 0xffffffffa01fae60 [ 205.248034] RSP <ffff8800bad33d50> [ 205.248034] CR2: ffffffffa01fae60 [ 205.248034] ---[ end trace ec1ae60827e0b798 ]--- [ 205.254616] systemd-udevd[1223]: worker [24652] terminated by signal 9 (Killed) [ 205.255711] systemd-udevd[1223]: worker [24652] failed while handling '/devices/pseudo_0/adapter0/host10/target10:0:0/10:0:0:0/block/sda/sda1' [ 205.454695] scsi_debug: host protection [ 205.455316] scsi11 : scsi_debug, version 1.82 [20100324], dev_size_mb=50, opts=0x0 [ 205.456558] scsi 11:0:0:0: Direct-Access Linux scsi_debug 0004 PQ: 0 ANSI: 5 [ 205.460037] sd 11:0:0:0: [sda] 102400 512-byte logical blocks: (52.4 MB/50.0 MiB) [ 205.464032] sd 11:0:0:0: [sda] Write Protect is off [ 205.464732] sd 11:0:0:0: [sda] Mode Sense: 73 00 10 08 [ 205.472036] sd 11:0:0:0: [sda] Write cache: enabled, read cache: enabled, supports DPO and FUA qemu: terminating on signal 15 from pid 25397
Killed Job ### WATCHDOG MARKER END ### No buildstatus set, either the base system is broken (kernel/initrd/udev/glibc/bash/perl) or the build host has a kernel or hardware problem...
On Thursday 07 May 2015, Ruediger Meier wrote:
Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
cu, Rudi
On Montag, 1. Februar 2016, 09:43:53 CET wrote Ruediger Meier:
On Thursday 07 May 2015, Ruediger Meier wrote:
Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
SP1 is not yet imported at all. Sorry, had not time for it yet.
So this is either a generic SP0 issue or a build host issue.
Do you have any pointer to a logfile?
On Monday 01 February 2016, Adrian Schröter wrote:
On Montag, 1. Februar 2016, 09:43:53 CET wrote Ruediger Meier:
On Thursday 07 May 2015, Ruediger Meier wrote:
Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
SP1 is not yet imported at all. Sorry, had not time for it yet.
So this is either a generic SP0 issue or a build host issue.
Do you have any pointer to a logfile?
It's in home:rudi_m:ul-all/ul-plain
[...] SLE_12_SP1 ppc64le succeeded SLE_12_SP1 x86_64 failed SLE_12 ppc64le failed SLE_12 x86_64 failed [...]
This package is util-linux inclusive root checks (sudo). It's a lot stress for OBS build hosts but it discovered a few kernel bugs already. (Like currently Leap and Factory also fail because of another kernel bug ... already fixed.)
The oops does not happen always but AFAIR only on SLE 12. scsi_debug module is involved, the oops can only happen if the log does not show lines containing "missing scsi_debug module".
cu, Rudi
On Montag, 1. Februar 2016, 09:49:59 CET wrote Adrian Schröter:
On Montag, 1. Februar 2016, 09:43:53 CET wrote Ruediger Meier:
On Thursday 07 May 2015, Ruediger Meier wrote:
Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
SP1 is not yet imported at all. Sorry, had not time for it yet.
So this is either a generic SP0 issue or a build host issue.
SP1 is imported since monday.
And SP0 re-imported to solve some missing packages.
On Wednesday 03 February 2016, Adrian Schröter wrote:
On Montag, 1. Februar 2016, 09:49:59 CET wrote Adrian Schröter:
On Montag, 1. Februar 2016, 09:43:53 CET wrote Ruediger Meier:
On Thursday 07 May 2015, Ruediger Meier wrote:
Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
SP1 is not yet imported at all. Sorry, had not time for it yet.
So this is either a generic SP0 issue or a build host issue.
SP1 is imported since monday.
One build is currently hanging again osc remotebuildlog home:rudi_m:ul-all/ul-plain SLE_12_SP1 ppc64le
And SP0 re-imported to solve some missing packages.
cu, Rudi
On Mittwoch, 3. Februar 2016, 16:46:42 CET wrote Ruediger Meier:
On Wednesday 03 February 2016, Adrian Schröter wrote:
On Montag, 1. Februar 2016, 09:49:59 CET wrote Adrian Schröter:
On Montag, 1. Februar 2016, 09:43:53 CET wrote Ruediger Meier:
On Thursday 07 May 2015, Ruediger Meier wrote:
Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
SP1 is not yet imported at all. Sorry, had not time for it yet.
So this is either a generic SP0 issue or a build host issue.
SP1 is imported since monday.
One build is currently hanging again osc remotebuildlog home:rudi_m:ul-all/ul-plain SLE_12_SP1 ppc64le
I have created a bugreport for this against the kernel people:
https://bugzilla.suse.com/show_bug.cgi?id=964976
On Wednesday 03 February 2016, Adrian Schröter wrote:
On Mittwoch, 3. Februar 2016, 16:46:42 CET wrote Ruediger Meier:
On Wednesday 03 February 2016, Adrian Schröter wrote:
On Montag, 1. Februar 2016, 09:49:59 CET wrote Adrian Schröter:
On Montag, 1. Februar 2016, 09:43:53 CET wrote Ruediger Meier:
On Thursday 07 May 2015, Ruediger Meier wrote:
Hi,
I get kernel oopses very often for SLE_12 builds on x86_64 and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
SP1 is not yet imported at all. Sorry, had not time for it yet.
So this is either a generic SP0 issue or a build host issue.
SP1 is imported since monday.
One build is currently hanging again osc remotebuildlog home:rudi_m:ul-all/ul-plain SLE_12_SP1 ppc64le
I have created a bugreport for this against the kernel people:
Thanks! I'am on CC there but unfortunately I have no access to the bug report.
Last year I have collected a few hundred SLE_12 build logs. 17% of them have such oops, see, http://akne.unxz.net/~rudi/tmp/obs-log-SLE12.tar.xz $ grep -l "Oops" log-*/bl_*SLE_12* | wc -l 112 $ grep -L "Oops" log-*/bl_*SLE_12* | wc -l 538
Would be nice if one could reproduce that on a real machine (non-OBS). I don't own SLE.
cu, Rudi
On Mittwoch, 3. Februar 2016, 18:54:07 CET wrote Ruediger Meier:
On Wednesday 03 February 2016, Adrian Schröter wrote:
On Mittwoch, 3. Februar 2016, 16:46:42 CET wrote Ruediger Meier:
On Wednesday 03 February 2016, Adrian Schröter wrote:
On Montag, 1. Februar 2016, 09:49:59 CET wrote Adrian Schröter:
On Montag, 1. Februar 2016, 09:43:53 CET wrote Ruediger Meier:
On Thursday 07 May 2015, Ruediger Meier wrote: > Hi, > > I get kernel oopses very often for SLE_12 builds on x86_64 > and ppc64le. Is this a known problem?
This kernel oops still happens. Where can I bug report this for SLE_12_SP1?
SP1 is not yet imported at all. Sorry, had not time for it yet.
So this is either a generic SP0 issue or a build host issue.
SP1 is imported since monday.
One build is currently hanging again osc remotebuildlog home:rudi_m:ul-all/ul-plain SLE_12_SP1 ppc64le
I have created a bugreport for this against the kernel people:
Thanks! I'am on CC there but unfortunately I have no access to the bug report.
Last year I have collected a few hundred SLE_12 build logs. 17% of them have such oops, see, http://akne.unxz.net/~rudi/tmp/obs-log-SLE12.tar.xz $ grep -l "Oops" log-*/bl_*SLE_12* | wc -l 112 $ grep -L "Oops" log-*/bl_*SLE_12* | wc -l 538
Would be nice if one could reproduce that on a real machine (non-OBS). I don't own SLE.
Using
osc build --vm-type=kvm
will setup the same environment on your system, if you use openSUSE_13.2, you would even have the same host (but I doubt it does matter here).
Adrian Schröter adrian@suse.de writes:
And SP0 re-imported to solve some missing packages.
There's still one problem left:
nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel
Andreas.
On Donnerstag, 4. Februar 2016, 14:38:55 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
And SP0 re-imported to solve some missing packages.
There's still one problem left:
nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel
you spoke about SP1?
Should be solved now.
Adrian Schröter adrian@suse.de writes:
On Donnerstag, 4. Februar 2016, 14:38:55 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
And SP0 re-imported to solve some missing packages.
There's still one problem left:
nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel
you spoke about SP1?
No, SP0.
https://build.opensuse.org/project/monitor/home:Andreas_Schwab:aranym
Andreas.
On Montag, 29. Februar 2016, 10:26:16 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
On Donnerstag, 4. Februar 2016, 14:38:55 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
And SP0 re-imported to solve some missing packages.
There's still one problem left:
nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel
you spoke about SP1?
No, SP0.
https://build.opensuse.org/project/monitor/home:Andreas_Schwab:aranym
okay, solved there as well
Adrian Schröter adrian@suse.de writes:
On Montag, 29. Februar 2016, 10:26:16 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
On Donnerstag, 4. Februar 2016, 14:38:55 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
And SP0 re-imported to solve some missing packages.
There's still one problem left:
nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel
you spoke about SP1?
No, SP0.
https://build.opensuse.org/project/monitor/home:Andreas_Schwab:aranym
okay, solved there as well
I still see the same unresolvable dependency.
Andreas.
On Dienstag, 1. März 2016, 10:56:42 wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
On Montag, 29. Februar 2016, 10:26:16 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
On Donnerstag, 4. Februar 2016, 14:38:55 CET wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
And SP0 re-imported to solve some missing packages.
There's still one problem left:
nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel
you spoke about SP1?
No, SP0.
https://build.opensuse.org/project/monitor/home:Andreas_Schwab:aranym
okay, solved there as well
I still see the same unresolvable dependency.
# eosc r home:Andreas_Schwab:aranym 13.2 x86_64 succeeded 12-ga x86_64 succeeded 11-sp3 x86_64 succeeded
You may run into a webui caching issue? Have you clicked on the reload button?
Adrian Schröter adrian@suse.de writes:
# eosc r home:Andreas_Schwab:aranym 13.2 x86_64 succeeded 12-ga x86_64 succeeded 11-sp3 x86_64 succeeded
$ osc r -v home:Andreas_Schwab:aranym SDL_image 13.2 x86_64 disabled 12-ga x86_64 unresolvable: nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel 11-sp3 x86_64 disabled
Andreas.
On Dienstag, 1. März 2016, 11:14:20 wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
# eosc r home:Andreas_Schwab:aranym 13.2 x86_64 succeeded 12-ga x86_64 succeeded 11-sp3 x86_64 succeeded
$ osc r -v home:Andreas_Schwab:aranym SDL_image 13.2 x86_64 disabled 12-ga x86_64 unresolvable: nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel 11-sp3 x86_64 disabled
that is just an old state,
osc buildinfo home:Andreas_Schwab:aranym SDL_image 12-ga x86_64
works (we have currently a larger backlog on backend home server)
Adrian Schröter adrian@suse.de writes:
On Dienstag, 1. März 2016, 11:14:20 wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
# eosc r home:Andreas_Schwab:aranym 13.2 x86_64 succeeded 12-ga x86_64 succeeded 11-sp3 x86_64 succeeded
$ osc r -v home:Andreas_Schwab:aranym SDL_image 13.2 x86_64 disabled 12-ga x86_64 unresolvable: nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel 11-sp3 x86_64 disabled
that is just an old state,
It doesn't say that.
Andreas.
On Dienstag, 1. März 2016, 13:55:04 wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
On Dienstag, 1. März 2016, 11:14:20 wrote Andreas Schwab:
Adrian Schröter adrian@suse.de writes:
# eosc r home:Andreas_Schwab:aranym 13.2 x86_64 succeeded 12-ga x86_64 succeeded 11-sp3 x86_64 succeeded
$ osc r -v home:Andreas_Schwab:aranym SDL_image 13.2 x86_64 disabled 12-ga x86_64 unresolvable: nothing provides libwebpdecoder1 = 0.4.3 needed by libwebp-devel 11-sp3 x86_64 disabled
that is just an old state,
It doesn't say that.
a manual backend imported file is not createing dirty flags in all repos building against it...
buildservice@lists.opensuse.org
-
Adrian Schröter -
Andreas Schwab -
Ruediger Meier