Bug#928125: losetup causes Dead systemd-udevd processes, blocks forever

Michael Biebl biebl at debian.org
Sun Apr 28 18:41:47 BST 2019


Am 28.04.19 um 18:00 schrieb Brad Barnett:
> 
> Package: udev
> Version: 232-25+deb9u11
> 
> After today's upgrade to debian 9.9, I attempted to create a disk image.
> 
> I have a script, it does the following:
> 
> ###
> dd if=/dev/zero of=/tmp/ff1.raw bs=1G seek=8 count=0
> sync
> sleep 1
> parted /tmp/ff1.raw mklabel msdos
> parted -s /tmp/ff1.raw mkpart primary linux-swap 1 100
> parted -s -- /tmp/ff1.raw mkpart primary ext2 101 -1
> parted -s -- /tmp/ff1.raw set 2 boot on
> sleep 5
> losetup -Pf /tmp/ff1.raw --show
> ###
> 
> The above has been run several times per month, for years.
> 
> After today's upgrade I rebooted, ran the above, and it locked on losetup.
> 
> I see this:
> 
> # ps auxwwf
> 
> root       346  0.0  0.1  45656  3420 ?        Ss   11:30   0:00 /lib/systemd/systemd-udevd --daemon
> root      2596  0.0  0.0  45564  2120 ?        S    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2597  0.0  0.0  45564  2184 ?        D    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2599  0.0  0.1  45644  3096 ?        S    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2602  0.0  0.0  45564  2120 ?        D    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2603  0.0  0.0  45656  1936 ?        D    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2607  0.0  0.0  45656  1936 ?        D    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2609  0.0  0.0  45564  2120 ?        D    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2610  0.0  0.0  45656  1936 ?        D    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> root      2612  0.0  0.0  45656  1936 ?        D    11:31   0:00  \_ /lib/systemd/systemd-udevd --daemon
> 
> And
> 
> root      2591  0.0  0.0   8064   780 pts/2    D    11:31   0:00 /sbin/losetup -Pf /tmp/ff1.raw --show
> 
> I also see this (last 3 'blocks' pasted):
> 
> Apr 28 11:34:17 buildvm kernel: [  242.831744] INFO: task systemd-udevd:2612 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.831807]       Not tainted 4.9.0-9-amd64 #1 Debian 4.9.168-1
> Apr 28 11:34:17 buildvm kernel: [  242.831859] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> Apr 28 11:34:17 buildvm kernel: [  242.831927] systemd-udevd   D    0  2612    346 0x00000004
> Apr 28 11:34:17 buildvm kernel: [  242.831932]  ffff8a2dbb10c400 0000000000000000 ffff8a2dbacc5240 ffff8a2dbfa18980
> Apr 28 11:34:17 buildvm kernel: [  242.831939]  ffffffffaf411500 ffffaf84409e7b58 ffffffffaee15a99 1e580f2a959617cf
> Apr 28 11:34:17 buildvm kernel: [  242.831946]  00000000000402c0
> ffff8a2dbfa18980 0000000000000000 ffff8a2dbacc5240
> 
> Apr 28 11:34:17 buildvm kernel: [  242.831952] Call Trace:
> Apr 28 11:34:17 buildvm kernel: [  242.831960]  [<ffffffffaee15a99>] ? __schedule+0x239/0x6f0
> Apr 28 11:34:17 buildvm kernel: [  242.831967]  [<ffffffffaee15f82>] ? schedule+0x32/0x80
> Apr 28 11:34:17 buildvm kernel: [  242.831973]  [<ffffffffaee1623a>] ? schedule_preempt_disabled+0xa/0x10
> Apr 28 11:34:17 buildvm kernel: [  242.831977]  [<ffffffffaee17c94>] ? __mutex_lock_slowpath+0xb4/0x130
> Apr 28 11:34:17 buildvm kernel: [  242.831982]  [<ffffffffaee17d2b>] ? mutex_lock+0x1b/0x30
> Apr 28 11:34:17 buildvm kernel: [  242.831988]  [<ffffffffc03e11c5>] ? lo_open+0x15/0x50 [loop]
> Apr 28 11:34:17 buildvm kernel: [  242.831994]  [<ffffffffaea48ce3>] ? __blkdev_get+0xd3/0x470
> Apr 28 11:34:17 buildvm kernel: [  242.832001]  [<ffffffffaea492e6>] ? blkdev_get+0x126/0x330
> Apr 28 11:34:17 buildvm kernel: [  242.832008]  [<ffffffffaea27e04>] ? unlock_new_inode+0x44/0x70
> Apr 28 11:34:17 buildvm kernel: [  242.832014]  [<ffffffffaea47e3a>] ? bdget+0xfa/0x110
> Apr 28 11:34:17 buildvm kernel: [  242.832020]  [<ffffffffaea49530>] ? blkdev_get_by_dev+0x40/0x40
> Apr 28 11:34:17 buildvm kernel: [  242.832026]  [<ffffffffaea08e74>] ? do_dentry_open+0x234/0x340
> Apr 28 11:34:17 buildvm kernel: [  242.832030]  [<ffffffffaea1ada7>] ? path_openat+0x777/0x15b0
> Apr 28 11:34:17 buildvm kernel: [  242.832037]  [<ffffffffae9c8211>] ? page_add_file_rmap+0x11/0x110
> Apr 28 11:34:17 buildvm kernel: [  242.832042]  [<ffffffffaea1cea1>] ? do_filp_open+0x91/0x100
> Apr 28 11:34:17 buildvm kernel: [  242.832047]  [<ffffffffae9bb263>] ? handle_mm_fault+0xcc3/0x1350
> Apr 28 11:34:17 buildvm kernel: [  242.832052]  [<ffffffffaea07b3a>] ? __check_object_size+0xfa/0x1d8
> Apr 28 11:34:17 buildvm kernel: [  242.832058]  [<ffffffffaea0a3ae>] ? do_sys_open+0x12e/0x210
> Apr 28 11:34:17 buildvm kernel: [  242.832064]  [<ffffffffae803b7d>] ? do_syscall_64+0x8d/0xf0
> Apr 28 11:34:17 buildvm kernel: [  242.832069]  [<ffffffffaee1a84e>] ? entry_SYSCALL_64_after_swapgs+0x58/0xc6
> Apr 28 11:36:18 buildvm kernel: [  363.661408] INFO: task losetup:2591 blocked for more than 120 seconds.
> Apr 28 11:36:18 buildvm kernel: [  363.661487]       Not tainted 4.9.0-9-amd64 #1 Debian 4.9.168-1
> Apr 28 11:36:18 buildvm kernel: [  363.661541] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> Apr 28 11:36:18 buildvm kernel: [  363.661610] losetup         D    0  2591      1 0x00000004
> Apr 28 11:36:18 buildvm kernel: [  363.661620]  ffff8a2db7926c00 0000000000000000 ffff8a2db7956ec0 ffff8a2dbfa18980
> Apr 28 11:36:18 buildvm kernel: [  363.661628]  ffffffffaf411500 ffffaf8440a7fbd8 ffffffffaee15a99 ffffffffaee1a964
> Apr 28 11:36:18 buildvm kernel: [  363.661635]  00ff8a2db7957a00 ffff8a2dbfa18980 ffff8a2dbacc5240 ffff8a2db7956ec0
> Apr 28 11:36:18 buildvm kernel: [  363.661642] Call Trace:
> Apr 28 11:36:18 buildvm kernel: [  363.661656]  [<ffffffffaee15a99>] ? __schedule+0x239/0x6f0
> Apr 28 11:36:18 buildvm kernel: [  363.661663]  [<ffffffffaee1a964>] ? __switch_to_asm+0x34/0x70
> Apr 28 11:36:18 buildvm kernel: [  363.661670]  [<ffffffffaee15f82>] ? schedule+0x32/0x80
> Apr 28 11:36:18 buildvm kernel: [  363.661677]  [<ffffffffaee1623a>] ? schedule_preempt_disabled+0xa/0x10
> Apr 28 11:36:18 buildvm kernel: [  363.661682]  [<ffffffffaee17c94>] ? __mutex_lock_slowpath+0xb4/0x130
> Apr 28 11:36:18 buildvm kernel: [  363.661689]  [<ffffffffaee17d2b>] ? mutex_lock+0x1b/0x30
> Apr 28 11:36:18 buildvm kernel: [  363.661697]  [<ffffffffaeb15e46>] ? blkdev_reread_part+0x16/0x30
> Apr 28 11:36:18 buildvm kernel: [  363.661705]  [<ffffffffc03e1803>] ? loop_reread_partitions+0x23/0x50 [loop]
> Apr 28 11:36:18 buildvm kernel: [  363.661712]  [<ffffffffc03e1b3b>] ? loop_set_status+0x30b/0x3a0 [loop]
> Apr 28 11:36:18 buildvm kernel: [  363.661719]  [<ffffffffc03e1e88>] ? loop_set_status64+0x48/0x70 [loop]
> Apr 28 11:36:18 buildvm kernel: [  363.661726]  [<ffffffffc03e3138>] ? lo_ioctl+0xe8/0x6f0 [loop]
> Apr 28 11:36:18 buildvm kernel: [  363.661734]  [<ffffffffaeb16645>] ? blkdev_ioctl+0x265/0x970
> Apr 28 11:36:18 buildvm kernel: [  363.661740]  [<ffffffffae9bb37f>] ? handle_mm_fault+0xddf/0x1350
> Apr 28 11:36:18 buildvm kernel: [  363.661748]  [<ffffffffaea481a9>] ? block_ioctl+0x39/0x40
> Apr 28 11:36:18 buildvm kernel: [  363.661753]  [<ffffffffaea201b2>] ? do_vfs_ioctl+0xa2/0x620
> Apr 28 11:36:18 buildvm kernel: [  363.661759]  [<ffffffffaea207a4>] ? SyS_ioctl+0x74/0x80
> Apr 28 11:36:18 buildvm kernel: [  363.661765]  [<ffffffffae803b7d>] ? do_syscall_64+0x8d/0xf0
> Apr 28 11:36:18 buildvm kernel: [  363.661771]  [<ffffffffaee1a84e>] ? entry_SYSCALL_64_after_swapgs+0x58/0xc6
> Apr 28 11:36:18 buildvm kernel: [  363.661776] INFO: task systemd-udevd:2597 blocked for more than 120 seconds.
> Apr 28 11:36:18 buildvm kernel: [  363.661841]       Not tainted 4.9.0-9-amd64 #1 Debian 4.9.168-1
> Apr 28 11:36:18 buildvm kernel: [  363.661894] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
> Apr 28 11:36:18 buildvm kernel: [  363.661962] systemd-udevd   D    0  2597    346 0x00000004
> Apr 28 11:36:18 buildvm kernel: [  363.661969]  ffff8a2db7924400 0000000000000000 ffff8a2db79ff180 ffff8a2dbfb18980
> Apr 28 11:36:18 buildvm kernel: [  363.661976]  ffff8a2dbb91cec0 ffffaf844086bd88 ffffffffaee15a99 3959c0b6fb0308ab
> Apr 28 11:36:18 buildvm kernel: [  363.661983]  00ffaf844086bd70 ffff8a2dbfb18980 ffff8a2dbb219498 ffff8a2db79ff180
> Apr 28 11:36:18 buildvm kernel: [  363.661990] Call Trace:
> Apr 28 11:36:18 buildvm kernel: [  363.661998]  [<ffffffffaee15a99>] ? __schedule+0x239/0x6f0
> Apr 28 11:36:18 buildvm kernel: [  363.662006]  [<ffffffffaee15f82>] ? schedule+0x32/0x80
> Apr 28 11:36:18 buildvm kernel: [  363.662013]  [<ffffffffaee1623a>] ? schedule_preempt_disabled+0xa/0x10
> Apr 28 11:36:18 buildvm kernel: [  363.662017]  [<ffffffffaee17c94>] ? __mutex_lock_slowpath+0xb4/0x130
> Apr 28 11:36:18 buildvm kernel: [  363.662022]  [<ffffffffaee17d2b>] ? mutex_lock+0x1b/0x30
> Apr 28 11:36:18 buildvm kernel: [  363.662029]  [<ffffffffc03e2ff4>] ? lo_release+0x44/0xa0 [loop]
> Apr 28 11:36:18 buildvm kernel: [  363.662036]  [<ffffffffaea48b2d>] ? __blkdev_put+0x1ed/0x2d0
> Apr 28 11:36:18 buildvm kernel: [  363.662043]  [<ffffffffaea491b1>] ? blkdev_close+0x21/0x30
> Apr 28 11:36:18 buildvm kernel: [  363.662048]  [<ffffffffaea0dc48>] ? __fput+0xd8/0x220
> Apr 28 11:36:18 buildvm kernel: [  363.662055]  [<ffffffffae898d5f>] ? task_work_run+0x7f/0xa0
> Apr 28 11:36:18 buildvm kernel: [  363.662061]  [<ffffffffae803754>] ? exit_to_usermode_loop+0xa4/0xb0
> Apr 28 11:36:18 buildvm kernel: [  363.662067]  [<ffffffffae803bcd>] ? do_syscall_64+0xdd/0xf0
> Apr 28 11:36:18 buildvm kernel: [  363.662073]  [<ffffffffaee1a84e>] ? entry_SYSCALL_64_after_swapgs+0x58/0xc6
> 
> The above repeats for all D systemd-udev tasks, and losetup, EG:
> 
> Apr 28 11:34:17 buildvm kernel: [  242.829350] INFO: task losetup:2591 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.829715] INFO: task systemd-udevd:2597 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.830018] INFO: task systemd-udevd:2602 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.830342] INFO: task systemd-udevd:2603 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.830679] INFO: task systemd-udevd:2607 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.831086] INFO: task systemd-udevd:2609 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.831413] INFO: task systemd-udevd:2610 blocked for more than 120 seconds.
> Apr 28 11:34:17 buildvm kernel: [  242.831744] INFO: task systemd-udevd:2612 blocked for more than 120 seconds.
> Apr 28 11:36:18 buildvm kernel: [  363.661408] INFO: task losetup:2591 blocked for more than 120 seconds.
> Apr 28 11:36:18 buildvm kernel: [  363.661776] INFO: task systemd-udevd:2597 blocked for more than 120 seconds.
> 
> I tried to strace this, but if I do?  It works fine.  Further, once I've
> done an strace, further losetups work fine -- without blocking.
> 
> Not really sure what's causing this, but without strace and losetup D
> forever, it's annoying to debug.  Suggestions welcome.
> 
> Note that after the above INFO/hangs, the above systemd-udev and losetup
> tasks sit D forever.  It's been 20+ minutes, no activity in kern.log,
> just dead.
> 

This looks like a kernel regression.
Do you use a Debian kernel? From which version did you upgrade?
If you downgrade to the previous kernel version, does the problem go away.


-- 
Why is it that all of the instruments seeking intelligent life in the
universe are pointed away from Earth?

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://alioth-lists.debian.net/pipermail/pkg-systemd-maintainers/attachments/20190428/5326baf4/attachment-0001.sig>


More information about the Pkg-systemd-maintainers mailing list