systemd-udevd crashed with SIGABRT in __epoll_wait_nocancel()

[Eliah Kagan]

I don't know if this should be considered a bug in systemd or apport.

systemd-udevd appears to crash with this bug on each boot. This is a fully updated Xubuntu Artful amd64 system (installed yesterday from the daily-live) with systemd 234-2ubuntu1, udev 234-2ubuntu1, apport 2.20.6-0ubuntu4, and kernel version 4.11.0-12-generic. The system swaps to a 948 MiB swap file (/swapfile) and runs inside a VMware Workstation 12.5.7 virtual machine (with open-vm-tools) on a Windows host. The host system has not approached memory exhaustion. I suspect these details may be relevant, because the host system sometimes experiences unexplained slowdowns when I/O is high. Note that I only suspect, and do not actually know for sure, that this bug may have been triggered by the host system's I/O problems.

I only know about this crash because Apport notified me. Starting the system seems to take longer than it should; specifically, once the non-graphical login prompt is shown, it sometimes takes up to a minute for LightDM to appear. But I don't know if that is related, and beyond that, I have not observed anything going wrong on the system. When I originally installed this Ubuntu (guest) system, Apport did not inform me of a crash. Since then, the host system engaged in more intensive I/O, and also I enabled artful-proposed, and now the crash occurs. I don't know if either of these changes are relevant.

This most resembles bug 1705870, which occurred in Artful with some slightly earlier package versions. Depending on the results of the symbolic retrace, I might mark this as a duplicate of that bug. It also resembles bug 1663995, bug 1506798, bug 1433320, and some of its (many) duplicates. That bug (i.e., bug 1433320) was considered to affect apport rather than systemd (see https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1433320/comments/10) and is considered fixed. (To a lesser extent, this bug resembles bug 1437896, which that bug mentions, and which is considered a bug in systemd.)

I have made /bin/sh a symlink to bash on this system instead of dash (because I am using this system to test http://www.linuxfromscratch.org/lfs/view/development/chapter02/mounting.html, which expects it). Aside from making startup scripts run a bit slower and increasing memory footprint slightly, I suspect this is irrelevant to this bug, but it's a significant (and *unusual*) change so I figured it is worth mentioning.

ProblemType: Crash
DistroRelease: Ubuntu 17.10
Package: udev 234-2ubuntu1
ProcVersionSignature: Ubuntu 4.11.0-12.17-generic 4.11.12
Uname: Linux 4.11.0-12-generic x86_64
ApportVersion: 2.20.6-0ubuntu4
Architecture: amd64
Date: Mon Jul 31 17:41:49 2017
ExecutablePath: /lib/systemd/systemd-udevd
InstallationDate: Installed on 2017-07-31 (0 days ago)
InstallationMedia: Xubuntu 17.10 "Artful Aardvark" - Alpha amd64 (20170731)
Lsusb:
 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
 Bus 002 Device 004: ID 0e0f:0008 VMware, Inc.
 Bus 002 Device 003: ID 0e0f:0002 VMware, Inc. Virtual USB Hub
 Bus 002 Device 002: ID 0e0f:0003 VMware, Inc. Virtual Mouse
 Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
MachineType: VMware, Inc. VMware Virtual Platform
ProcCmdline: /lib/systemd/systemd-udevd
ProcEnviron:
 LANG=en_US.UTF-8
 PATH=(custom, no user)
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.11.0-12-generic root=UUID=3ebdfb0d-efad-4049-955f-56e299d1acea ro quiet splash
Signal: 6
SourcePackage: systemd
StacktraceTop:
 __epoll_wait_nocancel () at ../sysdeps/unix/syscall-template.S:84
 ?? ()
 ?? ()
 ?? ()
 __libc_start_main (main=0x1f82c781c0, argc=1, argv=0x7ffddfff7de8, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7ffddfff7dd8) at ../csu/libc-start.c:291
Title: systemd-udevd crashed with SIGABRT in __epoll_wait_nocancel()
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups:

dmi.bios.date: 07/02/2015
dmi.bios.vendor: Phoenix Technologies LTD
dmi.bios.version: 6.00
dmi.board.name: 440BX Desktop Reference Platform
dmi.board.vendor: Intel Corporation
dmi.board.version: None
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 1
dmi.chassis.vendor: No Enclosure
dmi.chassis.version: N/A
dmi.modalias: dmi:bvnPhoenixTechnologiesLTD:bvr6.00:bd07/02/2015:svnVMware,Inc.:pnVMwareVirtualPlatform:pvrNone:rvnIntelCorporation:rn440BXDesktopReferencePlatform:rvrNone:cvnNoEnclosure:ct1:cvrN/A:
dmi.product.name: VMware Virtual Platform
dmi.product.version: None
dmi.sys.vendor: VMware, Inc.

[Apport retracing service (apport)]

Thank you for taking the time to report this crash and helping to make this software better. This particular crash has already been reported and is a duplicate of bug #1705870, so is being marked as such. Please look at the other bug report to see if there is any missing information that you can provide, or to see if there is a workaround for the bug. Additionally, any further discussion regarding the bug should occur in the other report. Please continue to report any other bugs you may find.