Description of problem: I've tried to reproduce live cd not created issue and after starting ~ 130th livecd Boxes crashed. I've used tiny core 5.3 liveCD (9MB in size) Version-Release number of selected component: gnome-boxes-3.16.2-1.fc22 Additional info: reporter: libreport-2.5.1 backtrace_rating: 4 cmdline: /usr/bin/gnome-boxes --gapplication-service crash_function: vasprintf executable: /usr/bin/gnome-boxes global_pid: 2572 kernel: 4.0.4-301.fc22.x86_64 runlevel: N 5 type: CCpp uid: 1000 Truncated backtrace: Thread no. 0 (10 frames) #3 vasprintf at /usr/include/bits/stdio2.h:210 #4 virVasprintfInternal at util/virstring.c:476 #5 virRaiseErrorFull at util/virerror.c:723 #6 virReportSystemErrorFull at util/virerror.c:1469 #7 virNetClientProgramCall at rpc/virnetclientprogram.c:371 #8 callFull at remote/remote_driver.c:6522 #9 remoteDomainOpenGraphicsFD at remote/remote_driver.c:6362 #10 virDomainOpenGraphicsFD at libvirt-domain.c:10202 #11 gvir_domain_open_graphics_fd at libvirt-gobject-domain.c:1261 #12 ____lambda160_ at /home/zeenix/checkout/gnome/gnome-boxes/src/libvirt-machine.vala:69
Created attachment 1057263 [details] File: backtrace
Created attachment 1057264 [details] File: cgroup
Created attachment 1057265 [details] File: core_backtrace
Created attachment 1057266 [details] File: dso_list
Created attachment 1057267 [details] File: environ
Created attachment 1057268 [details] File: limits
Created attachment 1057269 [details] File: maps
Created attachment 1057270 [details] File: mountinfo
Created attachment 1057271 [details] File: namespaces
Created attachment 1057272 [details] File: open_fds
Created attachment 1057273 [details] File: proc_pid_status
Created attachment 1057274 [details] File: var_log_messages
Seems deep in libvirt territory.
Looking at the stack trace I can't see why it would be SEGV'ing, but I do see an interesting error message in the stack "Cannot duplicate FD 1011: Too many open files" It looks like Boxes has exhausted the (stupidly low) default file descriptor limit for its user environment. I wonder if there is a file descriptor leak somewhere in either Boxes or libvirt, or if the particular usage scenario genuinely needed so many FDs ? I guess the FD exhaustion might have tickled some, never used, code path which results in memory corruption causing the eventual SEGV.
Looking at the open_fds attachment there is a good mix of pipe and socket FDs open, but I can't tell what they're for.
Wondering if this could be related to https://bugzilla.gnome.org/show_bug.cgi?id=746800
(In reply to Daniel Berrange from comment #14) > Looking at the stack trace I can't see why it would be SEGV'ing, but I do > see an interesting error message in the stack > > "Cannot duplicate FD 1011: Too many open files" > > It looks like Boxes has exhausted the (stupidly low) default file descriptor > limit for its user environment. I wonder if there is a file descriptor leak > somewhere in either Boxes or libvirt, or if the particular usage scenario > genuinely needed so many FDs ? I guess the FD exhaustion might have tickled > some, never used, code path which results in memory corruption causing the > eventual SEGV. Ah, this is then likely a side-effect of https://bugzilla.gnome.org/show_bug.cgi?id=748646 , which I really should solve sooner than later.
(In reply to Zeeshan Ali from comment #17) > (In reply to Daniel Berrange from comment #14) > > Looking at the stack trace I can't see why it would be SEGV'ing, but I do > > see an interesting error message in the stack > > > > "Cannot duplicate FD 1011: Too many open files" > > > > It looks like Boxes has exhausted the (stupidly low) default file descriptor > > limit for its user environment. I wonder if there is a file descriptor leak > > somewhere in either Boxes or libvirt, or if the particular usage scenario > > genuinely needed so many FDs ? I guess the FD exhaustion might have tickled > > some, never used, code path which results in memory corruption causing the > > eventual SEGV. > > Ah, this is then likely a side-effect of > https://bugzilla.gnome.org/show_bug.cgi?id=748646 , which I really should > solve sooner than later. Done! I'll do a 3.16 bug fix release soon so you can test if this bug is fixed with these patches.
I never got around to rolling out another 3.16 release but now that I'm about to roll-out 3.18.0 and F23 is soon out with it, I don't think i'll be fixing this in F22.
Fedora 22 changed to end-of-life (EOL) status on 2016-07-19. Fedora 22 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. If you are unable to reopen this bug, please file a new report against the current release. If you experience problems, please add a comment to this bug. Thank you for reporting this bug and we are sorry it could not be fixed.