Thunderbird Randomly Segfaults At Startup

Home » CentOS » Thunderbird Randomly Segfaults At Startup
CentOS 10 Comments

Anyone else seeing this? This is happening on a straight “Desktop” install of CentOS 6.6 fully updated. I have made no changes other than installing Thunderbird. Thunderbird will occasionally (10 to 20% of the time) fail to start, and the abrt report indicates a signal 11 (SIGSEGV). I don’t see any other reports of this. Is no one using Thunderbird these days?

Do abrt reports submitted from CentOS go anywhere useful these days? I
saw some discussion a while back, but don’t recall the results. Should I submit one?

10 thoughts on - Thunderbird Randomly Segfaults At Startup

  • I have the same problem with Thunderbird and Firefox after the update from C6.5 to C6.6. I’ve wined and complained about it on this list in the past but have not seen any useful response. Other users on this same system have the same issue so it’s not my personal profile that’s causing the issue.

    The C6.6 update also broke web camera functionality. I understand there’s a kernel patch that fixed the web camera issue but it has yet to ship in any production C6 32 bit or 64 bit kernels.

    To work around the Thunderbird/Firefox issue I start them in a terminal:

    [mlapier@mushroom ~]$ thunderbird &
    [mlapier@mushroom ~]$ firefox &

    If they fail to start after a reasonable time I use the up cursor key and replay the command until it does work.

    [mlapier@mushroom ~]$ thunderbird &
    [1] 22072
    [mlapier@mushroom ~]$ [calBackendLoader] Using libical backend at
    /home/mlapier/.thunderbird/1qzdaguv.default/extensions/{e2fda1a4-762b-4020-b5ad-a41df1933103}/components/libical.manifest enigmail.js: Registered components mimeVerify.jsm: module initialized

    [mlapier@mushroom ~]$ firefox &
    [2] 22125
    [mlapier@mushroom ~]$ firefox &
    [3] 22158
    [2] Segmentation fault (core dumped) firefox
    [mlapier@mushroom ~]$ firefox &
    [4] 22244
    [3] Segmentation fault (core dumped) firefox
    [mlapier@mushroom ~]$ firefox &
    [5] 22340
    [4] Segmentation fault (core dumped) firefox
    [mlapier@mushroom ~]$ firefox &
    [6] 22426
    [5] Segmentation fault (core dumped) firefox
    [mlapier@mushroom ~]$

    CentOS release 6.6 (Final)

    Linux mushroom.patch 2.6.32-504.3.3.el6.i686 #1 SMP Tue Dec 16 22:55:44
    UTC 2014 i686 i686 i386 GNU/Linux

    21:54:52 up 1 day, 13:27, 3 users, load average: 2.02, 2.07, 2.14

    ^ ^ Mark LaPierre Registered Linux user No #267004

  • Another data point: If I install CentOS 6.6 from the distribution ISO and do _not_ do an update, Thunderbird seems to start reliably — 100 starts and no failures. As soon as I install the current set of updates, it starts failing.

  • I haven’t seen this, but I’m curious if it has something to do with your configuration. What I would suggest trying is working from a clean config.

    mv ~/.thunderbird ~/.thunderbird.backup

    Try it like that. If it starts correctly, then it is your config and you’ll probably have to recreate your config to make it work. If it doesn’t, then rename .thunderbird.config back to .thunderbird. No harm done.

    You can also install the latest version of thunderbird from and install it in /opt or /usr/local or where ever you are comfortable.

    – Jason

  • Downgrading nss-softokn to 3.14.3-17.el6 does _not_ solve the issue for me. Thunderbird still segfaults frequently on startup.

  • I did reboot. After some experimenting, I found it was the _library_
    package, nss-softokn-freebl, that needs to be downgraded to 3.14.3-17.el6. Surprisingly, the newer nss-softokn-3.14.3-10.el6_6 is happy to coexist with that older library.

  • Oh, boy. We do seem to have to treat Linux like Windows these days and
    _reboot_ after any update, not only kernel or glibc update. So much for
    “Unix-like”, sigh ;-(


    Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247

  • When you are trying to track down unexplained and only semi-repeatable behavior, you have to know that you are starting from a known state. Heck, I did at least 5 re-installs of the OS during the course of this.