[SOLVED] WIreshark 2.0 segfaults on Slackware64-current

1337_powerslacker · 11-29-2015, 08:17 PM

Quote:

Originally Posted by ljb643

I'm using 14.1, and no I have not made any progress in getting a build to work with Qt, nor am I getting any useful traceback information that would help someone figure it out. The gdb traceback seems to imply something is seriously wrong, like the stack got overwritten.

Those distros split the application up into multiple packages. Perhaps you didn't download the one that has the binary? There is a core wireshark-2.0.0 package, but also wireshark-ui-gtk and wireshark-ui-qt (which contain the UI binaries).

I stand corrected. I didn't know that was the case with other distros; that could be because Slackware doesn't bother with such nonsense, and puts it all into one package. Oh well. I'm going to keep on running 1.12.8 until I see that 2.0.0 has been fixed.

ljb643 · 11-30-2015, 09:06 PM

Someone just opened a Wireshark bug report for the same problem:
https://bugs.wireshark.org/bugzilla/...g.cgi?id=11833
I can't be sure, but it does NOT looks like he is running Slackware, so this is probably not a Slackware-specific problem.

ljb643 · 12-01-2015, 06:38 PM

And... the bug is now marked Fixed, Resolved. That was really fast. I tried the patch which is linked from the bug report, and it does result in a working Wireshark-2.0 with Qt UI on Slackware. (I'm not suggesting anyone needs to try it, but it tells me that this will be fixed in the next Wireshark release, and also that it wasn't a problem specific to Slackware.)

1337_powerslacker · 12-02-2015, 06:00 PM

Quote:

Originally Posted by ljb643

And... the bug is now marked Fixed, Resolved. That was really fast. I tried the patch which is linked from the bug report, and it does result in a working Wireshark-2.0 with Qt UI on Slackware. (I'm not suggesting anyone needs to try it, but it tells me that this will be fixed in the next Wireshark release, and also that it wasn't a problem specific to Slackware.)

Thanks for posting about the fix. I was able to apply the patch to the source, and re-compile, and it successfully compiled and ran without a segfault. Wonderful!

NOTE: To those who find it rather daunting to find the patch from the bug report (it is indeed linked, but rather difficult to find), so for convenience's sake, I will post it here. Another testimony to the power of open source!

Regards,

Matt