Summary: | Fresh SVN d/l, compile & install. Runs OK from cmdline. Quit GUI normally. Kile crash (segfault) after 6 minutes. | ||
---|---|---|---|
Product: | [Applications] kile | Reporter: | Paul Wragg <wagglet> |
Component: | general | Assignee: | Michel Ludwig <michel.ludwig> |
Status: | RESOLVED DUPLICATE | ||
Severity: | crash | CC: | andresbajotierra |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Unlisted Binaries | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Attachments: |
Terminal transcript of complete Kile dowload and session.
Output of System Check post install. Terminal transcript of 2.0.84, gave up after over an hour! |
Description
Paul Wragg
2010-01-25 23:06:37 UTC
Created attachment 40229 [details]
Terminal transcript of complete Kile dowload and session.
This is a transcript of ALL terminal input & output from initial download to running of latest SVN code.
Created attachment 40230 [details]
Output of System Check post install.
System check output.
Comment on attachment 40229 [details]
Terminal transcript of complete Kile dowload and session.
OK I'm a moron. At the end of this file, where I run Kile, I missed out the './'. Quite stupid. kile.transcript.2 is another script from the point of running Kile. Everything else has remained the same. Sorry!
(In reply to comment #3) > (From update of attachment 40229 [details]) > OK I'm a moron. At the end of this file, where I run Kile, I missed out the > './'. Quite stupid. kile.transcript.2 is another script from the point of > running Kile. Everything else has remained the same. Sorry! I don't think that 'kile.transcript.2' has been uploaded correctly... (could you maybe also choose a text mime type?) It would be interesting to see the output of Kile when it doesn't terminate correctly. Thanks. Hi, RE: kile.transcript.2 - I'm still waiting for Kile to segfault - it hasn't released my terminal in over 15 minutes! As soon as it's completed I'll post it. In the meantime I'll change the mimetypes! ;) Update: nearly ONE HOUR from quitting the Kile 2.0.84 GUI, the kile process as not terminated and released my terminal session. This is in stark contrast to version 2.0.83. I don't think 2.0.84 is going to quit at all... Output of ps aux at T + 50 minutes: kile-install/bin> ps aux | grep kile paul 18565 0.0 0.0 3068 708 pts/0 S+ 11:55 0:00 script kile.transcript.2 paul 18566 0.0 0.0 3072 436 pts/0 S+ 11:55 0:00 script kile.transcript.2 paul 18597 0.2 2.7 120584 56716 pts/1 Sl+ 11:55 0:04 ./kile paul 18625 0.0 0.2 36512 5696 ? S 11:55 0:00 kdeinit4: kio_file [kdeinit] file local:/tmp/ksocket-paul/klauncherT14323.slave-socket local:/tmp/ksocket-paul/kileZ18597.slave-socket kile-install/bin> Created attachment 40233 [details]
Terminal transcript of 2.0.84, gave up after over an hour!
Not much in this one. The Kile process ran for well over an hour after quitting the GUI. In the end I had to send a HUP from another window to release the terminal. I didn't just hit CTRL-C since I wanted to see if sending a signal would make it do something (like crash). Normal CPU usage was not affected. I/O, swap, mem and net usage were all baseline. It was like the process was idle, waiting for something.
It would seem that 2.0.84 is worse at quitting than 2.0.83 - the earlier version takes up to 15 minutes to quit (albeit ungracefully) but 2.0.84 hangs on until someone or something actively kills it. At least, these are my findings so far.
PS output after HUP:
kile-install/bin> ps aux | grep kile
paul 18625 0.0 0.2 36892 5784 ? S 11:55 0:00 kdeinit4: kio_file [kdeinit] file local:/tmp/ksocket-paul/klauncherT14323.slave-socket local:/tmp/ksocket-paul/kileZ18597.slave-socket
kile-install/bin>
It would almost be worth a VNC session / movie to give a true representation of what's happening (or not)...
Can you maybe try it again without using "script", i.e. just in a terminal, and then paste the output here? (I assume that Kile was compiled in debug mode and that it prints something when it is executed) The last lines of output will probably be crucial for fixing this issue... - Isn't this related to bug 220343 ? Regards (In reply to comment #9) > - Isn't this related to bug 220343 ? Regards Yes, it's the same thing. I just wanted to get some more information here but I've got what I wanted now in 220343. *** This bug has been marked as a duplicate of bug 220343 *** |