Summary: | KNotify crashed returning from sleep | ||
---|---|---|---|
Product: | [Frameworks and Libraries] Phonon | Reporter: | Unknown <null> |
Component: | general | Assignee: | Matthias Kretz <kretz> |
Status: | RESOLVED FIXED | ||
Severity: | crash | CC: | benwilby, colin, dheeru, martin.sandsmark, myriam, null, richts, sandroandrade |
Priority: | NOR | ||
Version: | 4.4.0 (KDE 4.4.2) | ||
Target Milestone: | --- | ||
Platform: | openSUSE | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: |
Description
Unknown
2010-05-06 10:06:19 UTC
Bus errors generally happen when bad things happen with SHM. It seems that the SHM region is not setup properly after suspend and that subsequently causes this problem. I'll ask around and see if anyone else has had this problem before (not seem it before) and what do to about it. *** Bug 241488 has been marked as a duplicate of this bug. *** With root user I can do anything on shm: szotsaki:/dev/shm # ls -lh összesen 2,0G -rw-r--r-- 1 root root 2,0G jún 28 22.26 preloadtrace.log -r-------- 1 aki users 65M jún 28 23.41 pulse-shm-99266583 -rw-r--r-- 1 root root 0 jún 27 21.19 warpclock szotsaki:/dev/shm # du -h 2,0G . szotsaki:/dev/shm # df -h Fájlrendszer Méret Fogl. Szab. Fo.% Csatl. pont /dev/sda6 483G 174G 285G 38% / szotsaki:/dev/shm # touch tempFile szotsaki:/dev/shm # echo "temptemptemptemptemptemptemp" > tempFile szotsaki:/dev/shm # cat tempFile temptemptemptemptemptemptemp With a normal user I can create files there but then I cannot write into them (I don't know if it's normal): aki@szotsaki:/dev/shm> touch temp2 aki@szotsaki:/dev/shm> echo "temptemptemptemp" > temp2 bash: echo: write error: No space left on device That 2 gig file looks suspicious. What puts that there and why? Whatever the problem is, abusing SHM like that is probably the root cause. There is likely an upper limit or a similar concept to root-reserved blocks in SHM mounts. Either way that's the problem. Try and work out how to avoid that massive file and the problem will likely go away. Hmm, I simply deleted the 2 GB file. It contained something like this: 0.028 384 boot.startprelo fork 405 0.028 405 boot.startprelo stat . 0.028 405 boot.startprelo stat /bin/pidof 0.028 405 boot.startprelo stat /bin/pidof 0.028 405 boot.startprelo stat /bin/pidof 0.028 405 boot.startprelo stat /bin/pidof 0.028 405 boot.startprelo stat /bin/pidof 0.028 405 boot.startprelo stat /bin/pidof 0.028 405 boot.startprelo execve /bin/pidof 0.028 405 pidof open 3 /etc/ld.so.cache ... 53.050 3364 plasma-desktop stat /etc/localtime 53.053 3467 kontact stat /etc/localtime 53.053 3467 kontact stat /etc/localtime 53.058 3358 kwin stat /etc/localtime 53.058 3358 kwin stat /etc/localtime 53.065 3358 kwin stat /etc/localtime 53.065 3358 kwin stat /etc/localtime 53.065 3358 kwin stat /etc/localtime 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.066 3363 knotify4 fstat 14 53.069 3358 kwin stat /etc/localtime 53.069 3358 kwin stat /etc/localtime 53.069 3358 kwin stat /etc/localtime 53.070 3467 kontact stat /etc/localtime 53.070 3467 kontact stat /etc/localtime 53.071 3364 plasma-desktop stat /etc/localtime ... But it has been created again. Interestingly, after a day usage its size is 1 GB. I haven't experienced KNotify crash, but at the time the file reaches its upper limit (probably 2 GB) I'll write here. *** This bug has been marked as a duplicate of bug 246448 *** Don;t think the duplicates stuff is right... I'll reopen this one and remark another bug as a dupe of this in a moment :) *** Bug 246762 has been marked as a duplicate of this bug. *** *** Bug 254609 has been marked as a duplicate of this bug. *** *** Bug 260298 has been marked as a duplicate of this bug. *** Are there some people who noticed this crash lately? On openSUSE 12.1 there's nothing interesting in /dev/shm (almost empty). In the meantime I switched to phonon-mplayer and maybe because of that I don't experience this bug any more. Is this still relevant? I think this bug got fixed in the meantime. |