Bug 1380 - got trunk r2965 to freeze up and use 100% of cpu
: got trunk r2965 to freeze up and use 100% of cpu
Status: RESOLVED WORKSFORME
Product: Logitech Media Server
Classification: Unclassified
Component: Misc
: 6.1.0
: Macintosh MacOS X 10
: P2 major (vote)
: ---
Assigned To: Vidur Apparao
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2005-04-14 10:52 UTC by Kevin Pearsall
Modified: 2011-03-16 04:18 UTC (History)
0 users

See Also:
Category: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Kevin Pearsall 2005-04-14 10:52:16 UTC
the last thing spit to stdout was:
Use of uninitialized value in string eq at
/Users/kpearsall/trunk/server/Slim/Player/Source.pm line 1651.
Use of uninitialized value in numeric gt (>) at
/Users/kpearsall/trunk/server/Slim/Player/Source.pm line 1758.
Use of uninitialized value in subtraction (-) at
/Users/kpearsall/trunk/server/Slim/Player/Source.pm line 1760.
Use of uninitialized value in numeric gt (>) at
/Users/kpearsall/trunk/server/Slim/Player/Source.pm line 1770.
Use of uninitialized value in string ne at
/Users/kpearsall/trunk/server/Slim/Player/Source.pm line 836.

- i was playing radioio80s at the time.
- r2965 on osx 10.3.8 on mac mini with 1gb ram
- very small library, use itunes is on and music folder is set to an alternate
location (my home dir)
- will attach a ktrace
- at some point this morning i was regressing the volume sync on song restart
issue, and had a slimp3 as well as an sb2 connected, set to sync.  at some point
i unplugged the slimp3, if that matters at all.
- left it going for about 15 minutes at 100% cpu and it never came back to
normal functioning...players can't connect now either, nor will the web
interface load.
Comment 1 Kevin Pearsall 2005-04-14 11:04:01 UTC
ok won't be attaching ktrace since it's enormous, but it just says this repeatedly:
 21491 perl     RET   read 0
 21491 perl     CALL  read(0x12,0x4916650,0x1)
 21491 perl     GIO   fd 18 read 0 bytes
       ""
Comment 2 Vidur Apparao 2005-04-25 10:15:53 UTC
Sounds like we're spinning in the select loop (an invalid socket in the read
list?). Unfortunately, this is hard/impossible to fix without a reproducible
test case.
Comment 3 Kevin Pearsall 2005-04-26 14:33:58 UTC
ok well i'm going to close this as worksforme since:
- i have crap for details
- i haven't seen this in a while
- nobody has called with a similar issue