[RDD] caed segfault on Ubuntu 10.04

Christian Pointner equinox at helsinki.at
Thu Jul 22 07:06:37 EDT 2010


Hi

Phil Ashby schrieb:
> The crash seems to happen in a different thread than the owner of fd 9,
> can you re-run in gdb and grab some stack traces, eg:
> 
[...]
i will do this as soon i'm back at work.

> That should show you/us where each thread is at the time of failure. Given
> your strace output below, it seems that an ioctl() call fails in pid 2310
> (returning EIO) which triggers the crash, however we can't see what fd is
> associated with the ioctl() nor what it refers to: should be a bit further
> back in the trace...
> 
as i mentioned below fd 9 belongs to the alsa device (/dev/snd/pcmC1D2p
in our case).

regards
 christian


> Phil.
> 
> equinox <equinox at helsinki.at> wrote:
> 
> Hello,
> 
> We upgraded our workstations to Ubuntu 10.04 and since then we are
> experiencing some problems with Rivendell. At about 1 in 3 tries to
> start the rivendell daemons caed crashes with a segfault. When caed
> crashes it always shows the same behaviour. The initialization completes
> and everything looks just normal but after about 20 seconds the daemon
> crashes due to a SIGSEGV. See the output of strace below
> 
> <snippet>
> ...
> [pid  2308] select(7, [3 4 6], [], [], {0, 19735}) = 0 (Timeout)
> [pid  2308] gettimeofday({1279709836, 849541}, NULL) = 0
> [pid  2308] gettimeofday({1279709836, 849590}, NULL) = 0
> [pid  2308] select(7, [3 4 6], [], [], {0, 19730} <unfinished ...>
> [pid  2309] <... ioctl resumed> , 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150, 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150, 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150, 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150 <unfinished ...>
> [pid  2308] <... select resumed> )      = 0 (Timeout)
> [pid  2308] gettimeofday({1279709836, 869536}, NULL) = 0
> [pid  2308] gettimeofday({1279709836, 869584}, NULL) = 0
> [pid  2308] select(7, [3 4 6], [], [], {0, 19736}) = 0 (Timeout)
> [pid  2308] gettimeofday({1279709836, 889539}, NULL) = 0
> [pid  2308] gettimeofday({1279709836, 889588}, NULL) = 0
> [pid  2308] select(7, [3 4 6], [], [], {0, 19732} <unfinished ...>
> [pid  2309] <... ioctl resumed> , 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150, 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150, 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150, 0x7f94f15bed10) = 0
> [pid  2309] ioctl(9, 0x40184150 <unfinished ...>
> [pid  2308] <... select resumed> )      = 0 (Timeout)
> [pid  2308] gettimeofday({1279709836, 909577}, NULL) = 0
> [pid  2308] gettimeofday({1279709836, 909700}, NULL) = 0
> [pid  2308] select(7, [3 4 6], [], [], {0, 19620} <unfinished ...>
> [pid  2310] <... ioctl resumed> , 0x7f94f0dbdd50) = -1 EIO (Input/output
> error)
> [pid  2310] --- SIGSEGV (Segmentation fault) @ 0 (0) ---
> Process 2310 detached
> [pid  2309] +++ killed by SIGSEGV +++
> +++ killed by SIGSEGV +++
> </snippet>
> 
> fd 9 which the ioctl command uses belongs to the open alsa device
> (/dev/snd/pcmC1D2p in our case).
> Can anybody help us out with this problem?
> 
> regards
>   christian
> 
> _______________________________________________
> Rivendell-dev mailing list
> Rivendell-dev at lists.rivendellaudio.org
> http://lists.rivendellaudio.org/mailman/listinfo/rivendell-dev



More information about the Rivendell-dev mailing list