Nick Hudson wrote:
> On Tuesday 16 November 2004 08:57, Martti Kuparinen wrote:
> [...]
> 
>>And from ktrace:
>>
>>  26117 master   GIO   fd 2 wrote 78 bytes
>>        "**Pth** SCHEDULER INTERNAL ERROR: no more thread(s) available to
>>sched\ ule!?!?
>>        "
>>  26117 master   RET   write 78/0x4e
>>  26117 master   CALL  fcntl(0x2,0x3,0)
>>  26117 master   RET   fcntl 6
>>  26117 master   CALL  fcntl(0x2,0x4,0x2)
>>  26117 master   RET   fcntl 0
>>  26117 master   CALL  __sigprocmask14(0x3,0x8078e58,0)
>>  26117 master   RET   __sigprocmask14 0
>>  26117 master   CALL  getpid
>>  26117 master   RET   getpid 26117/0x6605
>>  26117 master   CALL  kill(0x6605, SIGABRT)
>>  26117 master   PSIG  SIGABRT SIG_DFL
>>  26117 master   NAMI  "master.core"
> 
> 
> IIRC, that is usually a symptom of the program closing all fds. pth uses some 
> internally to help thread scheduling.

I've only updated perl58, db4 and cyrus-sals2. Could it be any of these?

Martti