Subject: Re: [netatalk-admins] Uninterruptably sleeping processes
From: Michael M Han (han@windy.ckm.ucsf.edu)
Date: Fri Jul 10 1998 - 12:44:10 EDT
What release of netatalk on what platform? There's a small fix for
1.4b2 and all derivatives for Solaris (on Solaris child afpd's may
exit with the EAGAIN signal, which netatalk doesn't know how to handle
w/o the patch). I heard a muttering here about a month ago about
similar behavior on FreeBSD.
Previously...
>
>I hope someone can help with this:
>
>If a Mac dies, and leaves a share hanging, it sometimes leaves an afpd
>process hanging in the 'D' state... AFAIK I can't get rid of these any
>way short of a reboot, and each one of these uninterruptable sleepers
>appears to bump the load up by 1.
>
>Any ideas on how to:
>
>a) get rid of the 'undead' processes
>b) stop them from becoming undead in the first place
>
>would be appreciated.
_________
mike (han@library.ucsf.edu)
I will not show off
- The collected wisdom of Bart Simpson
This archive was generated by hypermail 2b28 : Sat Dec 18 1999 - 16:32:54 EST