Managers,
I have recently been updating our machines here to TruUnix 5.0 due to Y2K
issues and have two problems.
1. Our machine serving /home and /usr/local/share is still running under
4.0D due to time constraints on my side (upgrade is scheduled for the end
of next week). The problem is that file locking between the 5.0 and the
4.0D machines does not work. rpc.statd and rpc.lockd run on all machines,
but whenever trying to lock a file on an NFS mounted file system that
resides on the 4.0D machine, the locking process hangs. Good example for
this is vi, which flocks the file before opening it. From the looks of it
the file itself is locked correctly on the server, but the system call
never terminates, as if there is no positive response back to the client
machine where flock was executed. Has anybody else experienced this problem
and has a bug-fix? If not I would have to upgrade the 4.0D machine earlier
than planned since lots of programs require working locks... (I had
originally planned to upgrade the 4.0D machine only after solving the next
problem).
2. The second problem is related to the TruUnix Patch #1 for 5.0. Several
of our machines are servers for RAID arrays (specifically, the OXYGEN
RAID controller from Advanced Unibyte). These arrays are recognized
without problems with the base 5.0 distribution. AFTER applying the
TruUnix Patch #1, however, the arrays are not being recognized
anymore. I THINK this might have to do with TruUnix Patch 45.00
(OSF500-46), which supposedly fixes an error in the SCSI device probe
code. I have not tried yet to remove this patch since I would first like
to know whether anybody else has experienced the same problem and since I
can only play with the machines during the weekend when the arrays are not
needed.
I will summarize any responses I might get,
cheers,
Joern
---
Dr. Joern Wilms
Univ. Tuebingen, Institute for Astronomy
Waldhaeuser Str. 64 (phone) +49 7071 29-76128
D-72076 Tuebingen, Germany wilms_at_astro.uni-tuebingen.de
Received on Thu Jan 13 2000 - 16:16:41 NZDT