On Fri, 10 May 1996 13:36:51 PDT David Boyes said:
>I would be very surprised to find this to be the case. Having
>recovered LISTSERV from several assorted disasters over the
>years, I have never encountered problems with VM:Backup being
>able to restore LISTSERV archive disks reliably.
 
VM:backup tells you if it had to switch to a physical dump, or it the
filespace/minidisk was skipped.  Switch/skip is a configuration option.
If backed up physically, the the entire minidisk has to be restored (to an
alternate location if recovering less than the entire minidisk), and even
then, the files might be semi-corrupted.  In the case of SFS, I think the
only option is to skip.  Note that SFS is NOT recommended for LISTSERV's
191.
 
>> themselves? I  was just  told by  one of the  people who  runs the  system on
>> which we run Listserv that if we were to shut Listserv down to back it up, it
>> would be down for  several hours so shutting our Listserv down  to back it up
>> isn't an option for us as far as I can see.
>
>This also puzzles me, unless the system LISTSERV is running on is
>*severely* CPU-overcommitted or I/O-constrained. Even if LISTSERV
>has several thousand files queued in the reader, the restart
>should take in the 5-10 minute range. If it's really taking
>several hours to restart LISTSERV, something else is wrong.
 
I think the down time he is talking about is for the dumps to complete.
LISTSERV initialization on a VM system is directly related to the size
of the signup files.  On UBVM, it can take 15-20 minutes.
 
 
Steve
([log in to unmask] | [log in to unmask] | (716)645-3564 ,
   | http://ubvm.cc.buffalo.edu/~tkssteve)