I believe you have two more paths to investigate:
1) Script something that parses daemon.log for entries that indicate saveset
failures and alert on it.
2) Within the nsr group resource there is a hidden attribute called
completion that gives information about completed savesets (including
success/failure info) for a given savegroup that is updated throughout the
run of the savegroup. Again, in this case you could script something that
uses nsradmin to query for running savegroups and then parse the completion
attribute and alert on failures found.
You can also combat this issue of more timely feedback a different way by:
1) Creating smaller savegroups; perhaps something like Daily_1800,
Daily_1900, Daily_2000, etc, for each hour and then put just a handful of
clients in each. These groups will run much faster and you'll get more
timely feedback.
2) Create a 2nd group called something like Daily_Slow and put the slower,
problem prone clients in it. Chances are you have one or more clients that
consistently have problems and pulling them aside will decrease the ability
for their problems to affect the group at large.
Good luck!
-Chris
----- Original Message -----
From: "Kalbus Johanes" <Johanes.Kalbus AT ITELLIUM DOT COM>
To: <NETWORKER AT LISTMAIL.TEMPLE DOT EDU>
Sent: Tuesday, October 22, 2002 10:01 AM
Subject: [Networker] AW: [Networker] Client failure notification
You won't get any luck doing what you wish with Legato Networker - the
completion report is only sent after the complete backup is run. Using
savepnpc will not help either - because the pre and post parts are run
before and after the backup.
Only thing you might try is to write a wrapper script for save which is
named identical on all the clients. This script should run the save command,
take the output send it to the networker and check it at the same time. In
Unix I'd use the "tee" command in a script to accomplish that. The doubled
output can be sent to you by mail in every case you define.
Until now I haven't done that, because we've split our backup groups.
Maximum 2 to 3 clients per group. So it does not take that long, gives you
the ability to control the start times more precisely and clutters your
daemon.log or messages file with lots of completion reports :-)
Yours
Johanes
-----Ursprüngliche Nachricht-----
Von: Michael Cope [mailto:Michael_Cope AT HOMEDEPOT DOT COM]
Gesendet: Montag, 21. Oktober 2002 18:12
An: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Betreff: [Networker] Client failure notification
Hi all,
I'm wondering if anyone has created a notification to alert when a client
within a group fails. I have a number of NT/2K servers within a group that
typically runs for about 10 hours. Unfortunately, this means I'm not aware
of client failure until the end of the backup when I receive the email. I
haven't had much luck with the manual or Legato's website on figuring out
creating notifications or the process to monitor that generates the error
message.
Thanks,
Michael
--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
--
Note: To sign off this list, send a "signoff" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
|
