We created a script for the TSM server to email our network admins every
morning. This way if the server backup completes we can tell if it had
problems with any files without querying the log. This is an example.

Cory Curtis
Salt Lake County, IS Dept.
Salt Lake City, Utah
Network Admin

03/16/01
***************************************************************
*
* Scheduled backup completion report follows --
*
***************************************************************
Tivoli Storage Manager
Scheduled Start          Actual Start             Schedule Name     Node
Name         Status
--------------------     --------------------     -------------
-------------     ---------
03/15/01   02:30:00      03/15/01   03:00:03      EXCHANGE_SCH-     SERVER1
Completed
                                                   ED

03/15/01   19:25:09      03/15/01   19:25:42      CLIENT_SCHED-     SERVER2
Completed
                                                   ULE

03/15/01   19:25:09      03/15/01   19:31:54      CLIENT_SCHED-     SERVER3
Failed
                                                   ULE

03/15/01   21:00:00                               CLIENT_SCHED-     SERVER4
Missed
                                                   ULE

03/15/01   21:00:00      03/15/01   19:45:10      CLIENT_SCHED-     SERVER5
Completed
                                                   ULE

The following clients missed their scheduled backup window in the previous
24 hours

Nodename       Status
-------------  ---------
SERVER3        Failed
SERVER4        Missed

The following clients have not accessed the TSM system for more than 2 days

Nodename                Platform       # days since last access
--------                --------       ------------------------
Workstation6             WinNT                    11
Workstation7             WinNT                    11

The following clients had errors during their last backup

SERVER3                   # of failures   12
SERVER8                   # of failures   14
SERVER9                   # of failures   22
SERVER10                  # of failures   12
SERVER11                  # of failures   3

-----Original Message-----
From: Pearson, Dave [mailto:[EMAIL PROTECTED]]
Sent: Friday, March 16, 2001 9:37 AM
To: [EMAIL PROTECTED]
Subject: Re: client notification


We created one that tells us what backup failed in the last 24 hours. Thanks
goodness most of the time it come back and say "All backups was successful"
(we created a script for this)

Dave Pearson
Snohomish County PUD
Everett, Washington
Production Control Analyst
[EMAIL PROTECTED]


> -----Original Message-----
> From: Rushforth, Tim [SMTP:[EMAIL PROTECTED]]
> Sent: Friday, March 16, 2001 8:24 AM
> To:   [EMAIL PROTECTED]
> Subject:      Re: client notification
>
> You could also query the server activity log to see files that failed.  We
> do perl query like:
> if (/ANE4959I/ && !/Total number of objects failed:           0/)
>
> which gives us a line for each client (inlcuding the node name) that has 1
> or more failed files.  You could then send to the appropriate users.
>
> Tim Rushforth
> City of Winnipeg
>
> -----Original Message-----
> From: Prather, Wanda [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, March 15, 2001 12:03 PM
> To: [EMAIL PROTECTED]
> Subject: Re: client notification
>
>
> Part of the problem is, who is the "user".
>
> Under WinnT, scheduled backups generally run under the System account.
> Any
> USER account may or may not be logged on at the time backups are run.  On
> UNIX, the backups generally run as root.
>
> What we do is query the eventlog on the server end, and generate mail
> messages to be sent to users whose backups fail or miss.  That works if
> you
> have a way to map the TSM node names into mail ids.  But, it doesn't take
> into account any individual files that fail to back up, we really don't
> have
> a solution for that yet.

Reply via email to