jcarlock
ADSM.ORG Member
I have written a rather large tsm script in perl (For UNIX/Linux currently). I have added a couple of options for portability, but I think its ready to have a larger group's input.
Knowledge of Perl and TSM's select queries is all you will need to add/modify/update. Some things are specifically tuned to my environment, specifically the executed command 'dsmpdmc' has an encyrpted version with the UID/PASS included, but this is a global variable easily adapted to your environment. Also, my Exchange backup validation, is VERY specific to my environment, you will need to manually update this...
Also, I use RegExp's to parse the output for the emails, you can remove this value completely from any scripts not giving proper output.
That being said, I will only be willing to give basic support for this script as my time is limited, but I thought this might be helpful.-
Here is the -help output:
TSM Consolidated Monitoring Script v0.9
This utility monitors a range of TSM critical status info.
$cmdhint
[OPTIONS]
-s : -scratch
Queries TSM for available scratch tapes and alerts
when there are less than $scratch_threshold tapes.
-c : -countses
Queries the number of active Node sessions actively
connected to TSM and alerts if there are more
sessions tha the threshold of $maxnumsessions.
-dr : -drivesoffline
Queries TSM for any LTO3 Tape drives that are offline
and alerts if there are any offline.
-dbch : -dbcachehit
Queries the current TSM database cache hit ratio and
alerts if the value is less than $cachehitthreshold%.
-dbpu : -dbpercentutil
Queries the current precent utilization of the TSM
database and alerts if it is using more than $dbputhreshold%.
-lpu : -logpercentutil
Queries the current precent utilization of the TSM
database log and alerts if it is using more than $logputhreshold%.
-w : -waitses
Queries TSM for the number of node sessions with a wait
time of over 60 minutes (3600 seconds).
-disks : -diskpoolusage(-disks)
Queries TSM for any disk pools utilizing over 95%
of there allocated disk space.
-dm : -datamovers
Queries TSM for any NDMP Datamovers that are not online.
-x : -exchangedaily
Queries TSM DB for the pervious day's backup and alerts
if the scheduled backup was unsuccessful.
-df : -dailyfailed
Queries daily failed status from TSM events and alerts
for any incomplete status from the TSM event table.
-ds : -dailysummary
Queries TSM to generate a daaily summary of Unsuccessful
backups/archives from the TSM summary table.
-ac : -access
Queries TSM for servers that have not accessed the server
in over 2 days.
-f : -frequent
Executes a set of jobs normally scheduled to run frequently.
Including:
- Scratch Tape Count
- High # of sessions
- Offline Drives
- DB Cache Hit
- DB Percent Utilization
- DB Log Percent Utilization
- High number of waiting sessions
- High Disk Pool Usage
- Offline Datamovers
-d : -daily
Executes a set of jobs normally run daily at 3:30 PM, including:
- Daily Exchange Backup Verification
- Daily Incomplete Events
- Daily Unsuccessful Backups/Archives Summary
- Daily Access Check
-a : -all
Runs all monitoring/reporting including daily and frequent jobs.
-h : -help
Prints this message.
Knowledge of Perl and TSM's select queries is all you will need to add/modify/update. Some things are specifically tuned to my environment, specifically the executed command 'dsmpdmc' has an encyrpted version with the UID/PASS included, but this is a global variable easily adapted to your environment. Also, my Exchange backup validation, is VERY specific to my environment, you will need to manually update this...
Also, I use RegExp's to parse the output for the emails, you can remove this value completely from any scripts not giving proper output.
That being said, I will only be willing to give basic support for this script as my time is limited, but I thought this might be helpful.-
Here is the -help output:
TSM Consolidated Monitoring Script v0.9
This utility monitors a range of TSM critical status info.
$cmdhint
[OPTIONS]
-s : -scratch
Queries TSM for available scratch tapes and alerts
when there are less than $scratch_threshold tapes.
-c : -countses
Queries the number of active Node sessions actively
connected to TSM and alerts if there are more
sessions tha the threshold of $maxnumsessions.
-dr : -drivesoffline
Queries TSM for any LTO3 Tape drives that are offline
and alerts if there are any offline.
-dbch : -dbcachehit
Queries the current TSM database cache hit ratio and
alerts if the value is less than $cachehitthreshold%.
-dbpu : -dbpercentutil
Queries the current precent utilization of the TSM
database and alerts if it is using more than $dbputhreshold%.
-lpu : -logpercentutil
Queries the current precent utilization of the TSM
database log and alerts if it is using more than $logputhreshold%.
-w : -waitses
Queries TSM for the number of node sessions with a wait
time of over 60 minutes (3600 seconds).
-disks : -diskpoolusage(-disks)
Queries TSM for any disk pools utilizing over 95%
of there allocated disk space.
-dm : -datamovers
Queries TSM for any NDMP Datamovers that are not online.
-x : -exchangedaily
Queries TSM DB for the pervious day's backup and alerts
if the scheduled backup was unsuccessful.
-df : -dailyfailed
Queries daily failed status from TSM events and alerts
for any incomplete status from the TSM event table.
-ds : -dailysummary
Queries TSM to generate a daaily summary of Unsuccessful
backups/archives from the TSM summary table.
-ac : -access
Queries TSM for servers that have not accessed the server
in over 2 days.
-f : -frequent
Executes a set of jobs normally scheduled to run frequently.
Including:
- Scratch Tape Count
- High # of sessions
- Offline Drives
- DB Cache Hit
- DB Percent Utilization
- DB Log Percent Utilization
- High number of waiting sessions
- High Disk Pool Usage
- Offline Datamovers
-d : -daily
Executes a set of jobs normally run daily at 3:30 PM, including:
- Daily Exchange Backup Verification
- Daily Incomplete Events
- Daily Unsuccessful Backups/Archives Summary
- Daily Access Check
-a : -all
Runs all monitoring/reporting including daily and frequent jobs.
-h : -help
Prints this message.