That would depend on the results of your business requirements study, site
and regional disaster business continuation plans, logical and physical
telecommunications pathway redundancy, service level agreements, and your
cost per hour versus time of outage graph for loss of service. And a few
other things.
Seriously, the monitoring functionality required and its allowed cost
horizon should fall out of the business requirements, not some generic
notion of what it is nice to keep an eye on. Now, that being said here are a
few things I don't think you should be without:
- Typical "heartbeat" transactions that verify your needed point-to-point
capabilities and suggest that in lieu of "locking issues" your physical
plant should be delivering reasonable OLTP response versus expectations.
- A history of disk growth graph quantized by the the size of storage it is
easy to add to your disk farm without disrupting your i/o signature.
(for example, don't plan to add one disk {or a few disks} that will get
all the hot new pieces of everything as soon as you add it).
- At least monthly a check with your vendors that they continue to
guarantee disk farm and computer replacement/augmentation parts for a moving
6 or nine month window.
- A history of your headroom for CPU utilization.
- An alarm level warning versus a workshift plan for CPU utilization level
(tell someone if use is unexpectedly very high, even if it is not screwing
up your heartbeat transacations.)
- Unexplained or unplanned changes in the rate of generation of redo.
- For recurring jobs, a history of job completion times, preferably with
desired, investigate, and alarm/deadline time thresholds.
- Holy cow, this could be a really long list. Try digging up the old MOSES
papers for a start. If you can't find those, get someone's old MVS glass
house operations manuals. Those guys always wore belts and suspenders (aka
braces, not the sock thingies.)
mwf
-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org]On Behalf Of VIVEK_SHARMA
Sent: Wednesday, July 07, 2004 5:50 AM
To: oracledba_at_lazydba.com; oracle-l_at_freelists.org
Subject: Monitoring activities for Large Centralized Production Hybrid
Database?
Folks
What monitoring activities need to be set in place for a Critical
production,=20
near 24/7, Centralized (500 GB) banking application Database?
Thanks
Vivek
Please see the official ORACLE-L FAQ:
http://www.orafaq.com
To unsubscribe send email to: oracle-l-request_at_freelists.org
put 'unsubscribe' in the subject line.
--
Archives are at
http://www.freelists.org/archives/oracle-l/
FAQ is at
http://www.freelists.org/help/fom-serve/cache/1.html
Please see the official ORACLE-L FAQ:
http://www.orafaq.com
To unsubscribe send email to: oracle-l-request_at_freelists.org
put 'unsubscribe' in the subject line.
--
Archives are at
http://www.freelists.org/archives/oracle-l/
FAQ is at
http://www.freelists.org/help/fom-serve/cache/1.html
Received on Wed Jul 07 2004 - 08:41:11 CDT
