Playing around with db2diag

Over the past few days I have been playing with db2diag. It is the log analysis tool and it allows you to search, filter, and format the DB2 diagnostic logs. One of the simple examples is that I wanted to see when errors or warnings were produced on my test system. I could do that by calling:

db2diag -l Error,Warning -fmt "%ts %level %db"

This would only print the timestamp, the "error" level, and, if it was for a specific database, the database name.

2010-02-05-13.05.51.947000 Warning

2010-02-05-13.05.52.449000 Error PUREDB

2010-02-05-13.05.52.449000 Error PUREDB

2010-02-05-14.10.24.619000 Warning

The output is much quicker to digest than the diagnostic itself. And I could also use the output as input for some other scripts. Have you used db2diag in the past?

Electrocardiography (ECG), Health Checks, Performance Expert, DB2, and Oracle

Earlier today I wrote about a 24 hour ECG. That got me thinking. Did you know that IBM Information Management Services actually offer health checks? You can have a "health" specialist (in this case not a physician!) look over your database installation. Like you would expect, if there are findings, recommendations are given and maybe a prescription is handed over.

Similar to the 24 hour ECG, there is also a product called DB2 Performance Expert that does exactly (and much more) what this wearable mini-computer does. It collects all the data points and lets you evaluate them, even assists you in that.

BTW: Some non-DB2 (but Oracle) customers come to see the health specialists, too. They are suffering from high costs and other issues. Read more about it here.

Electrocardiography (ECG), Twitter, and Databases

Someone in my family currently is having a 24 hour electrocardiography (ECG). Basically, the heart region is fully wired and connected to a "wearable" mini-computer/database and data is collected over a 24 hour period. That part of the 24 h ECG is fully automated. The person in question gets wired, the computer set up, and everything is ready to go. The more complicated part is keeping an activity log. Think of it as a silly form of Twitter.

You write down the time and the start of activities:
16:05 running
16:15 sitting and talking
16:40 reading
...
18:18 eating & drinking
...
x:yy sleeping

After the recording period is over, the data is evaluated manually. Someone will take a look at the plotted graphs, look for "interesting" peaks, gaps, dense graphs, etc. and correlate that with the activity log.

Wouldn't that be something that could be done by a database application and mostly automated? Looking for patterns, for irregularities, correlating timestamped data, and eventually reporting back a list of events to take a closer look? Instead of keeping a paper-based log, timestamped activities could be stored with the mini-computer/database (enter it with a wireless device or have that person wear a USB connector... ;-) or correlated with the twitter account...