We are seeing many reports that time out in 7.1.9. These reports ran fine in 7.1.5 with the default timeout setting - but now we have to adjust that figure upwards. Has anyone else seen this? Is there a fix for this?
Have you confirmed that these reports are defined exactly the same (e.g., timeout value, maximum number of rows, etc)? If yes, could it be due to the poor performance of your 7.1.9 environment? Normally comparing alert logs should tell you why certain reports time out. You should see error stack in the pega logs as well.
There could a number of reasons why a report is timing out. The indexes might need a rebuild, the statistics could be stale, too much fragmentation of the table etc. As Kevin suggests, the alert log is a good start. If you are using AES or PDC, it could even show if indexes are present or not. Doing an explain plan could also help find the problem for you.
Kevin and Rajiv your points are well taken and understood. I have reviewed the alert logs and PDC many times since the upgrade and continue to do so. However this specific problem where a report times out is new since the upgrade to 7.1.9. I don't agree that index problems, stale statistics or fragmentation appeared after upgrading; I was curious if anyone else had seen this timeout issue as a result of upgrading to 7.1.9. We have seen issues in other areas crop up after upgrading (for which HFIXs have been created) so we were curious if this fell in that category as well. If not, we will continue to pursue solutions using logs/PDC/explain plan, etc.