Archive logs filled fast

[Former Member]

We are running production system sap_basis version 702 sp13 on oracle version 11.2.0.2.0 . Our kernel release is 721 patch 600.

I am getting one issue regarding oracle archive log which is filled very fast compare to other days. It got filled up within 6 hour.

We are taking backup of archive log daily evening and clearing it.

We always keep the space of 140-180 GB in archive log directory. this is the first time i faced this issue while working with client.

I have checked the background jobs and there is no much difference on that compare to other days.

Please let me know, how to find the reason of filling archive log so fast?

Thanks in Advance

Neelkanth

[Former Member · Accepted solution]

Hi,

All the changes are written in log, so the main reason will be the applications / background jobs those running during that day as you mentioned that it is OK on other days. You may start analyzing this issue using the OSS note below. Also check in alert log file for any errors / warning messages related archive log generation during that particular day.

584548 - Unusually high number of redo logs

Regards,

Ganesan

[Former Member · Accepted solution]

Thanks Ganesan.

I have checked all the points mentioned in the note but unable to find anything as everything is perfect according to the note.

I am able to find only one thing that IO Redo Log per hour is more on friday compare to other days. I have checked the background jobs and done the dialog analysis between that period but not able to find anything.

Please suggest for the same.

Thanks

Neelkanth

[Former Member · Accepted solution]

Hi,

It seems between 17:00 and 20:00 on Friday, the redo log generation was too high and hence the frequency of log switching is also high. It is strange, when no background jobs and dialog session are responsible for this generation. Could you try to find out expensive SQL statements during that period in transaction DB02? Do you generate EWA for this system, you can find this information in EWA report too. I m sure that there are some activities at database that causes this issue.

Regards,

Ganesan

[Former Member · Accepted solution]

Thanks Ganeshan.

Issue has been resolved after analyzing user data.

Transferred data for one APO RFC to ECC was very high compare to other days.

APO user have processed much queues on the day logs got full.