cancel
Showing results for 
Search instead for 
Did you mean: 

Archive stuck

Former Member
0 Kudos

Hi,

Our stage system faced an almost archive stuck situation where the 'oraarch' directory was 99% full. I moved some of the redo log files to a different location and re-triggered the BRARCHIVE job which completed successfully (copy,delete,save). Now, I put back the redo log files which were moved from the oraarch directory but these files are not touched by BRARCHIVE job. I understand that the redo log files are incremental in nature which means that my latest redo log files contain all the changes. Are these files not being cleared by BRARCHIVE because of a clash in the sequence numbers? or is there any other means to force BRARCHIVE to process these files ?

Cheers!!

Aj

Accepted Solutions (0)

Answers (1)

Answers (1)

anindya_bose
Active Contributor
0 Kudos

You should have moved those files which generated last and then restore them back before BRARCHIVE take them.

Now, you can have one fresh backup. Otherwise if you need to restore you will be needing those redolog files.

But as you have those files in your server you can manually restore them before recovering incase of disaster.

Please shcedule a fresh backup.

Redo logs are not increamental.

Former Member
0 Kudos

Hi Anindya,

Thanks for your time. As per your explanation, let me cite an example.

Lets say, we have redo log files 20 to 30 in the oraarch directory at the time of an archive stuck situation. In this case, I should have moved the files, like 26 - 30 to a temporary location.

- Run BRARCHIVE (This processes the redo log files 20 - 25)

- Restore the file 26 - 30 to oraarch which will now have the redo logs 31 - ...(The new redo log files)

-Run BRARCHIVE

-This time it will include 26 - 30 as the sequence would be correct.

Pls correct me if I am wrong.

Thanks for correcting me on Redo log files not being incremental. I see why, because incase of a recovery we would need all the redo log files. If it is incremental, then the latest redo log would be sufficient for the recovery which is actually not the case.

Cheers!!

Aj

anindya_bose
Active Contributor
0 Kudos

Explanation seems fine.

You can prevent this in future by keeping a dummy file or folder of 50-100 mb size in the oraarch directory. In case you face a archive stuck, you delete the dummy folder not the redologs. https://service.sap.com/sap/support/notes/391

You might be interested in the below Link

Link:[archive stuck |http://help.sap.com/saphelp_nw70/helpdata/EN/e3/92fdd1fada9e4d87be08c217639c1f/content.htm]

Regards

Anindya

Edited by: Anindya Bose on Nov 10, 2009 1:33 PM