bdump folder filling up very quickly with redo log thread dumps
I can't seem to track this one down via google so i figured I would check with you guys
It seems my entire redo log and its thread dump are being written to trace files in the bdump. These files are at least 200 mb (so hard to just open in text editor and view what is happening). The files I am actually able to open as notepad++ can't view extremely large files, the only ORA I see is this:
ORA-00600: internal error code, arguments: , , , , , , , 
Anyone seen this before?
I haven't figured out why it is happening or even when it seems to happen. There are very few database transactions that occur on this instance, and none of them seem to specifically be causing the redo log to dump to a .trc file.
Issue started happening with a Physical to Virtual change and at that same time the host.domain changed as well (yeah for being acquired!). Prior to that there were no issues with this db.
ora-600 messages need to go to support, so raise a SR with oracle
I wanted to avoid that if someone else had ran into the same thing
However, I solved the issue: It seems the cleanup job that runs hourly and delete's from MGMT_METRICS_RAW was causing the ora-00600 (most likely due to the p2v or hostname change). I had to recreate the emctl repo using emca, and deconfig/reconfigure the dbcontrol. Once I had done that the thread stopped dumping every hour.
Hopefully this helps someone else that has to go through 2 levels of management to raise an SR with oracle.
Click Here to Expand Forum to Full Width