LogMiner question

Platform: Oracle 9.2 on Solaris 9

I've spent the morning working with LogMiner -- my first real-world application of it. (My only other time was just an initial "what can I do with this" about 18 months ago, which is documented in the ng archives). I just fed it a single archivelog file to try to analyze a sudden massive load of redo (and the resulting archivelog) activity.

Querying v$logmnr_contents with this statement

select (xidusn || '.' || xidslt || '.' || xidsqn) as xid,

        sql_redo as sql_redo
from v$logmnr_contents;

yeilded a fair number of the expected INSERT/UPDATE/DELETE statements, but I've got a whole bushel of the following:

3.14.574     Unsupported
3.14.574     Unsupported
3.14.574     Unsupported
3.14.574     Unsupported
3.14.574     Unsupported
3.14.574     Unsupported
3.14.574     Unsupported
3.14.574     commit;
3.31.574     set transaction read write;
3.31.574     delete from ....

I've actually just given the last few of the "Unsupported" ... there were actually several dozen of them from this xid, and the pattern repeats itself a few thousand times thru the log. If the meaning of 'unsupported' in this context is documented, I've not been able to find it.

Also, I don't know if this is coincidental or if it is actually what I'm trying to pin down, but at the same time this is going on, the application is writing its own error log and reporting a pot load of this:

Jan 13 12:04:32-Thread 1a-Project Framework******/initial: Component NNAVFWFTPServiceServer/******/********/FTPService/Servers/NNAVFWFTPServiceServerModel /NNAVFWFTPServiceModel: SQL exception with error code 17009, SQL state: (null) and message: Closed Statement.

(I've used the asterisks to mask out customer identifying information)

The developer tells me that the app was hammering this error every couple of seconds for a couple of days.

Thoughts?

Thanks.

• Ed Stevens

Cohn's Law: The more time you spend in reporting on what you are doing, the less time you have to do anything. Stability is achieved when you spend all your time doing nothing but reporting on the nothing you are doing. Received on Thu Jan 13 2005 - 14:45:59 CST