Thanks for the question, Andre Vincent.
Asked: September 22, 2016 - 3:00 pm UTC
Last updated: September 22, 2016 - 7:41 pm UTC
Version: 11g
Viewed 1000+ times
You Asked
The likely errors that raises "snapshot too old"s during datapump export are the well documented all over the internet such as :
-undo_retention parameter being too short
-redo logs groups are too small or there are too few
-undo tablespace being too small
Is there a way to monitor those failure cases while the job is running ?
The kind of solution I'm looking for would be a query that would report :
total percentage of the snapshot's undo tablespace usage until it needs to auto-extend or maxes out
total percentage of redo logs usage until it overwrites itself
time left until undo retention is reached by the snapshot
Any query that does part of these things would be welcome.
Does looking up such information requires oracle analytic packages?
Is it at all possible to query such information ?
Is it possible to know for sure before taking an EXPDP that there will be no snapshot too old ?
Looking forward to your answer!
Thanks
-Vincent
and Connor said...
redo log groups will not impact ora-1555. This is an undo size issue.
If you want to guarantee an expdp, you could combine a long retention with the 'guarantee' clause on the undo tablespace.
You can also examine v$undostat to pick up informatin for how long your longest queries run for, and how many ora-1555's you have encountered.
In most versions, ora-1555's are also logged in the alert log with the sql text.
Hope this helps.
Is this answer out of date? If it is, please let us know via a Comment