on 12-12-2007 6:24 PM
Hello,
One more question pl.
"Checkpoint not Complete" - Can an archiver stuck situation cause this condition? How about Block Corruption?
Thanks.
Regards,
Tuhsar
no,
what causes this: there is too much activity and Oracle does not have enough time to copy all online redologs to offline redologs because those online redologs fill to quickly.
possible solutions:
add mirrored online redologs
increase the size of existing online redologs.
... or both.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello Eric,
your solution is correct ... but your explanation is wrong.
>> there is too much activity and Oracle does not have enough time to copy all online redologs to offline redologs because those online redologs fill to quickly.
The error "checkpoint not complete" occurs if too many checkpoints are created in a short time period and the DBWR process has not written down all data to the datafiles which are located in the first redologgroup. Checkpoints are created, if a logswitch occurs.
There are some solutions to solve the problem:
- Add more redolog groups
- Increase the redolog size of the groups
- Increase the performance of the DBWR (maybe some more DBWR processes)
>> How about Block Corruption?
Block Corruption in case of an "checkpoint not complete" - no way, there should be no block corruption.
Your database stuck until the first redolog group can be overwritten (data is checkpointed and written down to the datafiles).
Regards
Stefan
User | Count |
---|---|
78 | |
9 | |
8 | |
6 | |
6 | |
6 | |
6 | |
6 | |
5 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.