Regarding critical bugs and table compression, take a look at the following note from MetaLink. What I like most is the recommended "workaround":
Bug 7123643 Wrong results (NULL) from select on compressed table
It is believed to be a regression in default behaviour thus:
Regression introduced in 10.2.0.4
Regression introduced in 11.1.0.6
Symptoms:
* Wrong Results
Related To:
* Compressed Data Storage
Description
This problem is introduced in 10.2.0.4 and 11.1.0.6 by the
fix for bug 5912350.
SELECT against a compressed table can return NULL data
for columns which contain data in certain scenarios.
This problem only affects compressed tables where the data has
been loaded using bulk load (CTAS, import, Insert as select etc...).
There is no persistent corruption.
Workaround:
Do not use compressed tables
By the way, as mentioned in the description, this bug was introduced by the fix for the bug
5912350: Dump (kdr9ir2f0rst4srp0) on select of compressed table having > 255 columns
Symptoms:
* Process May Dump (ORA-7445) / Abend / Abort
* Dump in or under kdr9ir2F0rst4srp0
Related To:
* Compressed Data Storage
Description
A dump can occur when selecting from a COMPRESSED
table with > 255 columns.
Workaround:
Do not use compressed tables with more than 255 columns.
Note:
This fix introduces the problem described in bug 7123643.
One off fixes for this bug (5912350) should be deinstalled and
replaced with the fix for bug 7123643.
So it looks like using table compression is still a bit risky. I hoped that it is more mature now that this feature has been there for quite a long time.
Partager