5.0.11 > 5.0.12

[Jbrabander]

We've done .9 to .11 to .12 with no issues.

[Bill Brock]

+-------------------------+
| Tables_in_zimbra_logger |
+-------------------------+
| amavis |
| amavis_aggregate |
| config |
| disk_aggregate |
| disk_status |
| mta |
| mta_aggregate |
| processing_history |
| raw_logs |
| service_status |
+-------------------------+
10 rows in set (0.00 sec)

mysql> check table mta;
+-------------------+-------+----------+---------------------------------------- ----------------------+
| Table | Op | Msg_type | Msg_text |
+-------------------+-------+----------+---------------------------------------- ----------------------+
| zimbra_logger.mta | check | warning | Table is marked as crashed and last rep air failed |
| zimbra_logger.mta | check | warning | Size of datafile is: 9937924 Shou ld be: 9658848 |
| zimbra_logger.mta | check | warning | Found 0 deleted space in delete link ch ain. Should be 274300 |
| zimbra_logger.mta | check | error | Found 0 deleted rows in delete link cha in. Should be 724 |
| zimbra_logger.mta | check | error | record delete-link-chain corrupted |
| zimbra_logger.mta | check | error | Corrupt |
+-------------------+-------+----------+---------------------------------------- ----------------------+

OK. Now what? I am searching for the answer. But if anyone wants to jump in with the answer, feel free to do so.

[uxbod]

repair table mta;

[Bill Brock]

mysql> check table mta;
+-------------------+-------+----------+----------+
| Table | Op | Msg_type | Msg_text |
+-------------------+-------+----------+----------+
| zimbra_logger.mta | check | status | OK |
+-------------------+-------+----------+----------+
1 row in set (0.19 sec)

Should I restart my server after repair.?

Thank you uxbod. You da' man!

[uxbod]

I would recommend that you check your other tables aswell; and with respect to a restart it should not be necessary though would not harm.

[Bill Brock]

I will definitely check the rest of my logs.

My question is how the mta log became corrupt. It was truly working right up until I did the upgrade. I even received my admin e-mail that morning before I did the upgrade.

Thanks again uxbod. You were right there when a Zimbra user needed assistance.

[uxbod]

No idea to be honest; only thing I can think of was that a zombie process was holding onto that table. IMHO in future before performing a upgrade I would shutdown ZCS manually and ensure all zimbra processes have closed properly. If not then send a kill -HUP to the PID to allow it to close. Once all have gone then perform the upgrade; obviously once a backup has been performed

No problem with the help it is what we are here for Glad all is working for you Bill.