| Author |
Topic |
|
Sarat
Constraint Violating Yak Guru
265 Posts |
Posted - 2003-08-29 : 15:54:39
|
| Hi, I reran the job and now it is successful. The only options I checked was 'Include indexesCheck the data and index pages in the database during the integrity tests. 'But I still what to know why the job failed. If in future this happens, should I ask users to be off the system until I make sure everything is ok?Thanks.Sarat.**To be intoxicated is to feel sophisticated, but not be able to say it.** |
|
|
tkizer
Almighty SQL Goddess
38200 Posts |
Posted - 2003-08-29 : 16:01:29
|
| If it fails in the future, do the things that I suggested in the other thread. If the database is corrupt (DBCC CHECKDB), then yes the users should be off the system until you can fix it. We had a database become corrupt due to a SAN misconfiguration and we didn't even find out that there was a problem until two days later. Users were on the system that entire time. At some point, SQL Server just crashed on us and this is when we were alerted. We do not run the integrity checks everyday, but had we done that we would have known about it sooner.Tara |
 |
|
|
Sarat
Constraint Violating Yak Guru
265 Posts |
Posted - 2003-08-29 : 16:08:26
|
| Hey I did this,--use msdbselect * from sysdbmaintplan_historywhere plan_name = 'AllDbs12PMBak'and activity = 'Check Data and Index Linkage'and succeeded <> 1--=> and it gave me a table name the following error:[Microsoft][ODBC SQL Server Driver][SQL Server]Table error: Object ID 114515887, index ID 1. The previous link (1:2432804) on page (1:2432964) does not match the previous page (1:1795321) that the parent (1:21381), slot 93 expects for this page.--I donot quite understand the message but would the fix be to rebuild/defrag the clustered index on the table?--I really appreciate your input about SAN config...We are about to that shortly!!!--Sarat.**To be intoxicated is to feel sophisticated, but not be able to say it.** |
 |
|
|
tkizer
Almighty SQL Goddess
38200 Posts |
Posted - 2003-08-29 : 16:15:49
|
| Instead of reindex or defrag, I would drop all of the indexes on object 114515887 and then recreate them. But if DBCC CHECKDB isn't reporting any errors anymore, you should be fine already.The SAN configuration thing was a human error. We have multiple SQL Servers using the same SAN. When the server people were doing maintenance on it, they fat fingered one of the drive letters and then one of the SQL Servers was seeing another SQL Server's drives. We log ship our databases to another site (another U.S. state even at our disaster recovery site), but the destination server at this site was down and waiting for the vendor to send us hardware. So we had to restore the database from a previous backup.Tara |
 |
|
|
Sarat
Constraint Violating Yak Guru
265 Posts |
Posted - 2003-08-29 : 16:45:52
|
| I am happy that DBCC CHECKDB was successful this time. But I did nothing to fix the error so how did it get fixed when I ran the command second time? Does SQL Server tries to repair certain things automatically?I do not have 'Attemp to repair minor errors' field checked in the maintainence plan!so what happened that took care of the error?**To be intoxicated is to feel sophisticated, but not be able to say it.** |
 |
|
|
tkizer
Almighty SQL Goddess
38200 Posts |
Posted - 2003-08-29 : 16:48:20
|
| I'm not sure. SQL Server does not repair anything with the DBCC CHECKDB command unless you specify it. Something else must have occurred.Tara |
 |
|
|
Sarat
Constraint Violating Yak Guru
265 Posts |
Posted - 2003-09-02 : 13:15:03
|
| I looked at the event viewer log and error log and the table index which was shown by integrity check job. Nothing gives me indication as to what went wrong with table that caused the integrity job failure.Is it possible that users were trying to do incorrect data entry or weird stuff or some sort of locking? I cannot close this matter until i reach some conclusion.PS: Could this be any related to index job hanging forever on sunday?The table are different though!Thanks, Sarat.**To be intoxicated is to feel sophisticated, but not be able to say it.** |
 |
|
|
tkizer
Almighty SQL Goddess
38200 Posts |
Posted - 2003-09-02 : 13:20:01
|
| Users adding incorrect data is not an integrity issue, it would just be bad data but not corrupt data. If you are unable to close this matter without knowing what caused the problem, then I would suggest calling Microsoft as it is impossible for us to know what happened.Tara |
 |
|
|
|