SQL Server detected a logical consistency-based I/O error

Question

SQL Server detected a logical consistency-based I/O error

Viewing 4 posts - 16 through 18 (of 18 total)

You must be logged in to reply to this topic. Login to reply

dbalmf SSCrazy Points: 2063 More actions · Answer 1

I came across this article a while back. It might help

http://www.sqlskills.com/blogs/paul/example-20002005-corrupt-databases-and-some-more-info-on-backup-restore-page-checksums-and-io-errors/

Ta

David

============================================================
David

http://www.sqlservercentral.com/articles/Best+Practices/61537/[/url]

MysteryJimbo SSC-Insane Points: 24203 More actions · Answer 2

GilaMonster (2/26/2014)
kenneth.mofokeng (2/26/2014)
I don't have clean backup. For some reason, my customer stopped weekly backup and the last backup is several month ago
Since you don't have backup , try to repair the database,
TRY
ALTER DATABASE corrupted_db SET SINGLE_USER WITH ROLLBACK IMMEDIATE
DBCC ChECKDB ('corrupted_db', REPAIR_ALLOW_DATA_LOSS)
ALTER DATABASE corrupted_db SET MULTI_USER
NO! NO! NO!
Recommending repairing with no idea of what's actually wrong is insane! We don't even know what the errors are, a sensible recommendation cannot be made on no data. When dealing with corrupt databases the extent of the corruption is essential to know before any useful recommendations can be made. To be honest, since CheckDB's throwing errors I suspect repair will not be a option, it'll likely fail as well.
DO NOT attempt a repair at this point.

I'm most amused at this as I've recently finished part listening, part watching this video of you on the SQLPass channel on youtube 🙂

http://www.youtube.com/watch?v=SQ5Pw1jeWdY

I already knew not to have this knee jerk reaction, its just the timing of it.

Brain2000 Old Hand Points: 364 More actions · Answer 3

I know it's been a few months, so you have probably fixed this one way or another.

I have the situation where I have to keep databases online. Data integrity is second (non-financial), so corrupted records can be compared and fixed when it is convenient. Therefore, I use something called "EMERGENCY" mode.

I would recommend making a copy of the MDF/LDF first, as this technique can cause data loss. In a pinch, it has served me well many times, and I have yet to lose anything too valuable. This is also useful when you can't get a database to even go into single user mode to attempt repairs, or when you can't run backups because it just throws torn page errors.

ALTER DATABASE corrupted_db SET EMERGENCY

ALTER DATABASE corrupted_db SET SINGLE_USER WITH ROLLBACK IMMEDIATE

DBCC ChECKDB ('corrupted_db', REPAIR_ALLOW_DATA_LOSS)

ALTER DATABASE corrupted_db SET MULTI_USER

mxy SSCarpal Tunnel Points: 4036 More actions · Answer 4

I have same errors recently, storage team didn't attach storage properly which caused the problem(Don't know why they are playing with storage in PROD using ISICI).

after the server reboot it is back to normal.