Viewing 15 posts - 481 through 495 (of 994 total)
ok - you need to hack the database back into the server. See this blog post: TechEd Demo: Creating, detaching, re-attaching, and fixing a suspect database
April 6, 2009 at 2:08 pm
Yeah, the log file's corrupt - one guess would be a drive wasn't honoring write-through requests to the log file. Anyway, without going back to your backup, you'll need to...
April 6, 2009 at 1:31 pm
It's possible for a page to be corrupted, by a few bits being flipped here and there, that doesn't affect the structural integrity of the records, page, or index/table the...
April 6, 2009 at 8:52 am
Gail (and others) - once a database is upgraded and checksums are enabled, torn-page protection is still in place until a page is subsequently changed - see Inside The Storage...
April 3, 2009 at 3:23 pm
It would help if you explained what the trace flag does... 818 enables some code which could help with the detection of stale reads. You could also try trace flag...
April 3, 2009 at 3:12 pm
This is definitely 2000, from the error messages. 2005 errors from CHECKCATALOG are completely different.
Did you create a user-defined type, and then someone mess with the system tables directly?
April 3, 2009 at 3:08 pm
This might happen when a CHECKDB-created snapshot fails to be removed if something goes wrong while CHECKDB is running. Doesn't happen very often - but when it does the hidden-snapshot...
April 2, 2009 at 9:49 am
Your corruptions are 'disappearing' because the rest of your nightly maintenance is doing stuff like rebuilding indexes (i'd guess) that are deallocating the corrupt pages, and moving the indexes to...
April 2, 2009 at 7:49 am
And programming is irrelevant - you can't do anything in SQL Server that will cause corruption - see Search Engine Q&A #26: Myths around causing corruption
March 28, 2009 at 4:16 pm
Yes and no. Yes, in that tempdb will be recreated and may use a different part of the drives hosted by the IO subsystem. No, because your IO subsystem still...
March 27, 2009 at 3:54 pm
It's a tempdb page. You could try running CHECKDB on tempdb, but the checks are somewhat limited on tempdb, and it's likely the page isn't allocated any more and so...
March 27, 2009 at 3:47 pm
This looks just like I/O subsystem corruption - the page is full of garbage.
I'd like you to turn on page checksums, which will catch when the I/O subsystem is causing...
March 26, 2009 at 12:46 pm
No idea how to tell what the physical address of the pages are on the hard disk - all I can say is that the offset of a page within...
March 25, 2009 at 2:08 pm
March 25, 2009 at 10:13 am
Just be very careful when repairing msdb - and make sure you know what repair deleted (remember that repair is basically "delete what's broken and fix up all the links")....
March 25, 2009 at 9:59 am
Viewing 15 posts - 481 through 495 (of 994 total)