Error 605
-
I received the following error related to a DTS package that runs every night:
DTSStep_DTSExecuteSQLTask_19 DTSRun OnStart: DTSStep_DTSExecuteSQLTask_20 DTSRun OnError: DTSStep_DTSExecuteSQLTask_6, Error = -2147467259 (80004005) Error string: Attempt to fetch logical page (1:13520) in database 'tempdb' belongs to object '-733668897', not to object '#v_ARFact
According to the KB, this error suggests that the #v_ARFact table is corrupt. This is a temporary table and I'm wondering if this is common or to what extent I need to explore the problem. This error occurred 7/31 and has not happened again since. I ran a dbcc checkdb on the related database and found no errors.
Anybody else see this? does this appear to be a transient 605 error?
Thanks
-
This was removed by the editor as SPAM
-
We've been getting these errors at a rate of about one per month for a while now. We believe that the issue is related to our disk controller's firmware. We have a HP MSA1000 Disk controller and the documentation states that:
Under the following conditions with SQL Server 2000, SQL may report use of stale cache data and if all conditions are met, SQL may report error ID’s 605, 644 and 823 when performing subsequent reads from the same SCSI block.
·
Extremely heavy I/O
·
Some % of MSA controller cache allocated to Read
· Multiple small simultaneous writes to same SCSI block
·
Write cache is full at time of request
There is a firmware upgrade available but our suppliers
-
Oops, sorry about the incomplete post...
I'd be really interested to know if your server has an HP MSA1000 disk controller.
We've been getting these errors on our clustered production server at a rate of about one per month for a while now. We believe that the issue is related to our disk controller's firmware. We have an HP MSA1000 Disk controller and the documentation states that:
Under the following conditions with SQL Server 2000, SQL may report use of stale cache data and if all conditions are met, SQL may report error ID’s 605, 644 and 823 when performing subsequent reads from the same SCSI block.
·
Extremely heavy I/O
·
Some % of MSA controller cache allocated to Read
· Multiple small simultaneous writes to same SCSI block
·
Write cache is full at time of request
There is a firmware upgrade available but we are not happy with the risk associated with the upgrade. The kit is all pretty old now and is due to be replaced soon and so we are living with the issue pending replacement.
-
Yes, we are running an
HP MSA1000 disk controller.
-
Well, that will be the problem then. As I mentioned there is a firmware upgrade available which purports to resolve the problem but our suppliers tell us that they have never previously performed an upgrade of the firmware and we have taken the view that we will try and live with the problem pending replacement of our servers.
I would be very interested to hear how you decide to deal with the issue.
Regards
Sean
-
Sean
I am a contractor working for a small company that uses sql server to support its ERP system and some other small applications. I've developed a data warehouse application that relies heavily on stored procedures and temporary tables and variables.
The systems engineer here has not seen this error before and his response was:
"I agree with the statement given about the risk being involved with a firmware upgrade. I would think if we are receiving errors in this job, we would receive them in other jobs or receive other error messages. I don't believe we are."
The only other jobs that run here are backup and database maintenance so I'm not sure that those jobs would gnerate the transient 605 error. The error has only occurred once in the past 3 months so for now I'll take a "wait and see" approach.
thanks for your help
Mardy
-
We had same error before. After upgrading the firmware, it fixed the problem.
ftp://ftp.compaq.com/pub/products/storageworks/techdoc/msa1000/MSA10004.32_ReleaseNotes.pdf
-
Hi XYZ
I'd be very interested to hear your experiences of the upgrade. Who carried it out? How long did it take? What was the downtime? Was it difficult? What precautions did you feel necessary?
Thanks in advance
Sean
-
Hi Sean,
Sorry, I did not directly involve into the firmware upgrading process. I reported the issues to our system administrator; he found some useful documents (see below links) from HP.
Here is the list what he did:
1), download the software cd from HP. (about 40 minutes to download)
2), make a bootable cd out of the iso file and boot the server to it to upgrade the firmware for the MSA 1000 controller, the EMU, the Array configuration utility, and the host bus adapter.
ftp://ftp.compaq.com/pub/products/storageworks/techdoc/msa1000/MSA10004.32_ReleaseNotes.pdf
Detailed upgrade instructions:
ftp://ftp.compaq.com/pub/products/storageworks/whitepapers/upgradeMSAenhancedfeatures.pdf
Down time: about 30 minutes plus two times re-boots.
Good Luck.
-
After several jobs failed again this last weekend (not just the ones I created), a little investigation work disclosed that the 605 error had occurred intermittantly as far back as March of this year. The SA also found that he was behind on firmware upgrades and, after reassurances from HP, he decided to upgrade the firmware last night. A detailed account of his remote upgrade follows:
I successfully flashed the firmware from 2.38 to 4.32. Was a little nervous cause my call and test with HP did not require the server to be reset (so he said), so after the controller was flashed, it never showed back up on Titan...hence all the data was missing from the server.
After slamming 4 Pabst Blue Ribbons, I decided to hold off on finishing my Application to BlockBuster and try rebooting Titan. It came back up fine and I was able to log in to Solomon Corporate, Field and SPD.
I then changed my stained shorts and filed my Application away, in case I needed to finish it in the future after some other maintenance I performed.
See you guys tomorrow.
Hope that helps
Mardy
Viewing 11 posts - 1 through 10 (of 10 total)
You must be logged in to reply to this topic. Login to reply