HSG80 Warning seen

[Anwar uddin]

Dear All,

I am having a warning in event viewier since 2 weeks the warning is
Data was recovered using error correction code on device \Device\Harddisk2.

We have HP storage 2200 with HSG 80 Controller. I have know idea why this warning is coming up. i checked the enclosure to see if any hard disk is failed but i dont see any thing this only in the case of one server which is running exchange 2003 cluster and yesterday the cluster lost connection with hard disks and the cluster service failed. once i restarted The controller and SAN switch and enclosure it started working fine no idea why this has happened.

Can any one let me know any software which tells me what has might caused the error?

Secondly the firmware of the controller is 8.7 and i want to download newer ver thta is 8.8 for HSG80 controller. but i could not find one to download.

Thirdly what is making to throw those warnings.

I really want this storage to run for 3 more months as we are waiting for EVA 8000 to be delivered and installed so please let me know what to do to get ove this error. The warning is only seen in one cluster not others who have the data on storage. when the exchange cluster was down all other servers who were getting data from storage weren't effected and they were able to get there data without any problem. this problem of loosing the connectivity was with exchange cluster only i checked the HBA, Fiber cable nothing seems to be problematic though

Thanks

[Uwe Zessin]

The warning means that a disk has developed a bad spot on the media, but its built-in redundancy was able to recover the data. You need to find out which disk drive is affected and better replace that part.

[Anwar uddin]

Thanks for the information. How would i know whichhard disk is damaged?? secondly i was looking out for firm ware now where could i find the one. Thirdly how would i know what made the exchange virtual drives fail how would know the reason for that?

Thanks for reply

[Martin Smoral]

1) run vtdpy from the console serial port and do a show device or devices. Look at the error column to find the disk getting errors
2) download the new FW from here
http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareIndex.jsp?lang=en&cc=us&prodNameId=433897&prodTypeId=18964&prodSeriesId=433888&swLang=13&taskId=135&swEnvOID=1005
3) Not sure what agent reports the errors to the event log.
regards,
Martin

[Anwar uddin]

Hi,

Thanks for the reply i did run vtdpy but i did not find the column called errors and secondly when i am already running vtdpy from console i cannot run any other commands like show device command.

I am sorry but i am kind of new for storage. So please be patient with me.

Thanks again

[Anwar uddin]

Hi once again

I found how to find the problem in my controller. I did captured the data and i am attaching both logs of two controller.

i can see the logs but i dont know what does this report or log means...
can some one tell me if there is some thing wrong with controller i did find out all the events since 2004 which are shown below. I guess i am already asking to much but i am kind of worried. Thanks for all you who are of my help

Please check the attached text file which gives detail about the events.

[BR739321]

The attached files are 'SHOW LAST ALL' output from FMU on the HSG80 controller pair and so don't provide any insight into disk errors. The SHOW LAST ALL output purely gives information on any crashes / restarts (including intentional ones).
If you're running Windows and you have the correct solution software installed then what you should see in the Application Event Log are HS_Service events. These can be either manually decoded (not easy unless you know what you're looking for) or processed using WSEA (part of the WEBES suite of tools available from HP) which will provide a bit-to-text translation of the HS_Service events. You'll then be able to identify the exact physical disk thats reporting the errors and make a decision as to what action is required.
If you are unable to review the HS_Service events then the other option is to leave a console permanently connected to the controllers maintenance port and capture the output, you'll see the errors scroll through on the console (this is the information that is then reported to the Hosts Error Log). The errors unless captured will be lost (there's no history stored on the controller on the version of ACS you are running).
With version 8.8 of ACS there are additional commands which allow you to see a history of device errors. As mentioned though, with the version of ACS you are running this functionality is not available.
From the FMU output supplied there is no indication of an issue on the HSG80's themselves (other than that you are running an unsupported version of ACS - V87F-0). The last restart was due to a powerfail interrupt on 22-FEB, you need to ensure that this wasn't caused by the HSG's being simply powered off without being shutdown cleanly as this could cause you problems.
If you have the HS_Service events then your service provider should be able to assist you with decoding these.