Drive downshift errors after firmware update to 2.34B

[Mike Green_5]

I have updated the firmware on the 6404 and 6402 smart array controllers from 2.26 to 2.34b. This was done to relieve the false battery failed messages from the bad firmware. It was my understanding that this would also stop the drive downshift message and that is not the case. I am still getting the event log error Event ID: 24683
Description:
SCSI bus fault occurred on Storage Box box 0, , Port 1 of
Array Controller [Exp-Module] in slot 3.
This may result in a "downshift" in transfer rate for one or more hard drives on the bus.

Any ideas on how to fix this one?
Thanks
Mike

[Jim Davis_8]

We have also upgraded to rectify the cache battery issue and are now experiencing the same fault on 2 different servers, both of which are mission critical to us. It has also been noticed that there has been a degredation in I/O performance of our application since the upgrade to 2.34 so the "downshift" message is correct!
We originally thought it may have been hardware related and had drive cage 0 replaced on the ML570 G2 but after a couple of weeks the errors returned.

My feeling is that it is related to the SA firmware as we have only just updated the 2nd server and it is now experiencing the same issue.

Please write more firmware ;-)

Regards

Jim

[Matt Carson]

We received the exact same errors on a DL380G4 in a remote site. The server in question also randomly fails 4 hard disks on the same channel. Each time two drives are failed, the system crashes, but it was coming up OK when the drives were reenabled at the F1/F2 prompts.

HP Tech Support required the latest firmware on the system board and embedded 6i RAID controller (from 2.26 to 2.36). The array controller firmware resolved the battery failure message, but the drives are still failing and we still had downshift errors.

The server crashed again during the second phone call and the system went down during an array rebuild. Now it's an OS reload.

HP Tech Support (Pakistan or India) won't replace parts (main board for embedded 6i) or SCSI backplane until I send someone to the site (3 hour drive) to reseat drives, cables and backplane.

My bet is on a failed 6i controller. Do we have a documented resolution for the previous posts?

Thanks,
Matt

[Jim Davis_8]

We have had some success by applying a workaround that was suggested by HP. It seems that since the 6404 firmware upgrade to version 2.34, a compatability issue with the Insight Diagnostics service is the cause of the "downshift" issue. The workaround is to disable the service and reboot the server. Both of the servers, that were experiencing this issue, now do not have the problem. The disk-disk backups are performing at an expected level and are now consistent. We have not seen another "downshift" error since.

The message from HP is that the Insight Diagnostics application is not critical to the operation of the server and can be left disabled. There may be a newer version of the diagnostics, that will work with the new firmware, released in the near future.

Well done HP....

Jim Davis

[Mike Green_5]

Thanks for the tip - Your message references firmware rev 2.34a and Iâ m running 2.34b do you happen to know if that is still the issue in rev b or not. Also I canâ t seem to find the Insight Diagnostics Service is that the exact name of the service or is it something else?

Thanks again for the help with this
Regards
Mike

[Riesterer]

We also have the same error with the 6400 Controller FW 2.34B.
HP-Support in Germany answers, this worked as designed!! Too many drive commands at the same time, buy a second controller!!! ;-(((
Your workaround with disabling Insight Diagnostics - which service is this exactly?
I don´t find it. Where do I this disable?

Thanks...

[Jim Davis_8]

Gentlemen

The service is "HP Insight Diagnostics". It sits with all the other HP agents in the services list. Mind you, we have loaded PSP7.3 which may meant that the service may be labeled as something else by another PSP version. It may be worth updating to PSP7.3, if you're not there already, to help identify it as well as having an updated system.

Just a suggestion...

Jim

[Jim Davis_8]

Mike

Not sure if the firmware applied is A or B. There did not seem to be anything identifying 2 different versions. I just downloaded 2.34. Maybe it's just an aussie thing. ;-)

Jim

[JOHN SMITH_10]

Hello,

Is the following how you disabled the Insight Diags?

http://h18004.www1.hp.com/support/files/server/us/download/22447.html
(shutdown at the bottom of this page)

tks John

[Jim Davis_8]

Guys

My previous post in relation to disabling the HP Insight Diagnostics service was incorrect to which I appologise. To modify\disable the service you first have to go into the System Management Homepage of the server and select the HP Insight Diagnostics link on the left-hand-side. This takes you to where you can view the config of the system and on the top-right there is a "Schedule" button. This is where the dignostic service is configured to run at a specified time and weekly/monthly interval. Default is 06:00 Wed. We simply unchecked all days scheduled and saved the setting. If you still want the service to run just reschedule it to a more quite time; say 01:00 Sun........

Regards

Jim

[Riesterer]

Hello,

I have deinstalled the "HP INSIGHT DIAGNOSTIC" in ADD/Remove SW, and I always have about once a week the "DOWNSHIFT ERRORS".
Has anybody some news from HP, new FW or something???

Regards
Thanks

[Johnathan]

Guys, I'm having this problem on an ML350G4...re-reading the post I noticed that I didnt disable the service in the right way, I'll try it now.

(Does the new FW version 2.48 solve this problem?)

[Riesterer]

Hi,

I have installed the new FW 2.48 on the SmartArray 6400(DualChannel) two weeks ago, but since the new firmware is installed, i got two times again the downshift errors. Also I have recognised , that 4 of the disks have timeout errors. One of them 28 in the Field: OTHER.
I dont know if this is produced from the Controller , or from the disk himself.
Regards

[Paul Jerome]

I'm getting this error on an ML350 G4p as well.
It's

I'm not seeing an insight diagnostics service. I have a bunch of HP services but none named diagnostics.

The hardware in this case is SmartArray 6400 2ch on firmware 2.62.
Insight management agents v 8.15 and insight array diagnostics U
Utilities v8.10

[inkblot]

has there been any resolution to this error, I'm getting the same on 5 of our servers, four of which are critical.

Thanks

Dale

[Cyberbot]

i have the same issue,
is there is any solution yet?
my RAID control version is 8.26.1.0