EVA8000 problem in replacing new drive

[mahdieh]

Hi,
We have a problem in our EVA8000. When I changed the failed drive with a new one, Command view shows the below error in the controller event log.

An HSV210 controller has failed in communicating with the Cabinet (Rack) Bus Interface Controller.

The Drive Enclosure Environmental Monitoring Unit is able to communicate with a physical disk drive but this HSV210 controller is unable to communicate with that physical disk drive on the Fibre Channel bus.

The disk act strangely as its LEDs is green for 30 seconds and orange for 30 seconds continuously. I changed the place of drive. the LED become green in the new bay but the status of the drive is as follow:
The disk drive in this bay is either failed or not mated properly to the bay.

I should mention that 2 weeks ago, we had 4 failed disks in 1 week.

do you think it is an EVA problem, or the new disk's problem.

Regards

[Torsten.]

Since the EMU can talk to the drive but the controller cannot, it looks like an eva problem. Ask hp support to have a look at your logs.

[Víctor Cespón]

Without looking at the event log and why and how each disk failed, we cannot tell.

This disks looks like a DOA, does not communicate with the controllers.

[Matt Palmer_2]

Hi, were all disks that failed in the same shelf or slot? If so there could be a shelf or port issue in that bay for the specified shelf.

regards

Matt

[Daniel_279]

I see XCS code is NOT mentioned - chances are you have this issue/problem below.

And this can cause disks to fail prematurely - as stated EVA controller logs would need to be reviewed

http://bizsupport1.austin.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&taskId=110&prodSeriesId=3664583&prodTypeId=12169&prodSeriesId=3664583&objectID=c01914371

Search on the following if link above has issues


SUPPORT COMMUNICATION - CUSTOMER ADVISORY

Document ID: c01914371

Version: 2
ADVISORY (REVISED): HP StorageWorks 4000/6000/8000 and 4100/6100/8100 Enterprise Virtual Array EMU communication events are sometimes reported following upgrade to XCS 6.220

[mahdieh]

The failed disk are located in different enclosures. I attach the related log files.

[Jan Soska]

hello, we had similar issue with Eva6000 and xcs6220 firmware. If I remember it correctly, we fixed it with reinstalling disk firmware (the same version, no new one available) for each problematic disk. But we think the real solution was that during firmware "upgrade" was disk disconected for a while and reconnected again.

Jan

[Víctor Cespón]

Firmware is 5110, very old, consider updating it to 6220 ASAP

0-10-08 started having errors on 2010-jul-02 07:08:37 and failed on 2010-jul-02 07:09:20
0-06-09 On 2010-jul-03 16:20:51
0-05-10 On 2010-jul-06 10:13:36
0-04-02 On 2010-jul-08 16:27:57

Up to here acceptable, disks had errors like:
ASC: 0x18 ASCQ: 0x05 (Recovered data - recommend reassignment);
ASC: 0x11 ASCQ: 0x00 (Unrecovered read error);
ASC: 0x32 ASCQ: 0x00 (No defect spare location available);

But after replacing 0-10-08, several disks stopped responding to controllers. And they are on different loops.

2010-jul-19 16:55:59 0-10-08 not responding
2010-jul-20 08:45:15 0-09-14 not responding
2010-jul-20 09:44:14 0-05-10 not responding
2010-jul-20 09:52:06 0-03-14 not responding

I would recommend to update the firmware and power-cycle the EVA.

[Armin Norouzi]

vcespon,

You mean firmware bugs may harm disk drives physically?

thanks

[Uwe Zessin]

No, but too sensitive S.M.A.R.T parameters can declare a drive bad even if the errors are not _that_ bad.