XCS upgrade on EVA8000

[Rampy(Venu)]

Follow torsten guidelines in this case.It is right.
http://bizsupport2.austin.hp.com/bc/docs/support/SupportManual/c01721447/c01721447.pdf

[Antonio_137]

Hello Rob,
So you think that I did wrong asking for such thing on a public forum beeing an HP employee,well,I don´t see anything wrong about that.
By the way,you seem to be an expert on the matter,so what is you answer to my question...?
"Nemo nascitur doctus",maybe you did....??
Regards

[IBaltay]

Hi,
the XCS 6000 was deactivated on december 2007 for the potential Data Availability Issue When HP StorageWorks 4000/6000/8000 Enterprise Virtual Array Storage Systems Running XCS v6.000 and v6.010 Have Insufficient Free Disk Space:
http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&objectID=c01057608&jumpid=reg_R1002_USEN

The recommendation was to upgrade it immediately...

[IBaltay]

Hi,
HP StorageWorks 4x00/6x00/8x00 Enterprise
Virtual Array updating product software guide
(XCS 6.220)

http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01721447/c01721447.pdf?jumpid=reg_R1002_USEN


HP StorageWorks 4x00/6x00/8x00
Enterprise Virtual Array release notes
(XCS 6.220)

http://h20000.www2.hp.com/bc/docs/support/SupportManual/c01721444/c01721444.pdf?jumpid=reg_R1002_USEN

[AccuMegalith]

Just my worthless input on the XCS 6000 "data availability" issue:

That advisory was worded entirely too mildly. In November, 2007, we were migrating from 72 GB to 300 GB disks in an EVA 4000 with XCS 6000. We even had an HP engineer onsite, who was guiding the process. We migrated as much data as we could off the EVA to local storage, then deallocated half the 72 GB drives. This was going to leave us with about 30 GB, less than one spindle, free. The HP engineer said that was fine.

As soon as the free space dropped below 72 GB, all the LUNs went offline, and the controllers started screaming with various cryptic alarms. We got third-tier support on the phone, and after much hemming and hawing, they told us the firmware bug in XCS 6000 was this: running free space down below one disk's size corrupted all instances of the metadata. If there was only one disk group, as in our (typical) case, there was insufficient redundancy to recover. The only solution to bring the EVA back online was to uninitialize the system and lose all the data.

I spent the whole Thanksgiving weekend reloading the EVA. Good times, good times.