EVA5000 replaced failed disk, but cannot remove the failed disk from the disk group's list

[Paul Henderson_2]

One of the disks in my EVA5000 failed. It is called "Disk 012" in Command View EVA. This disk was physically replaced, and the new disk showed in Command View EVA as "Disk 058". Fine. Everything leveled off after a while, and the system is now running normally.

But "Disk 012" still appears in the disk list under my disk group in the Command View EVA list. If I click on it, I get a return saying "Operation Failed!".

I'd like to delete this disk from the disk group so that I can get rid of all the warning symbols in the Command View EVA display, but cannot figure out how to delete it. I've searched the online help, but cannot find any answer.

[SAKET_5]

Hi Paul,

My understanding of EVAs says that the problem you are faced with is likely to occur if between disk removal and insertion, you have disregarded the "Disk Surface Scan Interval" which is usually 1 minute. Did you actually wait for 1 minute between when you took out the failed disk and when you inserted the replacement one in?

The easiest approach that I would take is to reboot the SMA. This should not cause any major services downtime as the SMA is there for storage management, except you are running BC scheduled jobs, etc via it.

Let us know how you go with this approach, i.e. if you decide to take it.

Regards,
Saket.

[Paul Henderson_2]

HP Services replaced the disk, and I observed that they did wait the requisite 1 minute (until the beeping stopped) before inserting the new disk.

Per your instructions, I did a 'restart' of the SMA from its main page's Maintenance section, and the failed disk disappeared!

I guess this follows the old maxim... in the Windows world (the SMA runs windows), you need to reboot after every operation ;-).

[Mark Grossman]

Paul,
we have ahd similar experiences on our eva3000. Now, we not only wait the 1 minute but we make sure to click on the gui until it is refreshed correctly before proceeding.
Gotta luv windows.

Mark

[Uwe Zessin]

Yes, that's the general suggestion, but let's be fair: it is a problem of CV-EVA, not Windows.

In most cases it is sufficient to restart the agent on the SMA (Settings -> Manage tools, stop and then restart CV-EVA).

[Mark Grossman]

uwe,
yes i agree - but it is more fun to bash windows. The SMA is very delicate, although rebooting usually cleans it up.
Mark

[Uwe Zessin]

Well, the SMA goes end-of-life this month, but the fun will continue ;-)

[Edgar_8]

Hi Uwe,

Can you elaborate on your last comment that the SMA is discontinued?

Thanks in advance!

[Mark Poeschl_2]

Paul: The correct procedure for removing a failed disk is to:

1) "Ungroup" the bad disk from whatever disk group it's in.
2) Replace the disk and wait the requisite 1 minute.
3) You might want to wait for the original disk group to finish re-leveling after the removal in step 1, but I don't think the failure event might take care of that for you. In any case you could look at the disk group status to tell for sure.
4) There should now be a new "ungrouped" disk that you can add to the original disk group - and again there will be a re-leveling operation.

The problem came about when the bad disk wasn't ungrouped before replacing it. It may well be that a CV restart or SMA reboot will clear things up.

Edgar:
The turnkey SMA solution is being discontinued. CV-EVA is still the required software package for managing EVA, but HP is now saying it can be installed on any standard Windows host that meets certain minimum requirements.

[SAKET_5]

In Mark's response:

1) "Ungroup" the bad disk from whatever disk group it's in.
------
This is usually taken care by the VCS when the disk fails - the failed disk gets kicked out of its orginal disk group and lands into the "Ungrouped Disks" folder. Yes, if it is a case of pre-dictive failure you might be proactively ungrouping the disk in question which will first perform a data migration of the data contained on the failing disk to the rest of the physical disks in the disk group. This will then be followed by a disk group data levelling.
------
2) Replace the disk and wait the requisite 1 minute.
-----
I think what Mark meant to say is that remove the disk and wait the amount of time specified for EVA "Disk replacement delay", (which has a default value of 1 minute) and then replace the failed disk with a new one. This is not to say that you can't change the default disk replacement delay value to something that you fancy!...not that I can think of any reason why you would want to modify this value...!

[Mike Reznak]

Hi,
I've experienced the same problem and only the reboot of appliance helped me.
I've tried everything else, what was mentioned here. But it was on older version of VCS (think 3.0)


Mike

[Ivan Ferreira]

I had that problem too and restarting the command view eva solved the problem.

[JPS-LTU]

Hi there,

weve had this problem too on a number of occasions. Restarting command View EVA should fix the problem for you... open a browser window on the SMA and select Settings -> Manage tools. Stop and restart cv - eva.

We have just upgraded to version 4.0.1 so dont know if has been fixed in that release

cheers
jason