EVA 4000 - cache mirroring

[Ezio CERRATO]

Hi, on my command view eva I can see only CONTROLLER B with a warning sign. Under cache mirroring, Mirror path 1 and mirror path 2 are reported NOT ACTIVE.

What can I do to reestabilish the situation and view my other controller?

Thanks

Ezio

[Uwe Zessin]

Find out what is wrong with the other controller. Maybe it has failed.

[Ezio CERRATO]

Good call, the controller that is not working is displaying:

FAULT ERROR

TC: 0403047F IDX: LF
PARAM 00: 0005A2F8
01: 020D0100

Any clues?

[Uwe Zessin]

Is this a system with older firmware? I see this code mentioned in some notes with improperly handled situations.

There is not much you can do right now except for restarting the controller and opening a call with HP.

[kumardsm]

---

@Uwe Zessin wrote:
Is this a system with older firmware? I see this code mentioned in some notes with improperly handled situations.

There is not much you can do right now except for restarting the controller and opening a call with HP.

---

Hi, We too having same problem, in our situvation even we are not getting any display in OCP (Operator Control Panel). What may be the probelm, please let us know, because of this we are not able switch on any VMs. Thanks in Adance,

Kumar

[Sebastian.Koehler]

If the OCP is dark and stays dark if a button is pressed, the controller is most likely dead. Check the LEDs on the back to be sure. When a reboot does not fix the issue. Call support to have it replaced. If the controller boots up and passes all TEs, open a call to have the cause checked.

Regards,
Sebastian

[kumardsm]

Hi Sabastian,

Thank you very much for your kind support, yes, even after restart also OCP is not displaying, find the below Controller backside LED indicators

Port - Controller A - Controller B

Inter Controller - ON - ON

CAB - ON - ON

DP1B - ON - ON

MP1 - OFF - OFF

FP1 - Amber - ON

FP2 - OFF - ON

MP2 - OFF - Amber

DP1A - OFF - ON

UID - OFF - OFF

Please let me know, if you understand anything from this information.

Regards,

Kumar

[Sebastian.Koehler]

LEDs alone will not help. Remove power from controller and wait about 2min before reattaching. If it does not go trough TEs and does not display anything in the OCP, you should have it replaced.

The only thing you might try is a standalone bootup of the suspect controller.

1. Remove power from controller
2. Detach all cables including the interlink from the controller
3. Attach power and boot it up / see if anything changes

Doing so could help identify suspect cables. Usually the TEs will report faulty components.

Regards,
Sebastian

[kumardsm]

Hi Sebastian,

Thank you very much for your continued support. Yes, we too feel, there may be Controller failure. Currently all our VM's are working in another Controller and I will try your suggestions on Sunday and update you.

But now our backup is running very slow, is this because of Controller failure or low free space in VM server or any other issues you suspect. Plz....

Thanks & Regards,

Kumar

---

@Sebastian.Koehler wrote:
LEDs alone will not help. Remove power from controller and wait about 2min before reattaching. If it does not go trough TEs and does not display anything in the OCP, you should have it replaced.

The only thing you might try is a standalone bootup of the suspect controller.

1. Remove power from controller
2. Detach all cables including the interlink from the controller
3. Attach power and boot it up / see if anything changes

Doing so could help identify suspect cables. Usually the TEs will report faulty components.

Regards,
Sebastian

---

[Thomas Callahan]

It's running slow because your one remaining controller now has the handle the load which was split amongst the two controllers. This is also most likely due to the fact that your write caching is now disabled, resulting in write through(writeback) mode, whereby all writes have to go directly to disk, instead of into cache, then into disk.

Open a call with HP immediately, and I would not recommend messing with cabling unless you know exactly what you are doing. One wrong move and you could take out the surviving controller, resulting in a full outage of storage.

Continuing to mess around with the failed controller could cause more problems than you want to even consider. Call the controller dead, and get HP involved.