Mitel SSD drive failure

[v2win]

Has anyone had the SSD drives fail?

I've had two drives for the same customer at different sites bought about a year apart fail recently and was just curious.

[ralph]

What I've seen from the last failure I had is the HD could be reformatted and re-used.
It's sitting in our lab now as a spare.

Ralph

[bobcheese]

So more like corrupted rather than failed then...?

[ralph]

So more like corrupted rather than failed then...?

That would be my guess.   But the last one I had appeared to get corrupted while running.   It stopped responding to ESM, Telnet, RTC, and phones wouldn't boot etc.

I'd be interested in knowing if anyone else is experiencing this.

Ralph

[marcolive]

Got one too.  This was a 32gb SSD in a MXeIII. 

[bobcheese]

Macrolive - what where the symptoms? We had one go the other day after an unexpected reboot. Also what version of sw?

[darrick]

Hey just thought I would post my 2 cents about the SSD drives. First time poster as well.

We've had 3 or 4 of the Solid state drives fail on us as well in the MXE3's. For the most part the 3300's have reset (power outage, customer reset on 1) and they just never come back up. They look like they load, but then they just get real slow loading, and after awhile reboot themselves. (plugged into the serial port in front and watching the items load) They all seem corrupted - Back at the shop we can do a base install and use the FixPartition flag on the bootloader and they work great. We have sent the other back to Mitel and they have sent back replacements. We have a few more out there that have never had a problem at all.

Us tech's tend to stick with the drives with moving parts for now.
Darrick

[Mitel100]

Some Solid State Drives manufactured by Renice Technology prior to September 1st, 2011 may contain resistor packs with poor quality  solder joints, which cause drive initialization failure or intermittent runtime failure.

We have had a few of these currently. I would report these back to Mitel

[bobcheese]

Hey Darrick, same with us same system failed to reboot after 2 individual 'unplanned' power downs. A fresh install and the disk works fine again.

The reason I ask re SW load is that these are on 5.0SP1PR1 and we also had a SATA disk 3300CXi which did not come up after a reboot the other day. Then a second power down and it was fine. Just wondered if it may be a SW issue maybe on loading bootfile

[marcolive]

Hi all,

This system was on MCD 5.0 SP1 and it hopefully, it didn't reboot. The first symptoms were that users were unable to listen to their voicemail messages.  When users called EMEM hunt group, the port answered and than hung up immediately. Some time after that, their main auto-attendant (also from EMEM) had the same behavior.

Customer called us. I restarted their EMEM (iPVM_Stop and iPVM_Start in VxWorks). Everything worked fine after that!

Looked at EMEM logs (FTP to /vmail/c/voxdrv and downloaded diag.dat & diag.sav) and I saw many occurrences of this line :

vtxd40: Dos Error Encountered DOS code=21 cstate=5

Mitel Diagnostics Guide tells that this is basically a read/write error from disk. From that, I did backup of the system (including voicemail).

The next day, same problem. This time, I had a "VTG_VM_ERROR" major alarm. Tried to stop voicemail. Unable to start it again. FTP to /vmail partition showed that it was totally empty. Also, I had many occurrences of this in MCD logs :

Source : Io2k
Description : Unexpected Result->io2kFileSeek() - Error in lseek. Params:  715 2945280 0 , Error: 0x3D0004 FileName =

Hopefully, the MXe didn't restart by itself.  Users were able to make/receive calls without problems. I have not tried to backup the system in this state. I changed their SSD to a 160gb SATA hard drive after their working hours. The controller was unable to boot with the old SSD.

Ralph & darrick are right. I was able to use this SSD again with FixPartition flag.

That's a very strange problem. Looks like data gets permanently corrupted while the system is running. We have a lot of 8gb SSDs in CX(i)s and never had problems.

This SSD was a "Renice" received in in January 2012. I think Mitel100 is on the right track!

[x-man]

"Mitel Diagnostics Guide tells that this is basically a read/write error from disk. From that, I did backup of the system (including voicemail). "

This mitel diagnostic guide...is it available on MoL? It could be a very useful thing to have .

Thanks

Tony

[marcolive]

I'm sorry! Mitel Troubleshooting Guide. It's available on http://edocs.mitel.com

[ralph]

Bumping this thread....
Just found out that we received a batch of drives that are known bad.   They will run for a while then fail on boot.
I have ~40 systems benched, programmed and ready to go with these drives.
Getting replacements today.   It's gonna be fun swapping them.

Ralph

[PFloyd]

We've had 2 go out in one 3300, once likely due to a brownout (the 3300 was being prepped for another location and was not on a UPS) and another time after a seemingly clean shutdown prior to shipment to the remote location.

We had another experience some partial corruption after a bad UPS did not appear to keep the 3300 up through another brownout, but the drive was never replaced. We did had to restore its configuration, though. Not sure what else was done.

[bobcheese]

Hey Ralph, we had one replaced by Mitel after this fault occurred and they admitted it was a bad batch. Then the replacement did the same thing!!! Don't want to teach you to suck eggs but make sure you yank the power a good few times before you send them out. We are now at the point were we don't see how we can replace with another SSD and may have to go back to RAID on this one.