Unsolved
1 Message
0
1323
November 19th, 2019 13:00
Can you force a too many failure disk online
Hello, new to the group and I have tried looking for an answer with out luck. I have a PS6010(single member) with 16 1tb disks in raid 6(out of warranty/contract). I had one failure and it rolled to the spare, and was in stable mode again with a failed disk, while waiting for a replacement, we had a major power outage and when we came back, the PS6010 is showing 2 more failed drives and a disabled drive for "too many failures". the controllers are showing the raid is down for too many failed disks and that they are still initializing(several hours later, no signs of completing), I assume because there is no raid available. but because of this, I can't get any commands to work, I have the su raidtool showing me the status.
My question is, I am in the process of doing drive recovery on the two failed drives(PCB replacement) in hopes they will come online long enough for me to the data off the raid, But is there anything that can be done for the "too many failure" drive? My understanding is this is the controller disabling the drive due to stability, can that be overridden for recovery process? When the controller disables this drive, is it logged in the controller by the firmware or some where on the drive that can be cleared to allow it to come online temporarily? If I replaced the PCB on the drive, would the raid see the data on the drive and bring it online, or would the raid see the original serial number and mark it as offline? All this may hinge on the order in witch they failed, If they all failed at the same time in the power outage, then just bringing 1 online in the raid 6 "should" get me to a degraded stage, but if the 2 failed drives died before the "Too many failure" drive, then I could see the data on the other two being out of date and causing corruption.... In the current "Initializing" state, is there any way to see the order of failure? And before anyone asks, I tried to open a one time case, and they wanted $1000 for 2 hours, and with out even knowing if I could get an answer to these, let alone my data back, that was too much risk for a box that was about to be retired.
Thanks in advance.



dwilliam62
4 Operator
•
1.5K Posts
0
November 19th, 2019 21:00
Hello,
I don't hold much hope that replacing the PCBs will bring the drives back online.
re: Initializing. It will remain in that state as long as the RAIDset is down. It's looking for metadata to complete to boot process.
Re: Support call. It is a risk but they can look at the array and get a better idea of the likely hood of you getting it back online. There are many factors, what model drives, what EQL firmware, etc...
You would probably have better luck sending out the drives to a recovery service. They have many tools that can bypass drive POST failures and allow the drive to be cloned.
Regards,
Don
dr-kiev
1 Rookie
•
13 Posts
0
November 29th, 2019 08:00
Agree with previsue opinion. All disks should be immediatly cloned including broken ones. Raid6 assume double degradation defence, so with all the rest disks without two data recovery still possible . We help with such cases, and we need access to disks or images only in order to virtually reconstruct array and gain access to data.