2 raid disk failures in 2 months

bigred8080

Honorable
Mar 13, 2012
67
0
10,630
I just checked the status of my raid and it was critical. Seems that the HDD in slot 2 is unconfigured bad. the last time it was slot 3.. WTH man. is this common? anyone have any ideas on what could cause this?

I have an MSI raid controller running raid 5 with 6 Seagate 3tb drives. I guess ill order another and hope nothing happens between now and then.

Could heat be an issue? The box is in a well ventilated area (utility room that's attached to the kitchen, on a 2 1/2' stand) I do notice the top of the box is warm at times, and I hear the CPU cooler kick into high gear, I just thought this was normal.

Please shower me with your thoughts lol

thanks
BR
 
Solution
I would be hesitant to place much weight on that information. Those statistics aren't very fleshed out. That pie graph doesnt state what it actually shows. It implies (I think) percentage of Total Hard Drive RMA's, in which case I would say "duh". WD and Seagate are the biggest companies in the HDD space and dominate the majority of the market, of course they are going to have the most RMA's simply because they have the most drives.

Also mixing and matching HDD's in a RAID array is exactly what I would recommend you don't do. Your storage array performance is determined by the weakest link, and introducing variables by using multiple HDD models and brands will lower your performance compared to using the same model.
I would be hesitant to place much weight on that information. Those statistics aren't very fleshed out. That pie graph doesnt state what it actually shows. It implies (I think) percentage of Total Hard Drive RMA's, in which case I would say "duh". WD and Seagate are the biggest companies in the HDD space and dominate the majority of the market, of course they are going to have the most RMA's simply because they have the most drives.

Also mixing and matching HDD's in a RAID array is exactly what I would recommend you don't do. Your storage array performance is determined by the weakest link, and introducing variables by using multiple HDD models and brands will lower your performance compared to using the same model.
 
Solution
My point was the graph isnt clear as to what it was representing. Percentages can still be influenced by sheer quantity, as I said if its percentage of total RMA's (how you would know about dead drives) then of course the biggest sellers are going to be higher up on dead drives.
Also where is this info coming from, and who is this guy telling us? Just saying the info may not be reliable.



I don't see how that makes sense.
 

rehanzaidi

Honorable
Sep 15, 2013
3
0
10,510
What is the server brand? have you checked that if the firmware is up to date? Secondly, are you checking only in the event viewer or have you tried the HP management console (if the server is from HP)?
I had the same issue that windows 2008 R2 was not showing any error but physical LEDs were showing that 2 drives are failed. But of course, in RAID5 if 2 drives are failed then its impossible that server will work but mine was working. I updated the firmware & after that every warning/error wasnt appearing anymore.

 

bigred8080

Honorable
Mar 13, 2012
67
0
10,630


Custom built i guess you would call it. And I am just going by I guess what you are calling the event viewer. My raid management software.

Well I guess I'll just replace it with the same and go with it. They are on sale for like 120 anyways, so it's not to bad