Anyone seeing drive "failures" TS-509?

[Dennis Wood]

With one of our TS509 units, we're starting to see drives marked as "bad" which degrade RAID status, but end up fine after a length SCAN and RAID5 rebuild. There's a thread over at QNAP with quite a few instances. Just hoping anyone seeing similar can post here with information. This issue may or may not be related to the latest firmware. Here's a copy/paste of the pertinent details:

I've now had 3 errors with read/write errors on a TS509, drives 1, 3, and 5. We operate two TS509 units, the other has 1TB green drives in it with zero issues. On the first drive "failure", thinking the drive was bad, I attempted to insert a larger drive and was unable to do this based on errors from the NAS. This would suggest further volume expansions would be impossible based on current firmware.

Errors did not show up in previous firmware (zero to date until 3.3.0 firmware). Firmware now is 3.3.0
Log shows host resets right after a scheduled SMART scan (quick scan), followed by improper shutdown message in log.

Drives in the misbehaving unit are enterprise drives:

Hard Disk Model
Seagate Barracuda ES 2
Model Number
ST31000340NS

For now, I've disabled all SMART tests on this NAS.

Cheers,
Dennis Wood.[/I]

 

[totenkopf4]

Regarding drive "failures" TS-509?

On My "509" I have had two issues w this unit, the first was upon purchased it was discovered that the drive slides or "shoe" which holds the HD's would not match up w the Seagate Barracuda 7200.11 ST31500341AS 1.5TB mounting holes. This caused the drives to "pop out or not be seen by the QNAP.
This was rectified by QNAP sending 5 new sleds that matched the drive mounting holes. After that it was great - but then -over time - I did have fake hard drive failures, and after replacing w a spare, do the raid rebuild and test the removed drive; which was found to be ok! I returned this drive to my spares inventory. Later this happened again and I repeated the process. But soon after I have two drives fail and was not able to to a raid rebuild - the drives were ok and I purchased another one to add to the mix. All drives are ok, and have the same latest and greatest firmware since being installed in the QNAP.
What I believe is the issue is that the drive connector in the chassis is not "matting firmly w the drive". As I noticed that the amount of force to slide the Seagate drive into the electrical connector varied for each drive. So I used several other drives and found that the HITACHI drives always plugged into the connector w greater insertion resistance VS the Seagate drives.
SO I have switched to the HITACHI Deskstar 7K2000 HDS722020ALA330. In my "509". Now the question is it the over all length of the drives interconnect "fingers" that plug into the female socket on the back-plane of which I speak, or is it that QNAP connectors are now failing due to the number of connection insert / removal process that they go through. As I have learned in my many years of electronics that when you "mate" connectors such as these they can only do it for a certain number of "cycles" before they are considered worn out. I don't know if this helps but
I think somewhere in-between this information is the problem.
Additional info, my current firmware for the 509 is Version 3.2.7 which I believe is two levels below your current version.
I use the box for iSCSI drives only. And have had issues w QNAP iSCSI (lost whole volumes) in some of their older firmware. And have seen similar issues w THECUS as well (iSCSI that is)
Tote