Hi Folks,
I have a HD-H1.0TGL/R5 running the 2.16 firmware. It has been running without issue for several years.
I use Acronis True Image to do system back ups from my desktop to the Terastation. A few months ago I began to see file I/O errors in the Acronis screen. I would hit re-try and everything would back up without further issue.
About a week ago, I saw a red LED on disk one. A quick look at the GUI confirmed a failed disk. I initially tried SATA disks with a PATA/ SATA adaptor but these would go into a perpetual disk check loop, reach about 30%, then start again. It never completed.
I figured that it had to be an issue of the TS not liking the drive so I found the exact WD2500BB drive and installed that. According to the manual, the TS should just rebuild the RAID 5 array and go back to normal. Not the case. First the unit would not boot. After flashing the software, I was able to access the GUI and formatted the new disk to the XFS format. Still no go. I next deleted the RAID 5 array and re-created it. The TS went into disk check and went to about ~30% then locked up.
As an interesting side note, no share folder was created. I see the info folder. When I tried to manually create one, the screen said "Committing changes to memory" but no folder was created. I have since tried in RAID 1 and spanning mode with similar results.
Am I just missing something really basic or do I now own a really big paperweight? Any suggestions are appreciated.
garrman
Edit: I did a "Reset to factory defaults" and was able to set up a share. One step closer.
I next set up a RAID 5 array. The TS went into "Checking RAID array" right away. Once again, it made it to about 35-40% before it locked up and went off the network. What I found by trial and error was that if I set up in Spanning mode, the TS did not need to check the array.
So why will it set up in Spanning but not in RAID1 or RAID 5.
Jan 18, 2012 Update.
So I finally got the Terastation (HD-1.0TGL/R5) back into RAID 5. As far as I can tell, this is what worked.
1) Flashed to 2.16 firmware w/ telnet.
2) Deleted "old" array that never finished "Checking RAID". Note: during this checking raid, process raid5d would run at ~60% cpu usage, process raid5syncd would run at ~8% cpu usage.
3) Went to "Maintenance" tab and selected Initialization, than selected restore defaults.
4) Powered down Terastation by switch in back of unit.
5) Held down INT button for 30 seconds with no power. Hit power button, continued holding INT button for an additional 30 seconds.
6) At this point, unit was booted with all four disk access lights flashing red, diagnostic light was flashing slowly and the power light was flashing slowly.
7) Configured a RAID 5 array. Telneted into the Terastation and saw that process raid5d was running at ~ 43% cpu usage and process raid5syncd was running at ~36% cpu usage.
8) Opened the web ui, noticed that it said that the array was not formatted.
9) Allowed RAID sync to complete (something like 10-11 hours).
10) went in the web ui, selected Disk management, disk format. Formatted the array to XFS.
11) All has been well for the past 48 hours.
Not sure why this worked. Seems to be a combination of restarting the Terastation (clearing temp files maybe?) and no format during disk check . All I know is so far it worked.
Best of luck to you all. garrman
Exact same problem on my HS-DHTGL/R5, firmware 2.14
However, problem not solved. Any pointers?
Completely confused as to why RAID 5 will not complete disk checking, but other RAID setups will.
I had the "perpetual RAID5 checking" problem on a HD-H1.0TGL/R5 after a power outage. It would go up to ~20% (in a day!), and then spontaneously restart from 0, or the box becomes inaccessible. Following the advice found here, I did try forcing firmware update (it was already v2.16, but just in case) and tried to wait it out several times, but gave up in the end after several days. I noticed that during these checks the green status light for disk 3 was almost constantly on, while others would blink only briefly. Deleting the RAID5 and checking and formatting each disk individually worked fine - but I am not sure how thorough this formatting is - it finishes pretty quickly. Also, I tried making two RAID1 arrays (one with disks 1 and 2; second with 3 and 4), and noticed that the RAID check on 1-2 was moving forward much faster than 3-4. So, I decided to replace disk 3 - even though there was no clear indication of a fault there. After replacing disk 3, the RAID checking competed in ~6 hours. The red DIAG was still flashing indicating HDD error, but after a graceful shutdown (through the web UI) and power down with the switch on the back, it was then back to normal when powered up - and now I am copying back the data saved on an external drive.
Side comment: I wish Buffalo Technology came up with a better system of error indication - how exactly is one supposed to distinguish between "6 blinks in 4 seconds"(HDD error) and "7 blinks in 4 seconds" (motherboard)? Keep counting for 40 seconds?