I replaced a 2TB WD Green that was having issues today with a Seagate Black 2TB (seem to be having alot of failing WD Greens so trying out Seagate).....anyway...the format has been running all day and I dont remember the WD Greens taking this long. Is there a way to check on the status of the format....(get a progress). I dont see one via the GUI, but im guessing there is probably a CLI command that will show it.

I replaced a 2TB WD Green that was having issues today with a Seagate Black 2TB (seem to be having alot of failing WD Greens so trying out Seagate).....anyway...the format has been running all day and I dont remember the WD Greens taking this long. Is there a way to check on the status of the format....(get a progress). I dont see one via the GUI, but im guessing there is probably a CLI command that will show it.

Disks that are replaced (and re-constructed) are not formatted.  They are completely re-constructed by reading all the other disks.

 

A re-construction takes as long as a parity check. 

 

It sounds as if something else is occurring, so...

What version of unRAID are you running?

Did you press a "Format" button?

Post a screen-shot so we can see how it looks.

Attach a syslog to your next post

Describe in detail what you did to replace the one drive with the other.

 

Joe L.

A few weeks ago I was trying to install subsonic addon but never really got it working, then we went on vacation for 2 weeks. On saturday night we decided to watch a movie with some friends. When the movie started playing it was skipping quite frequently. First tried rebooting the computer on the TV in the living room but it continued to skip. I then suspected the subsonic addon was somehow the culprit as it was working great for about a year prior. I was in a hurry and not really sure how to properly uninstall the subsonic app so instead I copied my Pro.key, then erased the flash drive and reloaded it with 4.7 which is the same version it had before. The movie still continued to skip however it was only every couple of minutes and we decided to finish watching the movie like this. Once the movie was over and our friends went home and I continued troubleshooting.

 

First I tried playing the movie from the computer in my office with it reading from the tower, didn't skip as bad, but still doing it. I then copied the movie to my desktop and played it from there. Worked good. Looked in the disk shares to see which disk the movie resided on, which was disk 1, a WD Green 2TB drive. I shut down the server and removed the drive and plugged it into my desktop and ran Crystal Disk Info. It showed Caution on the drive with corrected sectors and uncorrected sectors I belive. I then removed the other drives in the array 1 by 1 and all showed healthy on SMART stats using Crystal Disk Info. I then decided to bring the array back online and attempt to play the movie from a different disk in the array just to confirm it was the disk and not something else. Since I am using shares and had a copy of the movie on my desktop, I attempted to delete the copy on disk 1 and then I would copy it back to disk 2 and play it from that disk to confirm it was the drive and not the movie or server somehow. When I attempted to delete the folder, it went very slow and eventually hung up, then showed as unformatted. I got nervous and removed the drive, plugged it back into my PC to see if I could copy the files before it completely failed but my PC showed it as unformatted as well. I then ran a program called DRW that I have used before to recover files from failed drives but it was moving extremely slow. I made up my mind that the drive was likely toast and decided to replace it with another Seagate 2TB black drive. I inserted this new disk into the same slot as the failed one and brought the array back online. It showed unformated and now my parity drive showed invalid. I thought the invalid status on the parity was likely bogus or normal during a reconstruction so I clicked the button to format and reconstruct the unformatted drive (disk 1), then verified that the LED indicator on the front of the ICY DOCK was red/active for only disk 1. The parity was green, disk 1 was red and all the other data drives were green as if it were formatting only the new drive. I figured it would eventually finish and then start the reconstruction which would read from all drives, however 7 hours later it was still showing the same. During this time I had been reading forums and saw several folks talking about "reiserfsck --check" command and decided to give it a try. I powered down the server again and replace the new drive with the old to try to this command to see if I could get the disk back to a read-able state. I ran the "reiserfsck --check /dev/md1" command and it said results would be put into a file (cant remember the name) but then I couldn't find it to view it. I then tried the "reiserfsck --fix-fixable /dev/md1" command and it said the drive could not be mounted and it was likely a bad drive and needed replaced. So I put the new drive back in and the LED indicator on the ICY DOCK immediately goes red for disk 1 and all other drives including parity are green. Main page viewer shows disk 1 is unformated and the parity drive is still showing invalid with all other drives showing green.

 

Sorry for the length of the post, but I was trying to be as detailed as possible.

syslog-2012-04-02.txt

I attached the syslog to the previous reply, but the size restriction wouldn't let me post the screen shots with it. Here is the Main page.

This is the unmenu main page.

One more detail I forgot to mention. All the drives in my array were previously MBR: unaligned. I was reading on the forums about 4k aligned is better so the first format attempt of the new drive was done using 4k aligned. The second attempt I started late last night and this time i used MBR: unaligned and it is still going. Not sure if that makes a difference or not.

I checked the SMART stats on Disk 1 again and it shows 32 pending sector counts and 1 uncorrectable sector count. I believe the data is still in tact on the Disk 1, but if not then surely its still good on the parity drive. Can we not just tell the array to trust the parity drive and rebuild disk 1 using it?

i don't think its doing anything.

 

guessing by the screenshots, you removed your parity drive and added the Seagate as an extra disk? or your parity was accidentally disconnected when you removed the WD drive.

 

check your connections and confirm all HDDs are detected in the BIOS as you boot up.

 

that is, unless Joe has any other suggestions

So, what should I do, I cant read disk 1 to get the files from it and I cannot rebuild it because it says that the parity drive is invalid. I'm thinking force the array to trust the parity drive and rebuild disk 1, however the manual says never do this with a failed disk or while upgrading a disk....