So I know I messed this up but how do I fix it. I removed two 1TB drives because they were on a PCI slot and I'm upgrading them to a PCI express card. However disk 8 was also on the pci card and I moved it to the onboard SATA. In the process I somehow created a duplicate disk of 8. Now it says it is 2TB however the % full is empty and shows negative free space. I think it believes I have 4TB. I know I have around 1TB of data on the drive.

 

Questions: what logs do you need to problem solve this? Any suggestions? How can I move all the data off of Drive 8 and just remove the drive completely? I use teracopy to move files however it does not have permissions to move and delete folders. How do I do this in the unraid cmd prompt? Any help and suggestions would be much appreciated.

 

For starters, the serial # for drive #8 indicates that it is indeed a 2TB drive; so it's not clear why you "... think it believes I have 4TB", or why you said you "somehow created a duplicate disk".    Why did you make those statements?

 

Your configuration also does not match what you've listed in your signature ==> you list 10 drives; but only show 8 in the system.  I presume this is simply an oversight ... or are some of your drives not assigned?

 

What size does the BIOS show for the hard drive in question (# ??

 

I will add that in your signature you have 5 2TB drives and in your current screen shot there are 5 2TB drives. I think the question garycase is getting at is are you sure this is a 1TB drive vise a 2TB drive as indicated? Although this doesn't answer your question of the reported negative space. Attach a copy of your system log to help isolate problem if the drive is actually reporting incorrectly.

Did you guys even look at his screen shots. It clearly shows -1,3xx,xxx free for disk 8.

 

No need to bust his nuts over petty details.

 

@op. Post up the syslog.

Yes, it clearly shows that (a) it's a 2TB drive and (b) it's got a negative value for free space.    But the display also does not match what his listed configuration is; and his description implies it's a 1TB drive (although the serial number shows that it's clearly a 2TB drive).

 

Without a bit more detail, it's hard to say exactly what the issue is ... but confirming that the BIOS sees it correctly; and checking to see what a newly configured flash drive shows, will isolate whether this is an issue with the drive or with the UnRAID configuration info.

 

If you read my post I acknowledge the negative space, and ask for a system log. I see no ball busting just questions for the op to look at, and perhaps be clearer on what he has as a problem. My goodness you act like we were harassing him/her.

OK so drive 8 is indeed a 2TB drive sorry I did not make that clear in original post. I am also running v4.7 waiting to upgrade to 5 when this is all cleared up. I have looked at my syslog but it just shows me moving a bunch of files off of drive 8. I attached my first 4 syslogs in this email and my next 5 will be in the following email. What do you mean by a newly configured flash drive? I ran the cmd fdisk -l -u /dev/sdh and it shows up as a 2TB drive, but I did get this error Partition1 does not end on cylinder boundary. Not sure how to send that message to a text file if you let me know how I'll do it.

 

I know I did this to myself. What happened was I had a 1TB and the 2TB (drive in question) on a PCI SATA card. The performance was awful, so I decided to get a new PCIe SATA card. In the meantime I had 2 1TB drives on the system that had nothing on them so I removed those 2 1TB drives and connected the 2TB drive to the motherboard. Now I know how to due this properly by running the setinit cmd, however I of course did not do that first. Under the myMain I removed the 2TB and the 2 1TB drives and then created a new drive with the 2TB serial number, hence why I think it is treating it as if I have 4TB of space on that one drive.

 

I have all my data backed up so no worries there. I receive the PCIe SATA card and some addition 2TB drives tomorrow. So should I just remove that drive and put in 1 of the new 2TB drives and let it rebuild. Then I could add the 2TB drive in question back into the array and hopefully it would have forgotten the previous settings?

 

I also just updated my configuration to be current.

syslogs.zip

Five more syslogs

syslogs.zip

I have all my data backed up so no worries there.

 

With all your data safely backed up, I'd just go straight to an update to v5.    That may, by itself, resolve this.  If anything doesn't work right; then just clear all the drives; create a new config; and copy all your data from your backups.

 

Ok, setting things representing the drives using MyMain will in no way screw up the stock unRAID interface.

 

How did you get the array working again if you did a bunch of changes in MyMain but didn't run ititconfig to reset the array after you removed the drives?

 

You know you're not properly describing what you did to get a valid array?

 

Personally, I think you screwed-up the filesystem on the disk. Look in the unRAID wiki for file system checks and run the reiserfsck command to check the file system on the disk.

 

After I made a bunch of changes in MyMain I did run initconfig and obtained a valid parity without losing any data. I will run the reserfsck cmd to check that disk.

unMENU does not re-read the sizes of the disks as you hot-plug them and switch them around. 

It is very easy to see how it is still using the original sizes of when it was first started.

 

You can re-start unMENU, using the button on its User-Scripts page, and it will likely show the correct free space.

You did reboot, right?

 

Joe - unRAID itself is showing a big negative free space on the disk so it can't just be unMENU.

You did reboot, right?

 

Joe - unRAID itself is showing a big negative free space on the disk so it can't just be unMENU.

4.7 does not support disks > 2.2TB.  That alone might be your only issue AND it is not hot-pluggable.

but its a 2.0TB drive.

Yes, it's only a 2TB drive, so that's not likely the issue.

 

This MAY have something to do with the blank slot (disk 7) ... it's been a long time since I've tried leaving a blank slot, and I've only done it with v5 (which works fine).

 

In any event, given that there are good backups available, and that something clearly isn't right with the current config;  I'd do what I suggested earlier:  Save your key file; redo the flash drive with the latest v5 configuration; add the keyfile to it; and then start the system with v5.    That will almost certainly resolve the problem .. and you'll be running the latest release and be ready to move to larger drives as needed.

 

I don't recall there being any issues with an empty slot.

 

I still feel it's rather premature to recommend more drastic measures until a reboot and a file system check. But, you can simply stop the array, unassign the disk, run the initconfig command and then start the array again and that disk will be gone. Then run a preclear on the disk and if that works without an issue the disk should be good to add again.

 

 

OK so I have rebooted several times and it does not fix the issue. I have run a file system check and there were  no problems found. So if I disconnect the drive and run initconfig cmd will it rebuild the data that was on that drive onto the other drives?

No.  It causes the parity disk to be invalidated. It is not used to rebuild a disk. It has the opposite effect of preventing any data disk from being rebuilt. Initconfig does not effect the disk contents at all.

No.  It causes the parity disk to be invalidated. It is not used to rebuild a disk. It has the opposite effect of preventing any data disk from being rebuilt. Initconfig does not effect the disk contents at all.

initconfig DOES immediately invalidate parity,  so don't press it if you've got a disk that is disabled or where you cannot get to its data.

 

It sounds as if unRAID is confused to the actual size of the disk vs. the partitioned size.

 

please post the output of

fdisk -lu /dev/sdX

(where sdX = the drive reporting the incorrect free space)

 

and

/root/mdcmd status | grep -i size

 

So if I disconnect the drive and run initconfig cmd will it rebuild the data that was on that drive onto the other drives?

 

As already noted, NO, all of that data will be gone.    Since you have all of the data backed up, that wouldn't seem to be a big issue; but DO be sure you save a directory of the drive before doing this, so it's easier to know what you need to copy to the array from your backups.

 

Ok so I have upgraded to v5.04. I ran check disk through the terminal and it found two errors which were fixed. For some reason the array has started but I can not access the web UI using Media/main it just shows a blank white page. I guess it is running a parity check? So this issue could be fixed but from the info below I'm not sure.

 

I ran the cmds suggested Sorry I don't know how to copy the text so I am typing this, pics are too large to upload.

fdisk -lu /dev/sdh resulted in

 

Disk /dev/sdh: 2000.4 GB 2000398934016 bytes

1 heads, 63 sectors/track, 62016336 cylinders, total 3907029168 sectors

Units = sectors 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

Disk identifier: 0x000000000

 

Device Boot                Start            End                Blocks                    Id        System

/dev/sdh1                      63          3907029167      1953514552+        83        Linux

Partition 1 does not end on cylinder boundary.

 

and /root/mdcmd status | grep -i size

Resulted in: Where disk 8 is the one of interest.

disksize.8=1953514552 and rdevsize.8=1953514552

The disk size is correct for a 2TB drive, so it seems this resolved your issue.

 

As for not being able to access //Media ... this may be an issue with the DNS through your router.

 

Try accessing it by the IP address ... or reboot your PC and see if you can then access //Media.

 

 

 

Yes typing in the ip address works and now the drive shows the correct space, so I guess the file check fix worked.