Missing drives and reappearing drives in Greyhole

[PatrickDickey]

Has anyone else ran into this issue either on Amahi 6 or on Amahi 7? I'm getting emails from my hda stating that a drive's UUID has changed, so greyhole is marking it as missing. And then I get a separate email stating that the drive has returned to the pool.

Snippet of the missing drive email:

This is an automated email from Greyhole.

One (or more) of your storage pool drives has disappeared:
/var/hda/files/drives/drive5/gh: expected partition UUID: b6557d0e-9433-41cf-be24-cd40eddac1f8; current partition UUID: 51aeb7d6-e19e-4b33-9bec-22130c4fb484

This either means this mount is currently unmounted, or you forgot to use 'greyhole --replace' when you changed this drive.

Snippet of the drive returned email:

This is an automated email from Greyhole.

One (or more) of your storage pool drives came back:
/var/hda/files/drives/drive5/gh was missing; it's now available again.

A fsck will now start, to fix the symlinks found in your shares, when possible.
You'll receive a report email once that fsck run completes.

I've got logwatch set up, and it shows that the drive is offlined due to errors. I don't know if this is happening because I moved the drives around on a USB Card, or if the drives are actually failing, or if the card/motherboard (it's an old computer) is failing. This all started after I upgraded from Amahi 6 to Amahi 7, but I don't think that's the reason.

Here's my /etc/fstab, so you can see that it's defined there:

Code: Select all

[root@localhost media]# cat /etc/fstab # # /etc/fstab # Created by anaconda on Sun Sep 29 13:51:49 2013 # # Accessible filesystems, by reference, are maintained under '/dev/disk' # See man pages fstab(5), findfs(8), mount(8) and/or blkid(8) for more info # /dev/mapper/hda-root / ext4 defaults 1 1 UUID=41bbae2f-7461-402c-9827-9c09dc34db73 /boot ext4 defaults 1 2 /dev/mapper/hda-home /home ext4 defaults 1 2 /dev/mapper/hda-swap swap swap defaults 0 0 UUID=5daf66e3-61f2-438e-a288-a218c922434e /var/hda/files/drives/drive1 ext4 defaults 1 2 UUID=ee6b982e-4a79-47af-a2e9-d408591c8935 /var/hda/files/drives/drive2 ext4 defaults 1 2 UUID=561975f5-db29-4f89-981c-3c110c64cddd /var/hda/files/drives/drive4 ext4 defaults 1 2 UUID=b6557d0e-9433-41cf-be24-cd40eddac1f8 /var/hda/files/drives/drive5 ext4 defaults 1 2 UUID=f30859b4-6f6d-41d2-aa80-a6afea739f65 /var/hda/files/drives/drive6 ext4 defaults 1 2 UUID=fc449823-d4bc-4dc3-9596-b818e388378e /var/hda/files/drives/drive9 ext4 defaults 1 2

And the results of ls -l /dev/disk/by-uuid

Code: Select all

[root@localhost media]# ls -l /dev/disk/by-uuid total 0 lrwxrwxrwx 1 root root 10 Oct 22 20:31 172a8d6f-9273-4e4e-972c-1603d0a59296 -> ../../dm-1 lrwxrwxrwx 1 root root 9 Oct 22 20:31 2011-02-17-17-10-30-00 -> ../../sr1 lrwxrwxrwx 1 root root 10 Oct 22 20:31 41bbae2f-7461-402c-9827-9c09dc34db73 -> ../../sda1 lrwxrwxrwx 1 root root 10 Oct 22 20:31 51aeb7d6-e19e-4b33-9bec-22130c4fb484 -> ../../dm-0 lrwxrwxrwx 1 root root 10 Oct 22 20:31 561975f5-db29-4f89-981c-3c110c64cddd -> ../../sde1 lrwxrwxrwx 1 root root 10 Oct 22 20:31 5daf66e3-61f2-438e-a288-a218c922434e -> ../../sdb1 lrwxrwxrwx 1 root root 9 Oct 23 21:15 705b88d0-2182-4c90-b4a4-74985e75bb01 -> ../../sdc lrwxrwxrwx 1 root root 10 Oct 22 20:31 7b4e7bbf-de85-4144-a49e-4a8c73e7f0f9 -> ../../dm-2 lrwxrwxrwx 1 root root 10 Oct 22 20:31 a9603d9e-9a60-4d50-a182-dabd1da0217f -> ../../sda3 lrwxrwxrwx 1 root root 10 Oct 24 13:06 b6557d0e-9433-41cf-be24-cd40eddac1f8 -> ../../sdh1 lrwxrwxrwx 1 root root 10 Oct 23 19:42 ee6b982e-4a79-47af-a2e9-d408591c8935 -> ../../sdi1 lrwxrwxrwx 1 root root 10 Oct 22 20:31 f30859b4-6f6d-41d2-aa80-a6afea739f65 -> ../../sdg1 lrwxrwxrwx 1 root root 10 Oct 23 21:15 fc449823-d4bc-4dc3-9596-b818e388378e -> ../../sdc1

It appears that it's swapping dm-0 for the actual drive at random times. So, I guess I'm confused by what's happening and how to fix this. It's preventing me from adding a drive, as now I have a fsck with about 9,000,000 items in the queue that won't go away, running.

Have a great day.
Patrick.

[PatrickDickey]

An update on this:

I thought maybe the USB port was going bad, so I swapped it out with another drive to verify. That drive started missing, so I stopped using the port altogether (it's on an USB Adapter card, and not internal to the computer). Now I'm getting other drives that are "missing" and then "returning". All of this started happening after I upgraded to Amahi 7 and the latest versions of greyhole.

So, I guess my questions are these:

1. Is anyone else experiencing this issue at all?
2. What are the odds that all of my USB ports have suddenly died around the same time that I upgraded?
3. Short of buying a new computer or buying a new adapter card, does anyone have suggestions for how to fix this?

Thanks, and have a great day.
Patrick.

[bigfoot65]

1. Is anyone else experiencing this issue at all?

I have never had this issue or heard of anyone else experiencing it.

2. What are the odds that all of my USB ports have suddenly died around the same time that I upgraded?

The ports probably did not die, but it could be a driver issue with Fedora 19. I have heard of such cases with other hardware.

3. Short of buying a new computer or buying a new adapter card, does anyone have suggestions for how to fix this?

Is it all USB ports? If not, pick one that appears to work and try connecting a hub to it. Have you considered getting away from USB and going to E-SATA? If you have to buy a card, it might work better. I know speeds are a lot faster than USB.

[PatrickDickey]

1. Is anyone else experiencing this issue at all?

I have never had this issue or heard of anyone else experiencing it.

2. What are the odds that all of my USB ports have suddenly died around the same time that I upgraded?

The ports probably did not die, but it could be a driver issue with Fedora 19. I have heard of such cases with other hardware.

3. Short of buying a new computer or buying a new adapter card, does anyone have suggestions for how to fix this?

Is it all USB ports? If not, pick one that appears to work and try connecting a hub to it. Have you considered getting away from USB and going to E-SATA? If you have to buy a card, it might work better. I know speeds are a lot faster than USB.

It doesn't matter which port I plug the drives into now. And one of the drives is brand new (purchased two weeks ago), so I'm fairly certain that it's not the drives. The main reason that I use USB is that most of the drives are external drives. I am seriously considering purchasing a SATA controller card, and ripping the drives out of their cases. My reasoning for thinking that the port on the card was the issue is that when I plugged a different drive into that port, it started disappearing.

The odd thing is, the first drive that was disappearing gave me messages like expected UUID:..... but found /dev/mapper (so it was finding my LVM there instead), and when I swapped drives, it found my /temp directory for the UUID.

The computer is quite old, so I may buy a newer tower from my local computer shop, and swap it out. Plus that *should* give me some extra SATA ports to play with, so I can hopefully eliminate some of the external drives.

Have a great day.
Patrick.

[bigfoot65]

Ok, that's what I figured. Never thought it was the drives. I was more focused on the controller. Could it be the computer cannot handle all the drives at once? I have seen that happen before.

I think you are right, to use a lot of drives may require a new machine. SATA is the best in my opinion. I have 4 drives in mine of which 3 are for Greyhole. Really don't see a need for more that 3 drives as I don't fill up one drive with my files.

[PatrickDickey]

Ok, that's what I figured. Never thought it was the drives. I was more focused on the controller. Could it be the computer cannot handle all the drives at once? I have seen that happen before.

I think you are right, to use a lot of drives may require a new machine. SATA is the best in my opinion. I have 4 drives in mine of which 3 are for Greyhole. Really don't see a need for more that 3 drives as I don't fill up one drive with my files.

I never thought about the controller not being able to handle all of the drives. But, since the computer is about 10 years old, I'd guess that is probably what's going on. I guess I'll be stopping into my local computer shop today to see what he has set up.

I'll post either way, when I figure out a solution. Plus, I can probably remove some of the drives, as two of them are over 1TB in size, so I probably don't need the smaller ones anymore.

Have a great day.
Patrick.

[bigfoot65]

Good to hear Amahi does work on such an old computer. Unfortunately, I think you are right in looking for a more current model. Sometimes we try to get too much out of those old dogs LOL

[PatrickDickey]

Good to hear Amahi does work on such an old computer. Unfortunately, I think you are right in looking for a more current model. Sometimes we try to get too much out of those old dogs LOL

It's going to be a few weeks before I can replace the computer now. Just because I'll either have to settle for some cheap POS from Walmart, or build up a bit more money for a decent built system.

So, right now, I have a 2 - 320GB External drives, a 80GB internal drive, a 2TB external drive, and a 3TB external drive hooked up. And I'm in a perpetual fsck, because it keeps losing drives and then finding them again. What are my options here? Is there any way to cancel all of the fsck's that are scheduled (using greyhole -C doesn't work, since they're running in response to a non-existent error condition).

I'm half tempted to pull one of the 320 GB drives off, and then use the greyhole --gone command to force it to rebuild those files. Then pull the other drive off, and use the command again to force it to rebuild the files. Both of the drives are completely full, due to the fsck's which are running, so it's not like it'll rebuild both copies on those drives. /dev/sdi and /dev/sdj are the two smaller drives.

Code: Select all

--------------------- Disk Space Begin ------------------------ Filesystem Size Used Avail Use% Mounted on /dev/mapper/hda-root 43G 16G 25G 39% / devtmpfs 1004M 0 1004M 0% /dev /dev/sdg1 1.8T 1.3T 510G 71% /var/hda/files/drives/drive5 /dev/sdb1 76G 72G 245M 100% /var/hda/files/drives/drive1 /dev/sda1 477M 107M 341M 24% /boot /dev/mapper/hda-home 21G 80M 20G 1% /home /dev/sdj1 294G 279G 0 100% /var/hda/files/drives/drive2 /dev/sdi1 294G 279G 0 100% /var/hda/files/drives/drive4 /dev/sdk1 2.7T 502G 2.1T 20% /var/hda/files/drives/drive9 /dev/sdb1 => 100% Used. Warning. Disk Filling up. /dev/sdj1 => 100% Used. Warning. Disk Filling up. /dev/sdi1 => 100% Used. Warning. Disk Filling up. ---------------------- Disk Space End -------------------------

Would that work, or will I just create more issues?

Have a great weekend.
Patrick.

[bigfoot65]

How about trying to clear the queue? Guidance can be found here.
https://wiki.amahi.org/index.php/Greyho ... leshooting

[kikkegek]

sorry to bring up an old thread, but I am having exactly the same problem.

drives (mostly drive2) keep disappearing and reappearing in the storage pool.

This is an automated email from Greyhole.

One (or more) of your storage pool drives has disappeared:
/var/hda/files/drives/drive2/gh: expected partition UUID: 7493b2b3-9ec4-4711-92ef-388390822826; current partition UUID: 7493b2b3-9ec4-4711-92ef-388390822826

This either means this mount is currently unmounted, or you forgot to use 'greyhole --replace' when you changed this drive.

and then comes back 3 minutes later

This is an automated email from Greyhole.

One (or more) of your storage pool drives came back:
/var/hda/files/drives/drive2/gh was missing; it's now available again.

A fsck will now start, to fix the symlinks found in your shares, when possible.
You'll receive a report email once that fsck run completes.

and ussually when this happens after the FSCK some old files that had been deleted always re-appear again...its getting pretty annoying.

Any news/progress on this bug/error?