Jclendineng

Here you go...It's a test server for playing with ZFS, so its not running the array currently. unraid-diagnostics-20230311-2058.zip

I am still getting this with the latest UD in Unraid 6.11.5 by the way... invalid cmd: /webGui/scripts/disk_log

I probably did something dumb here but I had to reinstall ABS from scratch and I can't get sockets working with default swag reverse proxy. 2023-03-08 22:00:47 ERROR Api called without a token /authorize 2023-03-08 22:00:50 ERROR Api called without a token /libraries 2023-03-08 22:00:50 ERROR Api called without a token /tasks I did the suggestion earlier in the thread with the same results. It works fine by IP address. I have a custom IP I use (macvlan) and this was working ~a month ago or so. Any help would be great!

Yes I have it in a private git repo that I call on startup to unlock, so its versioned and is the same. I also had a backup on a private cloud that also didn't work. Very odd but I'm assuming the simplest answer is ram issue or hardware issue, perhaps a controller is going bad, the board is pretty old, and its a gigabyte consumer board meaning anything possible

I thought memory (even though its relatively new ram...) but I pulled the drives, mounted them on a different server and attempted the unlock with the same results, so I have ruled out hardware (minus anything hardware related that might have done this)

I have 1 data drive, but I also have encrypted cache drive, with its own header, the keyfile doesn't work for either header anymore, but the header itself doesnt look corrupted meaning something must be wrong with the keyslot...in any case this is on me for not backing up the header. Edit. 1 data, 1 parity and 1 cache.

Yes, though I think I found the issue, or at least the most logical one. I don’t have ecc ram in my current server and a cosmic ray or random bit flip happened hosing my key slot. My key slot looks fine with the checker and the luks header looks fine so the only thing it could be is 1 bit maybe got changed causing my keyfile to fail. Unfortunate and I’ll have to format all my drives now but it happens and apparently in my research on this, happens a lot. If even 1 bit fails on the key slot the header is hosed. There are a couple threads on consistently backing up the header with mixed results, unfortunately I didn’t have a good header backup so all my data is gone. Sucks, but it’s price you pay for encryption, especially when it’s not fully supported by unraid. Future me will keep up to date header backups. Edit. I think it would be handy to have the script give the option to use more than 1 key slot when updating the key, instead of always deleting the old key by default, so if someone wanted to use a pass phrase and a key file for example they could. Unrelated to this but it would be a good addition to an already nice script (until unraid makes it officially a part of the gui as limetech has stated they will do)

I ran into an issue where I restarted unraid and I get "No key available with this passphrase.". This keyfile has worked for ages, and now it wont work, even manually, any ideas?

Could you share what you did, I restarted my server and unraid will no longer unencrypt. The keyfile thats worked for ages now fails. No key available with this passphrase.

I would turn incremental off, I enabled it, and when my monthly parity check was running it kept stopping (as expected) but then it completely freaked out, saying it couldn't write to the disk etc (new disk, cleared and all that), and unfortunately killed all the data forcing me to rebuild from parity. I do have parity so I didn't lose anything but be warned that something may not be completely stable here. Have a backup.

What sort of ZFS data recovery/protection features can we expect from the GUI, I know ZFS has self healing but if a drive fails and needs replaced in a raidz configuration, will the GUI support basic hdd replacement and monitoring? What about zfs errors. Will those be exposed in syslog?

I have a new server set up to test the zfs stuff, I 100% wont be doing testing on my main server 😂 then if I can get raidz working as expected then I’ll encrypt to test that stability and let it bake a bit after final release and copy my data over. I felt like ZFS official support deserved a server upgrade and it’s a good time to do it since I’ll need to reformat anyways for zfs.

I can finally have a proper all ssd pool now! ZFS RaidZ1 with 4 SSDs. Looking forward to that.

FYI this works great on latest Unraid

I second a harder code. If I know it’s Morse I know I don’t know it. You should use a cipher machine, preferable ex ww2

Thank you, I for one, as someone who doesn’t particularly have time to learn morris code at work appreciate that.

I forgot about that!! I have 256gb in my unraid server that ZFS would put to good use.

I was just being dumb. RAIDz1-3 would do what I want, I had a brain fart. So 3 drives in raidz1 would be best storage to parity efficiency

So I’m thinking about doing a flash based zfs array with this. How does parity with with ssds? TRIM and all that. Will a flash based zfs array in raidz work with parity? So 3 ssds plus a large platter drive for parity. The ssds will make up the zfs array and parity would be platter based. Would that be possible with zfs?

Following, in the exact same boat. I have a r630 I want to move to but don’t want to fill with hdds as ssds are more power efficient and a lot faster plus less moving parts. Unsure how unraid handles ssd’s in an array and if I can even use ssd’s with parity. Need to figure it out, or I’ll just use bare metal podman or rancher or something to manage docker. I couldn’t find anything conclusive on the forums about ssd’s and parity.

Interesting I was going to ask about encrypted zfs since I use encrypted xfs currently. So I’m curious why not zfs native? Isn’t it easier to manage then LUKs? Sent from my iPhone using Tapatalk

I would update unraid first and post diagnostics when it happens again, looks like you are on 6.11.1, 6.11.5 is the latest bug fix release and contains a lot of improvements. there were numerous issues with 6.11 that have been subsequently fixed in point updates, that's where I would start. I also see some virtio errors that probably aren't related but usually occur when there is a GPU or other device that's not being passed correctly to a VM. The app data backup/restore issue is also separate and you would ask in that thread, but if you have error checking on the backups that could cause that error since (depending on what you are backing up) it will always fail that (as an example if you backup USB or system, since checksums will fail as the files will have changed between start and end of backup)

OK I tried that and same issue. I even cleared all settings and re-setup the interfaces...

I am, its just utilization, but I have updated again and included the troubleshooting steps listed in the OP, thank you! Edit: I updated and it works...Im not 100% sure why. Possibly the APU needed reset, but I had thought that last time and rebooted which should reset it, which didn't fix. Ill reply with drag info if it pops up again...Thank you!

I am getting "Multiple NICs on the same IPv4 network" eth3 and eth7 both have IP addresses on the 10.0.2.0 network. This is rarely a valid configuration. I don't have an eth7 anymore, I did throw a NIC in to test something but I didn't assign any IP address and its been removed. I already deleted and regenerated the docker config file and network rules file to try and solve but the warning won't go away. Attached are diagnostics. hank you for all your work!T tower-diagnostics-20221226-0851.zip