Unraid OS version 6.12.0-rc6 available

I seem to be getting constant parity-checks when booting up my NAS with this RC6

 

I shut mine down at night mainly due to the fact that it doesn't need to run 24/7 for power consumption reasons (I have a second 24/7 one already running).

 

Whether i shut it down using a script, or using the 'shutdown' button on the main page every time i boot it back up I seem to get it complaining about an unclean-shutdown and requiring a parity check. Power is never yanked from the system (its on a UPS)

 

I have run this multiple times now and always end up with 0 errors.

 

Anyone else seeing this behaviour?

9 minutes ago, Joshndroid said:

Whether i shut it down using a script, or using the 'shutdown' button on the main page

Try stopping the array before hitting the shutdown GUI button.

20 minutes ago, JonathanM said:

Try stopping the array before hitting the shutdown GUI button.

 

Thanks, should I be doing this always anyway as a future reference? (as if so, is there a line i should add to my script to run this inline with my shutdown command after a sleep of say 30s)

18 minutes ago, Joshndroid said:

 

Thanks, should I be doing this always anyway as a future reference? (as if so, is there a line i should add to my script to run this inline with my shutdown command after a sleep of say 30s)

Shouldn't be necessary, it's just a troubleshooting step to see if the array stops in a reasonable period of time. There is likely something keeping the array from stopping, causing the shutdown to kill the array prematurely, triggering a parity check on startup.

 

If that's the case, you need to figure out what is keeping the array from stopping.

Have noticed on this release that when enabling vm manager and it creates the iso folder it creates it on the array not cache and when you set it to cache it just deletes the whole share, when the system is rebooted it re creates it again on the array.

 

Even noticed that if i set the iso to another folder on the cache in the vm manager it still recreates a different iso folder after a reboot.

7 hours ago, steveBBB said:

Have noticed on this release that when enabling vm manager and it creates the iso folder it creates it on the array not cache and when you set it to cache it just deletes the whole share, when the system is rebooted it re creates it again on the array

What share have you set to hold the iso file (the normal default is ‘isos’).   What is the Use Cache setting for the share holding the iso files.

 

We may be able to give better informed feedback if you attach your system’s diagnostics zip file to your next post in this thread.

Updated from 6.11 today - no major issues thus far.

Hi

quick question: just building a 2nd UnRAID server - is it safe to start with the current RC6 and use ZFS.

My question is towards the Pool(s) - to ensure i don't have to re-create them after the upgrade to the final version.

2nd question: i know ZFS can add ZFS-cache und ZFS-log drives... is that possible with the UnRAID implementation (via GUI).

 

Thank you in advance.
(I hope it´s allowed to ask those questions here - and it´s not limited to Bug-reports)

 

Best

Thorsten 

1 hour ago, Thorsten Gehrig said:

Hi

quick question: just building a 2nd UnRAID server - is it safe to start with the current RC6 and use ZFS.

My question is towards the Pool(s) - to ensure i don't have to re-create them after the upgrade to the final version.

2nd question: i know ZFS can add ZFS-cache und ZFS-log drives... is that possible with the UnRAID implementation (via GUI).

 

Thank you in advance.
(I hope it´s allowed to ask those questions here - and it´s not limited to Bug-reports)

 

Best

Thorsten 

 

1. I think so, but also depends on your use case. It is a RC not a stable release so I wouldn't put production data or important data while using a RC.

2. I actually would like to know this too, but I assume no as there has been no mention of it from what I've read.

I guess i found a bug:

on a new setuo UnRAID the "appdata" and "domain" (etc) are set to "primary storage=array".

If i change the primary storage to a 2nd pool and click on "apply" the restult is a pop-up confirming "share deleted".

(no new share is created - it´s just deleted)

 

Best

Thorsten

On 5/22/2023 at 9:20 AM, casperse said:

No But I do have a VM with Chrome having open tabs to Unraid?
I ended up having to cut the power (Not even SSH terminal could get access) so I rolled back to the stable version.
Everything works fine now? (Same open window on my VM.)

 

I have now made sure that CPU-0 / Core6 is not pinned to anything (Hoping this would give me access to the UI after upgrade). 

I now face a new issue, seems my "ident.cfg" on my flash is wiped clean, server named TOWER again and so on.
I then wanted to just use my flash backup and replace the ident.cfg with the org. one in the flash config folder.

But after upgrade to 6.12.0-RC6  I no longer have access to my "flash" when I acces my "Root share".
Anyone who can help me out? or is this a bug? (New mount points)

I haven't changed anything on the flash: (My windows user have Read/write access:

Edited June 4, 2023 by casperse

On 5/25/2023 at 11:35 PM, EDACerton said:

This is already being worked on:

 

 

 

I'm still strongly in favour of making this change optional; there are plenty solutions other than tailscale and plain wireguard (nebula, zerotier, netmaker, ...). Until now the unraid suggestion was to run these in docker - now all of those would require plugins.

Also, while I agree additional hardening is a good thing, ssh is imo not the top priority here - ssh is one of the few protocols that's actually safe to expose to the web - there are much more vulnerable services on Unraid that are still served on all interfaces (smb, web-interface, ...).

 

 

15 hours ago, Thorsten Gehrig said:

2nd question: i know ZFS can add ZFS-cache und ZFS-log drives... is that possible with the UnRAID implementation (via GUI).

It's possible but for now only manually using the CLI and then re-import the pool, instructions here.

11 hours ago, Thorsten Gehrig said:

I guess i found a bug:

This is fixed for rc7.

1 hour ago, casperse said:

But after upgrade to 6.12.0-RC6  I no longer have access to my "flash" when I acces my "Root share".

I've never used a rootshare but AFAIK it only gives access to all user shares, flash share is outside that, e.g., \\tower\flash

3 minutes ago, JorgeB said:

I've never used a rootshare but AFAIK it only gives access to all user shares, flash share is outside that, e.g., \\tower\flash

 

I did had access to the flash drive under the rootshare before.
But you are right I can just map to the flash from the hostname/IP, I just liked to have all in one rootshare 🙂

7 hours ago, JorgeB said:

It's possible but for now only manually using the CLI and then re-import the pool, instructions here.

 

Thanks for the additional info. Now I know for sure I won't be touching that until it is fully released. 🙃
IMO, this update should be pushed back until it is ready for full ZFS support.

I am now unable to connect to my mapped drives from Windows 11 since upgrading.  Could certainly be coincidence, but I've not had this issue until updating unraid, and every "windows" solution I've found has not helped.

 

I'm getting this when running diagnostics...

 

"The device or resource (Computer) is not setup to accept connections on port "The File and printer sharing (SMB)"

 

I am able to connect to unraid via Putty, but not with windows explorer.  Neither by connecting to the server directly, nor a mapped drive, nor can a create a new mapped drive connection.

 

I've tried

• turning on and off "File and Printer sharing for Windows Networks" from my adapter settings

 

1. Press the Windows key + R shortcut to open the Run command window.
2. Type in gpedit.msc and press OK.
3. Select Computer Configuration -> Administrative Templates -> Network -> Lanman Workstation and double click Enable insecure guest logons and set it to Enabled.
4. Now attempt to access \\tower

 

• reinstalling network drivers

 

so far, nothing has helped.

 

any suggestions on what to try next?

media-diagnostics-20230604-1645.zip

@JustinChase

have you tried connecting with the ip adress?

1 hour ago, Meannekes said:

@JustinChase

have you tried connecting with the ip adress?

I had not.

 

I reverted to 6.12.0-rc5 and it worked instantly after rebooting the server.  I didn't even have to reboot windows.

 

I guess this is now an official bug report.

 

Hopefully this gets fixed before final release.

Not strictly a bug, but certainly weird.

I changed my cache pool to ZFS, and redownloaded all docker containers.

I did not do any snapshots, but ZFS master reported that I have hundreds of snapshots and it shows each of the image file as a snapshot. I use directory for docker images.

Is this an expected behavior? Also please note that since each docker image counts as a snapshot, it takes ~10 seconds to load the main page. Granted, if I use docker.img or don't use ZFS master it will be less of a problem, but still weird.

55 minutes ago, goezzkim said:

Not strictly a bug, but certainly weird.

I changed my cache pool to ZFS, and redownloaded all docker containers.

I did not do any snapshots, but ZFS master reported that I have hundreds of snapshots and it shows each of the image file as a snapshot. I use directory for docker images.

Is this an expected behavior? Also please note that since each docker image counts as a snapshot, it takes ~10 seconds to load the main page. Granted, if I use docker.img or don't use ZFS master it will be less of a problem, but still weird.

 

After investigating more, it looks like it's an expected behavior(!!) of docker using zfs driver. I certainly didn't expect this, I think I should revert it back to btrfs.

Edited June 5, 2023 by goezzkim

1 hour ago, goezzkim said:

 

After investigating more, it looks like it's an expected behavior(!!) of docker using zfs driver. I certainly didn't expect this, I think I should revert it back to btrfs.

It's also caused super high disk usage for some. docker.img on ZFS is fine.