Unraid OS Version 7.0.0-beta.3 available

I have two UNRAID V6.12.13 servers ie 1 x prod and 1 x backup. I decided to move the backup server to V7.0-B3. Update went fine and machine rebooted ok. The array is made up of xfs disks and 1 parity drive and it did not start. If I try to start the array manually ie main --> array operation - I get a dialogue box with:

"Wrong Pool State

cache - too many wrong or missing devices"

 

The log shows the following "emhttpd: cache: ERROR:too many wrong or missing devices". Note on this machine I do not have a cache device let alone a cache pool ie the shares write directly to the array.

 

I rebooted and no change ie same problem where the array does not start. Same errors as above.

 

The Fix:

Went to main --> pool devices and noticed that there was "1" device in the pool but no device assigned. I set the devices pull down menu to "0" devices and after it corrected itself I could proceed to start the array without issue in main --> array operation. A reboot also confirmed that the array started automatically. All good.

 

In general log seams a bit more verbose but nothing in it worth writing home about (pretty much same warnings as previous version). No other issues encountered so far.

Edited October 7 by jim80z

2 hours ago, arsaurea said:

I can't enable Docker because it complains that these directories do not exist:

Looks like a bug, it starts fine with a vdisk, and it will create the folders, but not a docker directory.

4 hours ago, arsaurea said:

I've got the same problem

@arsaureatry adding a slash to the end, i.e., /mnt/user/system/docker/

 

That should allow you to start the service and create the shares

1 hour ago, JorgeB said:

@arsaureatry adding a slash to the end, i.e., /mnt/user/system/docker/

 

That should allow you to start the service and create the shares

Thank you, that worked!

25 minutes ago, arsaurea said:

Thank you, that worked!

Great, it's still a bug, but should be fixed for next release.

This is

Quote

There is a conflict with recent releases of Docker, ZFS, and the Linux Kernel. On Settings > Docker, we recommend that you use a Docker image rather than a Docker directory. If you choose to use a directory, avoid placing it on a ZFS pool (XFS or BTRFS are fine).

...is planned to be fixed before release? I hope so.

 

9 minutes ago, NLS said:

...is planned to be fixed before release? I hope so.

 

It's already been fixed, release notes need to be corrected, you can now use a docker directory on zfs without issues after upgrading to beta.3, just make sure the storage driver is set to overlay2 (default for new installs).

3 minutes ago, JorgeB said:

 

It's already been fixed, release notes need to be corrected, you can now use a docker directory on zfs without issues after upgrading to beta.3, just make sure the storage drive is set to overlay2 (default for new installs).

Ah thanks! How do I check what the drive is set? Can be changed?

4 minutes ago, NLS said:

Ah thanks! How do I check what the drive is set? Can be changed?

 

Sorry, corrected typo above, it's the docker storage driver, you can choose between native and overlay2, with a docker directory on zfs, always use overlay2 (default)

 

Noticed tonight that some of my docker containers used UTC and some CEST (system set to CEST, Europe/Berlin).

System time in dashboard correct.
I noticed it when some of my home automation fired off two hours late. This could have happened before, just me not noticing it.
Last system reboot was about 9h ago.

IF I use the TZ variable in any template, I use CEST and always the same so it shouldn't matter.

Don't know if bug or just some temp problem for me. Will open bug report (if not already reported?) if it happens again but I just want to mention it.

Edited October 7 by Niklas

When I woke up this morning, the web GUI displayed all the boxes/panels in the Dashboard, but they were blank except for the one showing Docker containers. The Main tab also had empty boxes. Array Devices, Pool Devices, etc., were all empty. I restarted nginx, and everything came back. Maybe some plugin or something. Monitoring.

Edited October 8 by Niklas

After upgrading from 6.12.13 to 7.0.0 Beta 3 I am getting the following error (Wrong Pool State - invalid config):

 

The pool contains the docker and appdata folders.

 

Diagnostics is attached.

 

*** Update: Rolling back to 6.12.13 worked fine.

 

 

tower-diagnostics-20241008-1605.zip

Edited October 8 by t3chn0m4g3

On 10/8/2024 at 4:11 PM, t3chn0m4g3 said:

After upgrading from 6.12.13 to 7.0.0 Beta 3 I am getting the following error (Wrong Pool State - invalid config):

 

The pool contains the docker and appdata folders.

 

Diagnostics is attached.

 

*** Update: Rolling back to 6.12.13 worked fine.

 

 

tower-diagnostics-20241008-1605.zip

Sounds like this bug (read the PS, it looks like your pool has 2 devices btrfs):

 

Edit: This happened to me for my pool that has 1 device (XFS).
It did not happen for the pool with 2xSSD (ZFS),

Edited Sunday at 03:11 PM by Niklas

@Niklas Thanks, will give it a try on the weekend.

1 hour ago, t3chn0m4g3 said:

will give it a try on the weekend.

It is that issue, re-importing the pool should fix it.

19 hours ago, JorgeB said:

 

Sorry, corrected typo above, it's the docker storage driver, you can choose between native and overlay2, with a docker directory on zfs, always use overlay2 (default)

 

Unfortunately, this option still has caveats. See my own report below, which is due to using Docker with ZFS folder and the overlay2 driver. Something causes this to happen. I'm back to the XFS image and it's a lot slower, but at least I'm consistently hitting the expected C-states again, much like I did on 6.12.13.

Additionally, it seems stopping Docker and starting it again is leaving leftover network routes:

 

This is causing containers not to be able to go online.

Just now, andrebrait said:

Unfortunately, this option still has caveats. See my own report below, which is due to using Docker with ZFS folder and the overlay2 driver.

Lets see if there are other reports of the same.

 

Just now, andrebrait said:

Additionally, it seems stopping Docker and starting it again is leaving leftover network routes:

@ich777any idea about this?

1 hour ago, andrebrait said:

Additionally, it seems stopping Docker and starting it again is leaving leftover network routes:

Do you already have a bug report for that?

 

AFAIK, stopping and starting docker won't clean up your network routes because they are saved by Docker.

Can you please create a dedicated bug report for that and mention me? Please also tell me what your expected behaviour is.
 

However it could take me a few days to respond.

11 hours ago, JorgeB said:

It is that issue, re-importing the pool should fix it.

@JorgeB just a quick question around BTRFS pools and this error/issue.On my production server I have a raid 1 pool for my cache of two BTRFS formatted ssd drives. My docker app data is also on this cache pool (behind a .apps directory). If I move from v6.12.13 to V7B3 will I be affected by this issue or other for that matter?  Thx

Edited October 9 by jim80z

11 hours ago, ich777 said:

Do you already have a bug report for that?

 

AFAIK, stopping and starting docker won't clean up your network routes because they are saved by Docker.

Can you please create a dedicated bug report for that and mention me? Please also tell me what your expected behaviour is.
 

However it could take me a few days to respond.

Done so: https://forums.unraid.net/bug-reports/prereleases/700-beta3-disabling-and-enabling-docker-leaves-behind-old-network-routes-containers-cant-connect-to-the-internet-r3283/

3 hours ago, jim80z said:

If I move from v6.12.13 to V7B3 will I be affected by this issue or other for that matter?  Thx

Most likely with a btrfs raid1 pool, would need to see the diags to confirm, but it's easy to resolve, you just need to re-import the pool as described here.

I didnt expect that my hardware gets an upgrade 🤣

28 minutes ago, Mainfrezzer said:

I didnt expect that my hardware gets an upgrade 🤣

Nice. You can keep everything in the cpu cache and skip the drives. Should be quite fast but don't lose power or reboot. 

 

L1-Cache: 256 KiB, L2-Cache: 1 MiB, L3-Cache: 8 MiB 

Edited Saturday at 11:49 AM by Niklas

21 minutes ago, Mainfrezzer said:

I didnt expect that my hardware gets an upgrade 🤣

Can you post output of dmidecode -q -t 7

1 hour ago, SimonF said:

Can you post output of dmidecode -q -t 7

Was already looking where the wrong number is from. There it is

Because lscpu and cpu info did display the correct size

Edit:
Seems to be a funny bug with the platform. It reads my Ryzen system just fine.

LSCPU for comparison.
ASRock - J4125-ITX
 

Architecture:                         x86_64
CPU op-mode(s):                       32-bit, 64-bit
Address sizes:                        39 bits physical, 48 bits virtual
Byte Order:                           Little Endian
CPU(s):                               4
On-line CPU(s) list:                  0-3
Vendor ID:                            GenuineIntel
BIOS Vendor ID:                       Intel
Model name:                           Intel(R) Celeron(R) J4125 CPU @ 2.00GHz
BIOS Model name:                      Intel(R) Celeron(R) J4125 CPU @ 2.00GHz Fill By OEM CPU @ 2.0GHz
BIOS CPU family:                      15
CPU family:                           6
Model:                                122
Thread(s) per core:                   1
Core(s) per socket:                   4
Socket(s):                            1
Stepping:                             8
CPU(s) scaling MHz:                   94%
CPU max MHz:                          2700.0000
CPU min MHz:                          800.0000
BogoMIPS:                             3993.60
Flags:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg cx16 xtpr pdcm sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave rdrand lahf_lm 3dnowprefetch cpuid_fault cat_l2 cdp_l2 ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust smep erms mpx rdt_a rdseed smap clflushopt intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves dtherm ida arat pln pts vnmi umip rdpid md_clear arch_capabilities
Virtualization:                       VT-x
L1d cache:                            96 KiB (4 instances)
L1i cache:                            128 KiB (4 instances)
L2 cache:                             4 MiB (1 instance)
NUMA node(s):                         1
NUMA node0 CPU(s):                    0-3

 

Edited Saturday at 01:04 PM by Mainfrezzer