nas-unraid-bkp-diagnostics-20210302-0904.zip

 

OK looks like my Docker.img is on my cache pool and during the upgrade I lost my cache assignment so all Docker Containers stoped working.

The reason I lost my cache is probably due to an unassigned NVMe drive that was not part of the 2 X SSD 500GB drives cache pool. --> This may be a bug.

I am backup and running on this Unraid as well as my primary which upgrade went just fine (I only had 2 SSD drives on that cache)

 

 

 

Well excited about upgrading to 6.9 but disappointing that all my docker containers are gone?

 

I am reverting back to 6.8.X and I hope my stuff shows back up

 

 

Edited March 2, 20215 yr by johnwhicker

Well I restored back to 6.8.3 and docker contained still gone.  Now that's so disappointing    Any ideas how I get these back?

Your diagnostics could tell us more on what is happening.

6 minutes ago, ChatNoir said:

Your diagnostics could tell us more on what is happening.

Good point, I apologize.  Is now attached. 

Man what a MESS.  And just now noticed that all my cache drives are unsigned. SERIOUSLY?  How can an upgrade be so unreliable?  What else should I look for missing?  My level of confidence in this upgrade is tanked, sorry

 

 

 

Test are in the release notes but I think they need to be highlighted better.

 

Note: A pre-6.9.0 cache disk/pool is now simply a pool named "cache".  When you upgrade a server which has a cache disk/pool defined, a backup of config/disk.cfg will be saved to config/disk.cfg.bak, and then cache device assignment settings are moved out of config/disk.cfg and into a new file, config/pools/cache.cfg.  If later you revert back to a pre-6.9.0 Unraid OS release you will lose your cache device assignments and you will have to manually re-assign devices to cache.  As long as you reassign the correct devices, data should remain intact.

 

I am not sure how easy it is to add the cache back once the array had been started. @JorgeB do you have any suggestions.

I have asked for this to be put onto the main post for 6.9 announcement.

12 minutes ago, johnwhicker said:

And just now noticed that all my cache drives are unsigned. SERIOUSLY?

That is in the release notes, it will happen every time you go from v6.9 to v6.8.

 

Just stop array, re-assign all cache pool devices, there can't be a "all data on this device will be deleted" warning for any cache device, then start array.

7 minutes ago, JorgeB said:

That is in the release notes, it will happen every time you go from v6.9 to v6.8.

 

Just stop array, re-assign all cache pool devices, there can't be a "all data on this device will be deleted" warning for any cache device, then start array.

 

Thanks. Well my disappointment continues.  I have no idea what data and if any data in the cache drives.  How do I verify?  Should I manually mount them and see what's there?

 

How can an upgrade go so WRONG and be designed so poorly?  At least during the upgrade they could've give a us warning or a banner with some of the key things to watch for or do? I mean who reads the notes?    What a MESS

10 minutes ago, johnwhicker said:

I have no idea what data and if any data in the cache drives.  How do I verify?  Should I manually mount them and see what's there?

Not sure what you mean, you just need to re-assign the devices.

6 minutes ago, JorgeB said:

Not sure what you mean, you just need to re-assign the devices.

 

Thanks. So I re-assigned the cache drives and it looks like I got my Docker Containers back?

 

So what is the right procedure for the cache drives when upgrading to 6.9?  It looks like the Docker Containers disappearing are related to the cache somehow?

When upgrading the Cache just gets made in to a multipool config with the name of cache, this is why you have to do a manual action if you revert as multipools don't exist on 6.8.3.

 

You shouldn't need to do anything when going to 6.9 re containers. But sounds like your mapping to the image file was affected some how.  Does it reside on your cache?

31 minutes ago, SimonF said:

When upgrading the Cache just gets made in to a multipool config with the name of cache, this is why you have to do a manual action if you revert as multipools don't exist on 6.8.3.

 

You shouldn't need to do anything when going to 6.9 re containers. But sounds like your mapping to the image file was affected some how.  Does it reside on your cache?

 

You're onto something.  So yes the docker.img is on the cache drive.  The question is why I am loosing my cache drives assignment after the upgrade to 6.9 ?  It looks that's the reason for all Dockers not working.  How do I fix this?

 

 

Edited March 2, 20215 yr by johnwhicker

You don't lose the cache upgrading, or at least you shouldn't, you lose going back to v6.8, if after the upgrade there's no cache pool post new diags.

According to those 6.8.3 Diagnostics in OP, you seem to have 3 SSDs, one an NVMe different in size than the other 2. Were these all in cache pool or was one of them Unassigned?

2 minutes ago, JorgeB said:

You don't lose the cache upgrading, or at least you shouldn't, you lose going back to v6.8, if after the upgrade there's no cache pool post new diags.

 

Here it goes Sir.

 

 

nas-unraid-bkp-diagnostics-20210302-1011.zip

@trurl From the post John's post on 6.9 thread all 3 are in the cache. I am wondering if its the mix of an nmve has caused an issue with the upgrade process.

Edited March 2, 20215 yr by SimonF

2 minutes ago, trurl said:

According to those 6.8.3 Diagnostics in OP, you seem to have 3 SSDs, one an NVMe different in size than the other 2. Were these all in cache pool or was one of them Unassigned?

 

The NVMe was unassigned as I plan to take it out when I access this box.  Would the unassigned nvme cause this issue?

Screenshot in your latest post in 6.9 release thread shows all 3 as cache. Would make more sense to make the NVMe a separate pool.

3 minutes ago, trurl said:

Screenshot in your latest post in 6.9 release thread shows all 3 as cache. Would make more sense to make the NVMe a separate pool.

 

Is because I manually re-assigned the cache drives and I added the nvme as well.  But the point is the 6.9 upgrade doesn't migrate /retain the cache configuration to where it needs to be.  Could it be a bug?

 

In 6.8.3 I had 2 500Gb SSD forming the cache pool and the NVMe was unassigned

After 6.9.X the cache was gone which caused my dockers to disappear as I had my docker.img on the cache.  I ended up manually re-assigning all3 and it works fine now.

Edited March 2, 20215 yr by johnwhicker

following

14 minutes ago, johnwhicker said:

re-assigning all3

See my comment on your post in the release thread.

17 minutes ago, trurl said:

Would make more sense to make the NVMe a separate pool.

 

I think the unassigned NVMe was the problem. I just upgraded my primary UNRAID and the upgrade went fine.  I only have 2 SSD drives for the cache on this one.