Not sure if its something to do with the update, But i cant check for plugin updates, page never loads & I get the following error

Sep 20 17:23:24 Poweredge nginx: 2018/09/20 17:23:24 [error] 6962#6962: *1736 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 10.1.0.6, server: , request: "GET /plugins/dynamix.plugin.manager/include/ShowPlugins.php?check=0 HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "10.0.0.222", referrer: "http://10.0.0.222/Plugins"

 

Did anyone else have their bridge configuration disappear? I had a few dockers with fixed IP addresses and they suddenly lost Bridge:0. I'm assuming that this happened with the new bridge utils I saw mentioned in the package updates.

 

I had to undo and redo the settings after backing down to 6.5.3. 

 

Love the new UI look though.

 

I posted a bug report.

10 minutes ago, ezhik said:

Upgraded from 6.5.3 to 6.6.0 and my NEW USB is detected as my Replaced (Blacklisted)... HOW?!

The answer to your question can be obtained by the process in the message body.  Be sure that you include the pertinent details on your USB drive.

6 hours ago, FreeMan said:

Only one real issue I noticed:

This happens due to the preclear plugin. The latest version of the preclear plugin should resolve it, otherwise uninstall this plugin to get a proper display.

5 hours ago, clowrym said:

Not sure if its something to do with the update, But i cant check for plugin updates, page never loads & I get the following error

dev-pack plugin is known to cause such issues, uninstall it.

4 hours ago, BCinBC said:

Did anyone else have their bridge configuration disappear? I had a few dockers with fixed IP addresses and they suddenly lost Bridge:0.

Do you mean "br0" or did you define your own custom network "bridge.0" ?

The newer version of Docker included in Unraid 6.6 has changes in the networking part. We adopted so networks generated by Unraid continue to work as before, but there is no guarantee for custom networks made by the user.

This happens due to the preclear plugin. The latest version of the preclear plugin should resolve it, otherwise uninstall this plugin to get a proper display.

Excellent! Preclear updated and all is good.

Sent from Tapatalk

Minor question - I was on 6.6.0-RC4 with the Next branch selected in the update options.  This stable relase is NOT flagged as an available update in the UI in this case - I learned about it from the forum.  Should it have been flagged in the UI as available and then the Next/Stable switch reset to Stable after updating? Or even just flagged as available (leaving the switch where it was)?

 

Edit - I just realised the notification was set to "Once per day" - probably the window to the next notification had not yet elapsed.  Therefore this is probaby a non-issue.

The update went OK, BTW.  Thanks for all the hard work.  Unraid continues to deliver. 

Edited September 21, 20187 yr by S80_UK
Correction

Upgraded to 6.6.0 from 6.5.3 via the browser with no problems.

Nice job guys!

2 hours ago, S80_UK said:

Should it have been flagged in the UI as available

When the "next" branch is selected, it will only report about new releases in the "next" branch.

Select "stable" in the GUI and it will search for new "stable" versions and tell you a new "stable" version is available.

 

The opposite is also true. When on "stable" release, you need to select "next" to get new releases in the "next" branch.

Edited September 21, 20187 yr by bonienl

7 minutes ago, bonienl said:

When the "next" branch is selected, it will only report about new releases in the "next" branch.

Select "stable" in the GUI and it will search for new "stable" versions and tell you a new "stable" version is available.

That should be changed though IMO.  When on Next if a stable version becomes available, the system should automatically pick that up.  Otherwise you wind up with people seemingly forever stuck on a RC when a Stable is available.

3 minutes ago, Squid said:

That should be changed though IMO.  When on Next if a stable version becomes available, the system should automatically pick that up.  Otherwise you wind up with people seemingly forever stuck on a RC when a Stable is available.

Actually, I think I am wrong. A notice is given when a new stable version becomes available...

Just now, bonienl said:

Actually, I think I am wrong. A notice is given when a new stable version becomes available...

It actually didn't for me when stable was released, and I did have to switch from Next to Stable for it to see the update (and I didn't have a chance to upgrade to Stable until last night)  I know that under 6.5 Next it did pick up that Stable was available no problems.

I wish the notes would state whether Zenstates fix is still needed for Ryzen or Threadripper processors....

Updated both of my servers without any issues. Looks great! Keep up the good work!  😃

8 hours ago, bonienl said:

dev-pack plugin is known to cause such issues, uninstall it.

I dont believe I have that one....

poweredge-diagnostics-20180921-0803.zip

2 hours ago, Squid said:

It actually didn't for me when stable was released, and I did have to switch from Next to Stable for it to see the update (and I didn't have a chance to upgrade to Stable until last night)  I know that under 6.5 Next it did pick up that Stable was available no problems.

When publishing a stable I usually clear out the previous next series which results in Check For Updates seeing a new stable.  Thought I'd leave as-is in case someone wanted a previous -rc.

1 hour ago, methanoid said:

I wish the notes would state whether Zenstates fix is still needed for Ryzen or Threadripper processors....

I think it is, though some report a certain power setting in newer bios also fixes it; search internet for "redhat ryzen bug" and join the rest of us confused by what works and what doesn't.  What you won't find is anything from AMD acknowledging the existence of the problem at all.

I am in the process of downgrading as the new version decided to put both my graphics cards in the same IOMMU group. There was no way to split them, i have been trying for 2-3 hours now :-(

had a problem with three containers, when trying to start them it gave me the "execution error: Containers do not exist". It was really odd since the app folder for each container was still there. To fix it i made a change in the template like erasing the port number and typing it back in. It pull the image again and they started to work. Only dockers affected were radarr, sonarr, and unifi.

1 minute ago, ijuarez said:

had a problem with three containers, when trying to start them it gave me the "execution error: Containers do not exist". It was really odd since the app folder for each container was still there. To fix it i made a change in the template like erasing the port number and typing it back in. It pull the image again and they started to work. Only dockers affected were radarr, sonarr, and unifi.

Has your cache drive filled up recently by any chance?  I'm running two of these and upgraded without issue, but I've had similar issues before when I've overloaded my cache.

 

Note, the containers live in your docker.img file - the presence of the data folder isn't proof that the dockers are present and correct.  Might be worth ticking the Advanced switch on the dockers page and seeing whether you've got any orphaned images that can be deleted.

The message "execution error: Containers do not exist" is a bit misleading. It occurs when a container image is corrupted in some way. Basically, the old (corrupted) image was deleted, and the new image resolved it.

1 minute ago, Cessquill said:

Has your cache drive filled up recently by any chance?  I'm running two of these and upgraded without issue, but I've had similar issues before when I've overloaded my cache.

 

Note, the containers live in your docker.img file - the presence of the data folder isn't proof that the dockers are present and correct.  Might be worth ticking the Advanced switch on the dockers page and seeing whether you've got any orphaned images that can be deleted.

will check the orphaned images, docker image is 100GB (why not) its currently using 10G . thanks for the info will check it out.

1 minute ago, bonienl said:

The message "execution error: Containers do not exist" is a bit misleading. It occurs when a container image is corrupted in some way. Basically, the old (corrupted) image was deleted, and the new image resolved it.

AH...

 

Thank you

11 minutes ago, jamesyt666 said:

I am in the process of downgrading as the new version decided to put both my graphics cards in the same IOMMU group. There was no way to split them, i have been trying for 2-3 hours now 😞

I just downgraded, Enabled ACS overrride and restarted and now all my devices are split apart again. It seems a bit odd that the new version does not work the same way...