[Plug-In] Community Applications


Recommended Posts

3 minutes ago, Wolfgang Weber said:

And legacy never stops. Even after 25 min.

 

22 hours ago, Squid said:

If it takes longer than two minutes, the "Updating" screen will never disappear, but it still does the thing in the background.  Give it say 10 minutes, then just reload the page and you're in business

 

Link to comment
22 minutes ago, rmilyard said:

So I have my DNS as static for both Google DNS server.  Clocks for awhile and when I click Legacy Mode just clocks forever.  I even let one server set over night.

 

23 hours ago, Squid said:

If it takes longer than two minutes, the "Updating" screen will never disappear, but it still does the thing in the background.  Give it say 10 minutes, then just reload the page and you're in business

 

Link to comment

Image result for hindsight

 

Never in my wildest dreams did I think that people were going to have problems with Legacy Mode downloads timing out on 6.4+  (I usually go by the assumption that everyone's hardware and internet connection is better than mine)  Nevertheless, the rather extended outage by Digital Ocean accelerated the need to slightly revamp how Legacy Mode downloads are handled.  There is now basically zero chance that niginx (unRaid) will ever timeout when switching to Legacy Mode.   Tested on Chrome, Edge, IE, Firefox.  Can someone running Safari go to CA's settings, enable developer mode, then hit Legacy Mode and confirm that it also works.  (You'll see the pop up individually download the various repositories instead of just the spinning wheel)

 

Other notable changes

  • In the event of an appfeed failure, CA will check for and display an applicable "Public Service Announcement" that will state if I know the feed is definitively down.
  • Appfeed download failure screen is more obvious that if nothing is wrong with your network settings to go to legacy mode.
  • Various PHP warnings that could have appeared when displaying stats on a appfeed failure without switching to Legacy Mode have been suppressed.
  • Download timeouts have been lowered from 60 seconds to 45.  
  • All new installs of CA now require you to actually accept the disclaimer before using the plugin.  
  • Like 1
  • Upvote 2
Link to comment

I updated the plugin earlier today and it all seems to have gone wrong.  The update process apparently went well but after it completed the Apps tab had gone and the plugin didn't display correctly in the plugin list.  I removed it and tried installing from the url in the first post but get the following:

 

Quote

 

plugin: installing: https://raw.githubusercontent.com/Squidly271/community.applications/master/plugins/community.applications.plg
plugin: downloading https://raw.githubusercontent.com/Squidly271/community.applications/master/plugins/community.applications.plg
plugin: downloading: https://raw.githubusercontent.com/Squidly271/community.applications/master/plugins/community.applications.plg ... done


Cleaning Up Old Versions


plugin: downloading: https://raw.github.com/Squidly271/community.applications/master/archive/community.applications-2018.04.02.txz ... done

+==============================================================================
| Installing new package /boot/config/plugins/community.applications/community.applications-2018.04.02.txz
+==============================================================================

Verifying package community.applications-2018.04.02.txz.
Unable to install /boot/config/plugins/community.applications/community.applications-2018.04.02.txz: tar archive is corrupt (tar returned error code 2)

plugin: skipping: ca-tooltipster.txz - unRAID version too high, requires at most version 6.3.2
Creating Directories

----------------------------------------------------
community.applications has been installed.
Copyright 2015-2018, Andrew Zawadzki
Version: 2018.04.02
----------------------------------------------------

plugin: installed
 

 

No matter what I do I can't get the plugin installed.  Any ideas?

Link to comment
3 hours ago, speedyyellow said:

I updated the plugin earlier today and it all seems to have gone wrong.  The update process apparently went well but after it completed the Apps tab had gone and the plugin didn't display correctly in the plugin list.  I removed it and tried installing from the url in the first post but get the following:

 

 

No matter what I do I can't get the plugin installed.  Any ideas?

First and foremost, you've got bigger fish to fry.  You're actively being hacked.  syslogs are nothing but attempts to log in.

 

Apr  3 04:41:15 Tower sshd[20618]: Failed password for root from 58.242.83.42 port 10376 ssh2

 

I suggest you close the ports you're not requiring, and take the server out of your DMZ, and then worry about other stuff.

  • Upvote 1
Link to comment
5 hours ago, Squid said:

First and foremost, you've got bigger fish to fry.  You're actively being hacked.  syslogs are nothing but attempts to log in.

 

 


Apr  3 04:41:15 Tower sshd[20618]: Failed password for root from 58.242.83.42 port 10376 ssh2

 

 

I suggest you close the ports you're not requiring, and take the server out of your DMZ, and then worry about other stuff.

 

Already taken care of that. Any ideas on the “corrupt archive” error message?

Link to comment

Not sure if anything changed in github but this morning I managed to get the plugin installed again.  I restored a version from the plugins-old-versions folder which got Unraid showing the plugin with an update available and hitting update worked properly this time.

 

Thanks for the assistance, and if nothing else the whole experience highlighted to me the ongoing hacking attempt.  I only opened up ssh access a few days ago to try and get some docker stuff going while I was at work.  I'll keep that shut off from now on I think.

Link to comment
On 4/1/2018 at 5:24 PM, Squid said:

A button appears called Legacy Mode (top right) which will let you use CA  (The note is there on the screen telling you the feed failed).  Reason CA doesn't automatically fall back is the majority of times the problem is on the user's end (DNS settings, etc) and I'd far rather have them fix that than have Legacy active, since Legacy will take ~30-60 seconds ish to start up (and if you have a really crappy internet speed, then it could take minutes and after 2 minutes, mginx in 6.4+ will automatically time out).  All vs appFeed mode which will take worst case ~10 seconds to get things running

So I fall into the above case with terrible internet, if I move to next build is there a stable way to move to the current mode instead of legacy so that I wont get the blank apps screen because of a timeout? Also why does my installed apps list fail out when trying to display (I understand if it is trying to pull updates and times out because of poor connection).

Link to comment
On 4/1/2018 at 5:24 PM, Squid said:

A button appears called Legacy Mode (top right) which will let you use CA  (The note is there on the screen telling you the feed failed).  Reason CA doesn't automatically fall back is the majority of times the problem is on the user's end (DNS settings, etc) and I'd far rather have them fix that than have Legacy active, since Legacy will take ~30-60 seconds ish to start up (and if you have a really crappy internet speed, then it could take minutes and after 2 minutes, mginx in 6.4+ will automatically time out).  All vs appFeed mode which will take worst case ~10 seconds to get things running

So I fall into the above case with terrible internet, if I move to next build is there a stable way to move to the current mode instead of legacy so that I wont get the blank apps screen because of a timeout? Also why does my installed apps list fail out when trying to display (I understand if it is trying to pull updates and times out because of poor connection).

Link to comment
1 hour ago, MBMWR said:

So I fall into the above case with terrible internet, if I move to next build is there a stable way to move to the current mode instead of legacy so that I wont get the blank apps screen because of a timeout? Also why does my installed apps list fail out when trying to display (I understand if it is trying to pull updates and times out because of poor connection).

You pretty much have to be on dial-up for the appfeed to ever suffer from a download timeout.  But, after 45 seconds trying to get the appfeed it's going to assume that it's not going to be able to get it, and the ask you to switch to legacy mode.

 

Legacy mode may take a couple of minutes on a very slow download speed, but it will never time issue an nginx timeout.  Rather it spends 45 seconds trying to download each repository.  If that repository times out, then the apps available within it will not be available.

 

CA will automatically switch back to appfeed mode when an update is available from the appfeed.  If you want to switch it back before that, then you can just press appFeed Mode in the top right.

 

CA never actually checks for updates for any installed container or plugin.  Rather it utilizes what dynamix says is available (if you have dynamix in its Notification settings checking for updates on a schedule).  For why CA timed out, is there something logged at that time.  If there is (nginx timeout), then this is a problem some users have (not me) on 6.5.  You're best off installing 6.5.1-rc3+ to alleviate that.

  • Like 1
Link to comment
On 4/3/2018 at 10:13 PM, speedyyellow said:

I updated the plugin earlier today and it all seems to have gone wrong.  The update process apparently went well but after it completed the Apps tab had gone and the plugin didn't display correctly in the plugin list.  I removed it and tried installing from the url in the first post but get the following:

 

 

No matter what I do I can't get the plugin installed.  Any ideas?

I'm getting the exact same thing from the last CA update (18-04-08). I have not tried reinstalling it however.

Also, my logs seem to have ended on 08? Both the syslog and dockerlog? The server's been online 24/7

 

unraid-diagnostics-20180410-0452.zip

Edited by Andreen
Add info
Link to comment
44 minutes ago, trurl said:

Your log space is completely full but it's not clear why, since syslog isn't that large. What else do you have in /var/log?

 

 

Apparently an atop log of 127M.......

 

du.txt

 

Edit: It's from one day too, too big to open, I've deleted it :S

 

Edit 2: I opened /var/log/atop/atop_20180326 and it wasn't readable by a text editor. Anyone know what it is? How can I prevent this from happening without removing atop? atop comes from the Nerd Pack Plugin

 

Edit 3: Actually I'm just gonna remove atop, not worth it. Thanks for pointing me in the right direction, trurl! :)

Edited by Andreen
Link to comment

I installed CA yesterday and tried to install a few docker containers. After clicking the install button to install a docker container, the page was blank. I tried a few other times with the same result and nothing installed. I tried uninstalling CA, reinstalling CA, rebooting the server and always had the same result. This shows up in my syslog when trying to install a docker:

 

Apr 10 19:23:22 FileSvr root: error: /Apps/AddContainer?xmlTemplate=default:/tmp/community.applications/tempFiles/templates-community-apps/linuxserversRepository/duckdns.xml: missing csrf_token

 

CA installs plugins fine as I tried that with FCP and ran that to see if I had any errors. Nothing was found except a few minor warnings. I made sure that no other windows or apps were connected to unRAID as I know that can cause problems with csrf tokens.

 

One other note is that i tired with 3 different browsers, Firefox, Edge, and Chrome.

 

Anyways, here are my diags. Thanks for the help.

filesvr-diagnostics-20180411-0512.zip

Edited by GHunter
Link to comment
6 minutes ago, GHunter said:

I installed CA yesterday and tried to install a few docker containers. After clicking the install button to install a docker container, the page was blank. I tried a few other times with the same result and nothing installed. I tried uninstalling CA, reinstalling CA, rebooting the server and always had the same result. This shows up in my syslog when trying to install a docker:

 


Apr 10 19:23:22 FileSvr root: error: /Apps/AddContainer?xmlTemplate=default:/tmp/community.applications/tempFiles/templates-community-apps/linuxserversRepository/duckdns.xml: missing csrf_token

 

CA installs plugins fine as I tried that with FCP and ran that to see if I had any errors. Nothing was found except a few minor warnings. I made sure that no other windows or apps were connected to unRAID as I know that can cause problems with csrf tokens.

 

One other note is that i tired with 3 different browsers, Firefox, Edge, and Chrome.

 

Anyways, here are my diags. Thanks for the help.

filesvr-diagnostics-20180411-0512.zip

If you happen to have bitdefender, you might try momentarily disabling it, then try to do a docker install.  If it works, add the ip of your unraid as an exception. (I ended up switching to something other than bitdefender because it became so disagreeable. Which is too bad because it has been my preference for quite a while) 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.