When manually Checking for Updates it is taking around 37 seconds to check 6 Plugins and 24 seconds to check 2 Docker Containers. Around a month or so ago it was fast and only took about 5 seconds but something changed during that time and it is consistently slow now. From searching the forum I believe it may be a DNS issue but I don't know how to troubleshoot it.

Please could someone check my Diagnostics and advise what I should check or alter please?

nas-plex-diagnostics-20240816-1612.zip

Edited August 16, 20241 yr by DanielCoffey
typo

Nothing obvious logged, is it the same in safe mode? Also try changing your DNS server to 208.67.222.222 and 208.67.220.220 instead of using your router.  Settings - Network Settings 

I stopped the Docker process and edited the Network Settings to point to those OpenDNS addresses and rebooted the server but nothing changed.

I entered Safe Mode, started the server and checked the time to update the Dockers and it was unchanged at 24s.

I have reverted to the normal unraid mode and rebooted the server. but there is still no change - 36s to check plugins for update.

Is you internet working normally, if you do for example a speed test with a docker?

Which speed test docker would you recommend with a simple setup please? I have CA but there are a few to choose from.

Sorry, never used any, but they should be similar.

Just an update. I have reset back to 8.8.8.8 and 8.8.4.4 for my DNS and rebooted both my server and router/switch. I am still getting slow responses for each plugin when checking for updates.

When I open a command prompt in the UI and ask unraid to tell me the dns server it responds fine. Ping works quickly too. I still don't know why the plugins are so slow to check for updates however.

I am still having issues with slow plugin checking and would appreciate some thoughts on the cause.

I can add that it is consistent, 24s to check two Dockers and 36s for 6 plugins. I can ping from the server to Google DNS and get 16ms responses as normal. Asking for OS Updates is instant and fast. Docker Updates are fast. It is just the checking that is slow.

 

Could it be an issue of my ISP resolving just the checking server? Is there an address I can ping to test it?

Try booting a flash drive with a stock Unraid install, no key needed, install the same plugins and retest, if it's faster it's a config issue, if it's the same, it's an external problem.

Thank you for the idea. Are there instructions for doing this without messing up my main flash drive (which I have just remembered is installed internally on a header... means pulling out the box to get at it)?

I have already loaded unraid in safe mode before and it made no difference.

You just remove your flash drive and connect a new one.

My friend have similar problem, sometime plugin update fail ( haven't docker ). And his Internet also unstable in random time ( sometime were normal, we use same ISP but in different location )

A simple way to use Mobile network as alternate ISP for testing was connect a USB ethernet adapter to Android phone act as ISP, useful to identify problem.

Edited September 1, 20241 yr by Vr2Io

Update 03 September : I have downloaded and created a Trial USB stick and added the exact same plugins. There was no change to the time taken to Check for Updates. Still 36 seconds for 6 plugins. That hopefully eliminates anything inside the UNRAID box as the issue.

 

I don't have any easy way to connect to a wifi hotspot without reconfiguring my router as an access point and totally messing my home network.

 

Is there a listing of the server address that the Check for Updates contacts so that I can ping and tracert to it from the UNRAID box?

Sorry, don't know that

4 hours ago, DanielCoffey said:

Update 03 September : I have downloaded and created a Trial USB stick and added the exact same plugins. There was no change to the time taken to Check for Updates. Still 36 seconds for 6 plugins. That hopefully eliminates anything inside the UNRAID box as the issue.

 

I don't have any easy way to connect to a wifi hotspot without reconfiguring my router as an access point and totally messing my home network.

 

Is there a listing of the server address that the Check for Updates contacts so that I can ping and tracert to it from the UNRAID box?

ping hub.docker.com and raw.githubusercontent.com

Now that was very revealing, thank you.

 

My unraid server seems to be set up for IPv4 but my Windows machine is IPv6 so I get different results depending which machine I ping/traceroute from.

 

hub.docker.com is 100% packetloss on both machines since I suspect the server does not respond to pings. Both machines get partial results from traceroute/tracert. Each line of traceroute is slow to appear, regardless of which machine I run it on.

raw.githubusercontent.com pings perfectly on both machines but Unraid is slow to start pinging. With traceroute, the Windows PC gets most of the hops normally but Unraid is again slow to start the traceroute, loses one, picks up again and then drops the thread.

Any thoughts as to the issue? If told how, I could attempt to turn on IPv6 on the Unraid box.

12 minutes ago, DanielCoffey said:

Now that was very revealing, thank you.

 

My unraid server seems to be set up for IPv4 but my Windows machine is IPv6 so I get different results depending which machine I ping/traceroute from.

 

hub.docker.com is 100% packetloss on both machines since I suspect the server does not respond to pings. Both machines get partial results from traceroute/tracert. Each line of traceroute is slow to appear, regardless of which machine I run it on.

raw.githubusercontent.com pings perfectly on both machines but Unraid is slow to start pinging. With traceroute, the Windows PC gets most of the hops normally but Unraid is again slow to start the traceroute, loses one, picks up again and then drops the thread.

Any thoughts as to the issue? If told how, I could attempt to turn on IPv6 on the Unraid box.

Sorry should have check hub pings.

 

How quickly does unraid resolve names? Mine returns straight away.

 

root@computenode:~# nslookup hub.docker.com
Server:         192.168.1.253
Address:        192.168.1.253#53

Non-authoritative answer:
hub.docker.com  canonical name = elb-default.us-east-1.aws.dckr.io.
elb-default.us-east-1.aws.dckr.io       canonical name = prodextdefgreen-k0uuibjyui-4ec6503f7037d339.elb.us-east-1.amazonaws.com.
Name:   prodextdefgreen-k0uuibjyui-4ec6503f7037d339.elb.us-east-1.amazonaws.com
Address: 44.219.3.189
Name:   prodextdefgreen-k0uuibjyui-4ec6503f7037d339.elb.us-east-1.amazonaws.com
Address: 44.193.181.103
Name:   prodextdefgreen-k0uuibjyui-4ec6503f7037d339.elb.us-east-1.amazonaws.com
Address: 3.224.227.198
Name:   prodextdefgreen-k0uuibjyui-4ec6503f7037d339.elb.us-east-1.amazonaws.com
Address: 2600:1f18:2148:bc00:a518:1574:fea8:ccdb
Name:   prodextdefgreen-k0uuibjyui-4ec6503f7037d339.elb.us-east-1.amazonaws.com
Address: 2600:1f18:2148:bc02:4f78:625e:1ac1:714c
Name:   prodextdefgreen-k0uuibjyui-4ec6503f7037d339.elb.us-east-1.amazonaws.com
Address: 2600:1f18:2148:bc01:2822:9e2a:c82a:af0

 

root@computenode:~# nslookup raw.githubusercontent.com
Server:         192.168.1.253
Address:        192.168.1.253#53

Non-authoritative answer:
Name:   raw.githubusercontent.com
Address: 185.199.110.133
Name:   raw.githubusercontent.com
Address: 185.199.111.133
Name:   raw.githubusercontent.com
Address: 185.199.108.133
Name:   raw.githubusercontent.com
Address: 185.199.109.133
Name:   raw.githubusercontent.com
Address: 2606:50c0:8003::154
Name:   raw.githubusercontent.com
Address: 2606:50c0:8002::154
Name:   raw.githubusercontent.com
Address: 2606:50c0:8000::154
Name:   raw.githubusercontent.com
Address: 2606:50c0:8001::154

root@computenode:~# 

Names lookup instantly too.

14 minutes ago, DanielCoffey said:

Names lookup instantly too.

I get similar results on EE to trace routes. My update checks are around 10secs.

I've been noticing this since updating to 6.12.13. Docker container update checks and plugin checks takes substantially longer than before, albeit I do not remember if it started immediately since it took a while before realizing update checks were consistently slower. Most everything else seems to be running as usual.
I have a stable and fast internet connection and anything outside of Unraid is seemingly unaffected and resolves DNS etc just fine (I use Pi-Hole to Cloudflare IPv4).

I too have this issue, both on 6.12.13 and after upgrading to 7.0.2 beta. My other (test) server on the same network/subnet and dns settings does not have the issue, checking updates for plugins and containers is fast as ever. Most annoying is that pulling a new app/container sometimes takes so long the process times out. I'm sure that earlier the problem wasn't there, it came more or less 'out of the blue' (for me at least;-). I might create a new usb and start from scratch. Thing is I had to transfer my Pro license already on account my new machine only boots UEFI and my USB key was created with legacy boot. So transferring twice in a couple of weeks...... don't know.

12 hours ago, rddz48 said:

So transferring twice in a couple of weeks...... don't know.

Note that you can just recreate the install on the same flash drive.