[SOLVED]Can't Install/Update Docker Images anymore

[aetek5]

Hi dear community,

 

in another thread I already described that I am having a network issue in my network and I don't know how to resolve it.

 

Anytime I am trying to install/update a docker container now I get following messages:

time="2020-11-30T15:11:13.984857499+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/rutorrent/manifests/latest: dial tcp 107.23.149.57:443: i/o timeout"
time="2020-11-30T15:12:25.833283901+01:00" level=error msg="Not continuing with pull after error: error pulling image configuration: Get https://registry-1.docker.io/v2/linuxserver/radarr/blobs/sha256:7d48a2cfcf77cf35346d7af3fadd99fd3c01c475fc6915a50727bb8b9f138500: dial tcp 52.1.121.53:443: i/o timeout"
time="2020-11-30T15:13:36.619321805+01:00" level=error msg="Not continuing with pull after error: error pulling image configuration: Get https://registry-1.docker.io/v2/linuxserver/rutorrent/blobs/sha256:d0b7d1021af16081266b04aedfc62b4887e384a600673a3d37f04f5d48ec8b0b: dial tcp 52.54.232.21:443: i/o timeout"
time="2020-11-30T15:17:10.387066410+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/radarr/manifests/latest: dial tcp 35.174.73.84:443: i/o timeout"
time="2020-11-30T15:17:26.616190119+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/rutorrent/manifests/latest: Get https://auth.docker.io/token?scope=repository%3Alinuxserver%2Frutorrent%3Apull&service=registry.docker.io: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)"
time="2020-11-30T15:19:50.252717636+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/radarr/manifests/latest: Get https://auth.docker.io/token?scope=repository%3Alinuxserver%2Fradarr%3Apull&service=registry.docker.io: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)"
time="2020-11-30T15:23:24.974569489+01:00" level=error msg="Not continuing with pull after error: error pulling image configuration: Get https://registry-1.docker.io/v2/linuxserver/radarr/blobs/sha256:7d48a2cfcf77cf35346d7af3fadd99fd3c01c475fc6915a50727bb8b9f138500: dial tcp 54.85.107.53:443: i/o timeout"
time="2020-11-30T15:27:04.392950433+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/rutorrent/manifests/latest: Get https://auth.docker.io/token?scope=repository%3Alinuxserver%2Frutorrent%3Apull&service=registry.docker.io: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)"
time="2020-11-30T15:34:06.269923069+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/radarr/manifests/latest: dial tcp 54.85.56.253:443: i/o timeout"
time="2020-11-30T15:35:57.794637496+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/rutorrent/manifests/latest: Get https://auth.docker.io/token?scope=repository%3Alinuxserver%2Frutorrent%3Apull&service=registry.docker.io: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)"
time="2020-11-30T15:46:19.070829097+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)"
time="2020-11-30T16:26:20.495384185+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)"
time="2020-11-30T18:53:30.690706066+01:00" level=error msg="Not continuing with pull after error: error pulling image configuration: Get https://registry-1.docker.io/v2/linuxserver/radarr/blobs/sha256:7d48a2cfcf77cf35346d7af3fadd99fd3c01c475fc6915a50727bb8b9f138500: dial tcp 52.20.56.50:443: i/o timeout"
time="2020-11-30T18:55:07.177314229+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/radarr/manifests/latest: dial tcp 54.236.131.166:443: i/o timeout"
time="2020-11-30T18:56:54.829101582+01:00" level=error msg="Handler for POST /v1.37/images/create returned error: Get https://registry-1.docker.io/v2/linuxserver/rutorrent/manifests/latest: dial tcp 35.174.73.84:443: i/o timeout"

Anyone had this issue before? Is it still an DNS issue?

 

Thank you so much for your time and help

ts140-diagnostics-20201130-1906.zip

Edited December 11, 2020 by aetek5

[Squid]

4 hours ago, aetek5 said:

Is it still an DNS issue?

It certainly won't hurt to try DNS addresses instead of using your router (eg: 208.67.220.220 / 208.67.222.222)

[aetek5]

Hey Squid,

thank you for your time and help. I change the network settings from my unraid to automatically get the IP from the Router and change the DNS to 1.1.1.1 in the router. After restart and reconnect I can update/install docker apps again. But I would love to have an static ip for the server to avoid that the ip changes everytime.

 

I try again some tinkering in the network settings why the behavior is so strange.

 

Thanks a lot.

[Squid]

Yeah, so set a static IP address for it (either in Network settings, or in the router).  I suggested simply setting static DNS in Network Settings.

[aetek5]

I did the following. I set the unraid server static DNS and the Router DNS to the Pihole IP. Everything is now working fine right now. It seems that setting the pihole server address as DNS in the UnRaid Server Network settings during the changes and the docker containers not running that it creates some sort of confusion. I will leave it like this for the moment.

 

Thank you for your time

[Squid]

By piHole, I'm assuming you're running that on your server.  You should also have your router set to a secondary public DNS server, as if you turn off the server (or stop the array), then everything connected to the router has lost its internet access.

[aetek5]

I thought that too that this could be an issue. But in the Pihole doc it is mentioned not to use a second DNS server in the router. Funny enough, everytime I stopped the docker and with that pihole, my router still had internet access and my pc too.

 

I am monitoring very weird behavior right now. I changed from the pihole to a public DNS on my router so everything what would be connected should work. But I tested different Ping and nslookup and got some timeouts here and there. The only thing left what could cause issue is my ISP. I checked some website and many are currently reporting a lot off internet issues. So with all horror the issue was not with my network but with my ISP.

I have PiHole set up on the router and UnRaid running on a public DNS and still get some DNS timeouts with the UnRaid Server and with my Router. But it works for now and I hope it was just the ISP having some issues.

 

Thanks a lot for your time and help.

[aetek5]

UPDATE:

 

For anyone having issues, I found the topic for Tips and Tweaks and corresponding wiki which mentions of two network settings to try to disable them:

 

https://wiki.unraid.net/index.php/Tips_and_Tweaks#Disable_network_offloading

 

The options are Disable network offloading and Disable network flow control. After disabling I don't occur to have the DNS issues after 24 hours.

[faulksy]

On 12/12/2020 at 8:10 AM, aetek5 said:

UPDATE:

 

For anyone having issues, I found the topic for Tips and Tweaks and corresponding wiki which mentions of two network settings to try to disable them:

 

https://wiki.unraid.net/index.php/Tips_and_Tweaks#Disable_network_offloading

 

The options are Disable network offloading and Disable network flow control. After disabling I don't occur to have the DNS issues after 24 hours.

Thanks a billion. My indexers in nzbhydra would constantly have DNS issues with the indexers and disable them. When trying to reinstall containers or even new ones they would fail. My nzbget download speed had also dropped by 80%. I suspected isp/rsp throttling. Disabled these NIC settings and everything seems okay so far. Issues only occurred after 6.9.x updates