moby: docker daemon unable to access registry - Client.Timeout exceeded while awaiting headers

I found out that the problem might be in /etc/resolv.conf. I had:

nameserver 10.0.0.10
nameserver 8.8.8.8
nameserver 8.8.4.4

but moving the non-working (yet) 10.0.0.10 nameserver to the bottom fixed it for me I think

nameserver 8.8.8.8
nameserver 8.8.4.4
nameserver 10.0.0.10

Setting the DNS server to fixed (8.8.8.8) from automatic in docker settings solved the problem for me.

I already hat MTU size of 1500 and still not working.

Apparently, enabling MTU probing in the kernel resolved the issue. To enable it just run:

 sudo sysctl -w net.ipv4.tcp_mtu_probing=1

To make this permanently, add the following line to /etc/sysctl.conf:

net.ipv4.tcp_mtu_probing=1

https://thesimplecomputer.info/pages/adventures-in-linux-tcp-tuning-page2

Exactly the same issue here.

@aasaanhai also worked for me. Simply set the default DNS to 8.8.8.8 on the “vEthernet (DockerNAT)” network adapter.

For Windows 10 Home, Putting 8.8.8.8 as the first dns resolves it.

I faced the same problem on a Windows 10 machine. To solve the problem I accessed the Docker client Settings screen, tab Network, and changed the DNS Server to fixed IP 8.8.8.8. After restarting docker everything start working properly.

I found out that the problem might be in /etc/resolv.conf. I had:

nameserver 192.168.0.1

after editing to

nameserver 8.8.8.8 nameserver 8.8.4.4 nameserver 10.0.0.10 This worked for me.

[ubuntu16.04, docker 18.01.0] In my case, I could solve by setting proxy like this. This way is different one from above.

$ sudo mkdir -p /etc/systemd/system/docker.service.d $ sudo vi /etc/systemd/system/docker.service.d/http_proxy.conf [Service] Environment=“HTTP_PROXY=http://<your_proxy_ip>:<port>/” $ sudo vi /etc/systemd/system/docker.service.d/https_proxy.conf [Service] Environment=“HTTPS_PROXY=http://<your_proxy_ip>:<port>/” $ sudo systemctl daemon-reload $ sudo systemctl restart docker

I got it solved by adding the proxy details in the docker config file /etc/default/docker.

@luxas I worked for my linux system, but not getting what to do on windows. Please help if possible.

Same problem faced running docker-machine on windows. Restarting docker-machine worked for me suggested by chenrui333. My docker version is 1.13.1

docker-machine restart

docker login as well as docker pull is working now.

I was getting this same error in Windows 10 Home using Docker Toolbox. Changing my DNS to Google’s 8.8.8.8 (secondary to 8.8.4.4) worked for me (per @aasaanhai above). Detailed instructions for how to do so here (see section II).

I’m also getting this issue:

Error getting v2 registry: Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)
Attempting next endpoint for pull after error: Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)
Not continuing with pull after error: Network timed out while trying to connect to https://index.docker.io/v1/repositories/armhf/busybox/images. You may want to check your internet connection or if you are behind a proxy.

docker info:

Containers: 0
 Running: 0
 Paused: 0
 Stopped: 0
Images: 0
Server Version: 1.12.1
Storage Driver: overlay
 Backing Filesystem: extfs
Logging Driver: json-file
Cgroup Driver: cgroupfs
Plugins:
 Volume: local
 Network: overlay null bridge host
Swarm: inactive
Runtimes: runc
Default Runtime: runc
Security Options: seccomp
Kernel Version: 4.4.19-1-ARCH
Operating System: Arch Linux ARM
OSType: linux
Architecture: armv7l
CPUs: 4
Total Memory: 922.2 MiB
Name: pi3arch
ID: 7MTA:UZIW:SVAS:H3U7:YBHE:FWXV:KWOS:4QAQ:HDLG:EWIV:DOAP:TWFR
Docker Root Dir: /var/lib/docker
Debug Mode (client): false
Debug Mode (server): false
Registry: https://index.docker.io/v1/
Insecure Registries:
 127.0.0.0/8

I’ve set the mtu to 1500 and net.ipv4.tcp_mtu_probing=1 with no success 😦 Any ideas?

Same issue here as reported by @luxas.

I found out that the problem might be in /etc/resolv.conf. I had:

nameserver 10.0.0.10
nameserver 8.8.8.8
nameserver 8.8.4.4

but moving the non-working (yet) 10.0.0.10 nameserver to the bottom fixed it for me I think

nameserver 8.8.8.8
nameserver 8.8.4.4
nameserver 10.0.0.10

This worked for me thanks

thanks @JohnVonNeumann i solved my issue. the message “Error response from daemon: Get https://registry-1.docker.io/v2/: net/http: request canceled (Client.Timeout exceeded while awaiting headers)” was caused by network connection with a proxy. with https://docs.docker.com/engine/admin/systemd/ (§ http proxy), it is ok. i am surprised docker use systemd on debian… and not environment variables. thanks again

Thank you luxas It works great!!!

I added below configuration on /etc/resolv.conf as suggested above and it works

nameserver 8.8.8.8 nameserver 8.8.4.4 nameserver 10.0.0.10 ​

Please refer to this: https://docs.docker.com/engine/admin/systemd/#httphttps-proxy I have resolved on Ubuntu 16.04 LTS

I’m getting this on Mac OS 10.12.5. I tried re-installing to the latest version of docker, and still no luck. Here’s the error I’m getting: Pulling wordpress (wordpress:latest)… ERROR: Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)

I already hat MTU size of 1500 and still not working.

Apparently, enabling MTU probing in the kernel resolved the issue. To enable it just run:

 sudo sysctl -w net.ipv4.tcp_mtu_probing=1

To make this permanently, add the following line to /etc/sysctl.conf:

net.ipv4.tcp_mtu_probing=1

https://thesimplecomputer.info/pages/adventures-in-linux-tcp-tuning-page2

Thank you… It fixed the issue 💯

docker login to private registry in our environment is failing intermittently. We are having same issue with private registry. Configuring Google’s DNS would work for docker.io but for internal registries how can we extend the login timeout value, esp with slow AD Authorizations.

We have configured our docker NGINX with proxy_ignore_client_abort on; to ignore client timeouts to verify if there is a server kick outs. But the client looks hard coded with 15s timeout irrespective of the server behaviour. Any way we can change/extend this timeout?

I added below configuration on /etc/resolv.conf as suggested above and it works

nameserver 8.8.8.8 nameserver 8.8.4.4 nameserver 10.0.0.10 ​

this one worked for me! thank you so much 😃

Thanks AllDev and Baul,

Let me try this, I will update you.

Thanks Buddy

On Mon, Jul 2, 2018 at 4:32 PM, baul notifications@github.com wrote:

@alldev0825 https://github.com/alldev0825 thanks. For old docker 1.13, can use /etc/sysconfig/docker to set proxy ,but for 17.06 docker-ce, it does not work,need setup proxy in the systemd dir as the link you supplied.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/moby/moby/issues/22635#issuecomment-401768028, or mute the thread https://github.com/notifications/unsubscribe-auth/AGz0zEiSo7eyIYO7opx2bMhAGv1ULj5Fks5uCf3UgaJpZM4IbXBr .

thx

8.8.8.8 didn’t work for me, but 8.8.4.4 did work. looks like whoever got this error- try different dns’s and hopefully one of them will work.

The solution from @brunolm worked for me! Thanks!

As kirstin said setting the DNS on Docker fixes the issue

DNS to 8.8.8.8 on the “vEthernet (DockerNAT)” network adapter.

You can do this from Powershell with

Get-DnsClientServerAddress | Where-Object { $_.InterfaceAlias -like "*(DockerNAT)" } | ForEach-Object { Set-DnsClientServerAddress -InterfaceIndex $_.InterfaceIndex -ServerAddresses @("8.8.8.8", "8.8.4.4") }

@alldev0825, thanks! It works for me.