-
Notifications
You must be signed in to change notification settings - Fork 18.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
service dns - no route to host when routing to the container executing the request #23965
Comments
As you can see below, I exec into web 1 and only can receive a curl for web 2 (little text at end of body tag)
|
I am also hitting this problem with 1.12-rc3. In my case, I want to run an Elasticsearch cluster using the primitives provided by swarm. This is so that I can easily scale the cluster up and down using The command I am using is:
The trick here is to set the name of the service to Because the containers cannot reach itself using the DNS name, in some cases, if you are unlucky, starting the service will cause a container to look up its own ip address using the
I hope this can be fixed for 1.12 as there are lots of use-cases where we want to use the DNS name of a service and the gossip protocol to discover other nodes/instances in a clustered application. |
ping @mavenugo PTAL |
Not sure if this is a DNS problem. I am also using Elastic and get a Java "No route to host" exception on a secondary instance, trying to connect to the first one on the same overlay network. Strange thing is, that I can exec into the secondary container and do a ping on the IP of the first one - that works.
Docker version: |
Any chance this one can make it into 1.12.2? |
This related to #25266? When you get the wrong IP address on lookups, |
@jhorwit2 @bluepuma77 @F21 @garthk This is most likely fixed in 1.12.2-rc. Please give https://github.com/docker/docker/releases/tag/v1.12.2-rc3 a try. Thanks! |
Unfortunately I have the same issue on 1.12.3. It occurres after running ~2h elasticsearch cluster on heavy load.
|
We have seen the same issue on docker 1.13.1
docker version
Exception is
|
I noticed that this can happen, when the system (e.g. just one node) is very very busy. |
Let me close this ticket for now, as it looks like it went stale. |
Output of
docker version
:Output of
docker info
:mac beta
Steps to reproduce the issue:
Describe the results you received:
When I run
curl web
it only returns the result for the other task. Whenever it tries to route to the current task it responds withcurl: (7) Failed to connect to web port 80: No route to host
Describe the results you expected:
Every curl should respond with nginx index page even if it's to the current container.
Additional information you deem important (e.g. issue happens only occasionally):
^Dockerfile for the test being run.
The text was updated successfully, but these errors were encountered: