I have a flask+gunicorn server, running on an ubuntu docker image (ubuntu:18.04)
In docker-compose I have this:
python_server:
build: ./python_server/
image: flask-server:${PYTHON_SERVER_VERSION}
command:
[
"gunicorn",
"-b",
"0.0.0.0:8080",
"wsgi:app",
"--workers",
"9",
"--worker-class",
"gevent",
"--worker-connections",
"1000",
]
expose:
- "8080"
------ OTHER CONTAINERS THAT USE THIS ONE HERE -------------
When I try to open the port in the Dockerfile (RUN ufw allow 8080) I get:
ufw not found
Other containers can't connect to it, in Node getting:
error: Error: write ECONNRESET
at WriteWrap.afterWrite [as oncomplete] (net.js:789:14)
forgotten to catch a Promise rejection:
write ECONNRESET
and I don't see any connections in the log
Is this a firewall issue?
The ufw command is unnecessary, just remove it.
You're missing port forwarding. "expose", as per the documentation, will "expose ports without publishing them to the host machine."
What you want is:
ports:
- "8080:8080"
Instead of the expose section.
To understand a bit about why you need port forwarding, see https://pythonspeed.com/articles/docker-connection-refused/
Related
I am trying to run Gitlab Docker image on mac Big Sur locally following the steps from documentation: https://docs.gitlab.com/ee/install/docker.html . And always cannot see locally https://gitlab.example.com. I tried both Gitlab EE and CE, and different versions of images, including latest. I also tried to use usual Docker basic run and docker-compose. I also updated Docker Desktop to latest version 4.10.1. I refined the logs and for all cases I have same error in Gitally log below:
~/gitlab/logs/gitaly/current:
{"level":"warning","msg":"[core] [Channel #2 SubChannel #4] grpc: addrConn.createTransport failed to connect to {\n \"Addr\": \"/var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.0\",\n \"ServerName\": \"/var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.0\",\n \"Attributes\": null,\n \"BalancerAttributes\": null,\n \"Type\": 0,\n \"Metadata\": null\n}. Err: connection error: desc = \"transport: Error while dialing dial unix /var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.0: connect: no such file or directory\"","pid":344,"system":"system","time":"2022-07-26T09:57:38.226Z"}
{"level":"warning","msg":"[core] [Channel #3 SubChannel #5] grpc: addrConn.createTransport failed to connect to {\n \"Addr\": \"/var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.1\",\n \"ServerName\": \"/var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.1\",\n \"Attributes\": null,\n \"BalancerAttributes\": null,\n \"Type\": 0,\n \"Metadata\": null\n}. Err: connection error: desc = \"transport: Error while dialing dial unix /var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.1: connect: no such file or directory\"","pid":344,"system":"system","time":"2022-07-26T09:57:38.228Z"}
{"level":"warning","msg":"spawned","supervisor.args":["bundle","exec","bin/ruby-cd","/var/opt/gitlab/gitaly","/opt/gitlab/embedded/service/gitaly-ruby/bin/gitaly-ruby","344","/var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.0"],"supervisor.name":"gitaly-ruby.0","supervisor.pid":400,"time":"2022-07-26T09:57:38.228Z"}
{"address":"/var/opt/gitlab/gitaly/gitaly.socket","level":"info","msg":"listening at unix address","time":"2022-07-26T09:57:38.234Z"}
{"level":"warning","msg":"spawned","supervisor.args":["bundle","exec","bin/ruby-cd","/var/opt/gitlab/gitaly","/opt/gitlab/embedded/service/gitaly-ruby/bin/gitaly-ruby","344","/var/opt/gitlab/gitaly/run/gitaly-344/sock.d/ruby.1"],"supervisor.name":"gitaly-ruby.1","supervisor.pid":401,"time":"2022-07-26T09:57:38.234Z"}
But I am not sure that this is the case - also checked other logs and seems to me no errors in other logs - but of course I will be happy to provide more logs if you want.
This is my 'docker ps -a' status, docker version: 4.10.1, is run with docker-compose:
web:
image: 'gitlab/gitlab-ce:latest'
container_name: 'gitlab'
restart: unless-stopped
hostname: 'gitlab.example.com'
environment:
GITLAB_OMNIBUS_CONFIG: |
external_url 'https://gitlab.example.com'
ports:
- '80:80'
- '443:443'
- '22:22'
volumes:
- '$GITLAB_HOME/config:/etc/gitlab'
- '$GITLAB_HOME/logs:/var/log/gitlab'
- '$GITLAB_HOME/data:/var/opt/gitlab'
list of docker active containers says container is healthy - but actually I see nothing in browser via https://gitlab.example.com.
sudo docker ps -a
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
b0fb6252c7f3 gitlab/gitlab-ce:latest "/assets/wrapper" 3 days ago Up 3 minutes (healthy) 0.0.0.0:22->22/tcp, 0.0.0.0:80->80/tcp, 0.0.0.0:443->443/tcp gitlab
I also tried to add docker host IP to my hosts file - but no success:
192.168.31.182 host.docker.internal
192.168.31.182 gateway.docker.internal
192.168.31.182 gitlab.example.com
I still see nothing in browser by https://gitlab.example.com
Adding more info to #sytech "That's just a placeholder URL... Did you replace gitlab.example.com with your actual GitLab instance URL?"
You either need to add an /etc/hosts, setting to route gitlab.example.com to localhost or modify that to be localhost
hostname: 'localhost'
environment:
GITLAB_OMNIBUS_CONFIG: |
external_url 'localhost'
These settings are telling the nginx proxy what url should be forwarded to gitlab. That's useful when you're running multiple sites from one server. If you are just running it locally, you should just use localhost.
I also tried to add docker host IP to my hosts file - but no success:
You are forwarding the ports from the docker ip. Just use localhost. It probably didn't work because the docker IP you're looking at is for a different docker network.
I am struggling with Go requests between containers.
The issue that I have that the rest of my containers can send request to the node Container that give response, but when I send request from my GoLang application to node I get that refuse error "dial tcp 172.18.0.6:3050: connect: connection refused".
So my whole docker set up is:
version: "3.3"
services:
##########################
### SETUP SERVER CONTAINER
##########################
node:
# Tell docker what file to build the server from
image: myUserName/mernjs:node-dev
build:
context: ./nodeMyApp
dockerfile: Dockerfile.dev
# The ports to expose
expose:
- 3050
# Port mapping
ports:
- 3050:3050
# Volumes to mount
volumes:
- ./nodeMyApp/src:/app/server/src
# Run command
# Nodemon for hot reloading (-L flag required for polling in Docker)
command: nodemon -L src/app.js
# Connect to other containers
links:
- mongo
# Restart action
restart: always
react:
ports:
- 8000:8000
build:
context: ../reactMyApp
dockerfile: Dockerfile.dev
volumes:
- ../reactMyApp:/usr/src/app
- /usr/src/app/node_modules
- /usr/src/app/.next
restart: always
environment:
- NODE_ENV=development
golang:
build:
context: ../goMyApp
environment:
- MONGO_URI=mongodb://mongo:27017
# Volumes to mount
volumes:
- ../goMyApp:/app/server
links:
- mongo
- node
restart: always
So my React app can send the request to "http://node:3050/api/greeting/name" and it get the response even that react app is not linked to the node app but when Golang app sends request to node docker container it gets connection refuse message GetJson err: Get "http://node:3050/api/greeting/name": dial tcp 172.18.0.6:3050: connect: connection refused
func GetJson(url string, target interface{}) error {
r, err := myClient.Get(url)
if err != nil {
fmt.Println("GetJson err: ", err)
return err
}
defer r.Body.Close()
return json.NewDecoder(r.Body).Decode(target)
}
type ResultsDetails struct {
Greeting string `bson:"greatingMessage" json:"greatingMessage"`
Message string `bson:"message" json:"message"`
}
func GetGreetingDetails(name string) ResultsDetails {
var resp ResultsDetails
GetJson("http://node:3050/api/greeting/"+name, &resp)
return resp
}
So how do I solve the Golang request to another Docker Node Container when docker doesnt see the host as the name of my container 'node'?
Update:
By accident i put Golang port, which it doenst run on any port since it is application that checks on database records. So it hasnt got any api, therefore it is not running on any port.
Is that could be the problem why my golang application cannot communication to other containers?
Since i have also another golang application which is api application and it is running on 5000 port and it is well communicating to my node application?
Network info:
After checking the network if node and golang share the same network and the answer is yes. All containers share the same network
(Unrelated to my issue) To anyone who has "dial tcp connection refused" issue I suggest to go though that guide https://maximorlov.com/4-reasons-why-your-docker-containers-cant-talk-to-each-other/. Really helpful. To those who this guide wont help prob read bellow this, maybe you trying to request the container api after just containers were built :D
For those who was interested what was wrong:
Technically reason why I was getting this error is because of the request that I was trying to run, was just when all containers were built.
I believe there is some delay to the network after containers are built. Thats why there host was throwing "dial tcp 172.18.0.6:3050: connect: connection refused" I've run that test on other containers that could possibly send request to that node container and they were all failing after the build time. But when re-requesting after few seconds all worked out.
Sorry to bother you guys. I really spent 3 days into this issue. And I was looking into completely wrong direction. Never thought that the issue is that silly :D
Thanks for you time.
I've met the same error in my harbor registry service.
After I docker exec -it into the container, and check if the service is available, and finally I found that http_proxy has been set.
Remove the http_proxy settings for docker service, then it works like a charm.
Failed on load rest config err:Get "http://core:8080/api/internal/configurations": dial tcp 172.22.0.8:8080: connect: connection refused
$docker exec -it harbor-jobservice /bin/bash
$echo $http_proxy $https_proxy
I'm using docker compose to run a simple web server project I created. This configuration has been working fine for months but suddenly stopped working after I haven't been to the office for two weeks.
It works when I map my ports like that - 8080:80, but I don't want to have to type out port 8080 every time. I used netstat -a -n -o | findstr /c:80 to find the process ID of the process listening to port 80, and tasklist /fi "pid eq 4" to find out what the name of the process is.
Turns out it's some system process, so I'm not sure what to do about that. I've uninstalled Skype and checked that the World Wide Web Publishing Service isn't turned on. Does anybody have an explanation or ideas as to how to fix this?
Thanks in advance.
update
when I run net stop http and kill all dependant services with it, port 80 is free. Services being stopped: Windows Remote Management (WS-Management), SSDP Discovery, Print Spooler, BranchCache and HTTP of course. Which of these could be the culprit?
update 2
I now stopped those services one by one, and after stopping every one of those it seems BranchCache is responsible for this. more testing ensues
docker-compose.yml
version: "3"
services:
vote-client:
build:
context: .
dockerfile: Dockerfile
ports:
- "80:80"
Dockerfile
FROM nginx
COPY ./html /usr/share/nginx/html
when I run docker-compose up this is my output:
docker-compose up --build
Removing vote-client_vote-client_1
Building vote-client
Step 1/2 : FROM nginx
---> 42b4762643dc
Step 2/2 : COPY ./html /usr/share/nginx/html
---> Using cache
---> a1aade2a299e
Successfully built a1aade2a299e
Successfully tagged vote-client_vote-client:latest
Recreating c2654f31dcff_vote-client_vote-client_1 ... error
ERROR: for c2654f31dcff_vote-client_vote-client_1 Cannot start service vote-client: driver failed programming external connectivity on endpoint vote-client_vote-client_1 (2188c8607a04ba2388a661504601431d6b30825d595dafae0c318f2d2b5685b0): Error starting userland proxy: Bind for 0.0.0.0:80: unexpected error Permission denied
ERROR: for vote-client Cannot start service vote-client: driver failed programming external connectivity on endpoint vote-client_vote-client_1 (2188c8607a04ba2388a661504601431d6b30825d595dafae0c318f2d2b5685b0): Error starting userland proxy: Bind for 0.0.0.0:80: unexpected error Permission denied
ERROR: Encountered errors while bringing up the project.
So there are a lot of posts around this subject, but none of which seems to help.
I have an application running on a wildfly server inside a docker container.
And for some reason I cannot connect my remote debugger to it.
So, it is a wildfly 11 server that has been started with this command:
/opt/jboss/wildfly/bin/standalone.sh -b 0.0.0.0 -bmanagement 0.0.0.0 -c standalone.xml --debug 9999;
And in my standalone.xml I have this:
<socket-binding name="management-http" interface="management" port="${jboss.management.http.port:9990}"/>
The console output seems promising:
Listening for transport dt_socket at address: 9999
I can even access the admin console with the credentials admin:admin on localhost:9990/console
However IntelliJ refuses to connect... I've creates a remote JBoss Server configuration that in the server tab points to localhost with management port 9990.
And in the startup/connection tab I've entered 9999 as remote socket port.
The docker image has exposed the ports 9999 and 9990, and the docker-compose file binds those ports as is.
Even with all of this IntelliJ throws this message when trying to connect:
Error running 'remote':
Unable to open debugger port (localhost:9999): java.io.IOException "handshake failed - connection prematurally closed"
followed by
Error running 'remote':
Unable to connect to the localhost:9990, reason:
com.intellij.javaee.process.common.WrappedException: java.io.IOException: java.net.ConnectException: WFLYPRT0053: Could not connect to remote+http://localhost:9990. The connection failed
I'm completely lost as to what the issue might be...
Interessting addition is that after intelliJ fails, if I invalidate caches and restart then wildfly reprints the message saying that it is listening on port 9999
In case someone else in the future comes to this thread with he same issue, I found this solution here:
https://github.com/jboss-dockerfiles/wildfly/issues/91#issuecomment-450192272
Basically, apparart from the --debug parameter, you also need to pass *:8787
Dockerfile:
CMD ["/opt/jboss/wildfly/bin/standalone.sh", "-b", "0.0.0.0", "-bmanagement", "0.0.0.0", "--debug", "*:8787"]
docker-compose:
ports:
- "8080:8080"
- "8787:8787"
- "9990:9990"
command: /opt/jboss/wildfly/bin/standalone.sh -b 0.0.0.0 -bmanagement 0.0.0.0 --debug *:8787
I have not tested the docker-compose solution, as my solution was on dockerfile.
Not sure if this can be seen as an answer since it goes around the problem.
But the way I solved this, was by adding a "pure" remote configuration in intelliJ instead of jboss remote. This means that it won't automagically deploy, but I'm fine with that
I am trying to do RPC with RabbitMQ inside 2 docker containers. I have set them up as follows:
chindow:
image: bablot/chindow
ports:
- "9999:9999"
links:
- rabbitmq:amq
rabbitmq:
image: rabbitmq:3
ports:
- "15672:15672"
Everything works as expected initially, but then, when I shut down the system with ctrl+c it results in the following error:
babelotserver_rabbitmq_1 exited with code 143
and then when I proceed to reboot the system:
chindow_1 | { [Error: connect ECONNREFUSED 172.17.0.3:5672]
code: 'ECONNREFUSED',
errno: 'ECONNREFUSED',
syscall: 'connect',
address: '172.17.0.3',
port: 5672 }
EDIT:
When I docker stop rabbitmq I get code 143
When I docker stop node.js I get code 0
Why are these 2 different? Is this meaningful?
Make sure you are using docker-compose format Version 2.
It will create a bridge network: see "Networking in Compose"
That will allow your containers to stop and re-start while still being able to see/contact each other.
By default Compose sets up a single network for your app.
Each container for a service joins the default network and is both reachable by other containers on that network, and discoverable by them at a hostname identical to the container name.