-
-
Notifications
You must be signed in to change notification settings - Fork 2.6k
The hub cannot start because: "SEVERE: Port 4444 is busy, please choose a free port..." (the port is not in use) #742
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
Comments
what command did you use to run? also, do me a favor and paste the result of the following command (preferably as the root user):
|
Hi @ddavison, I ran the previous command and the port is not in use. |
Hi @mach6 Can you take a look at this issue ? It seems to be an issue introduced by you. Please let me know if you need more details. |
i'm tempted to close this as i can't reproduce this. this looks like it's a setup issue on your end @alinpa , not necessarily an issue with our images |
@ddavison : it is enough to start the selenium grid with 1 node on two ec2 machines. Start the docker images and you will see the host is not discovered properly. Please check the following issue I already documented there: jenkinsci/selenium-plugin#128 (comment) |
We can reporduce this issue, port 4444 is reported as in use while plugin starts, hweover, a netstat reports no process serving on that port. We are using a jenkins deployment within a kubernetes cluster. |
@mtilev Please check my comments here: jenkinsci/selenium-plugin#128 (comment) The error message is wrong! |
@alinpa We implemented the workaround; nevertheless, this is an undocumented change in behaviour and thus either a bug or a documentation bug. |
@mtilev Did you implement my workaround or something new? |
@alinpa We followed your advice. Kind regars, m |
got same issue with node. v3.12:
with 3.11 looks fine:
k8s deployment:
|
Seeing the same issue on Docker node (both Chrome & Firefox) as @dober
Sadly the same issue is present on 3.13 and 3.14, however I can confirm it was working as expected on 3.11. |
any news? |
Hi all, Yes, this was a change in the upstream project where all the downstream projects had to adjust. The reasons why the change was done are out of my reach, but in the end it was done and here in docker-selenium we had to adjust the "host" parameter in the hub as well. The original issue was the port being used when starting the hub, and the Jenkins project as a downstream project needs to adapt as well. The other posts are related to nodes which are not related to the original issue. Would it be possible to have separate issues that mention only those ones with clear ways to reproduce them? For that, it will be ideal if you create a new issue with the template completely filled. |
Closing as we didn't get replies to the last comment. For the comments related to nodes, feel free to open a new issue with all the needed details to reproduce any potential problem. |
Can try in this way it worked for me : |
Meta -
Image(s):
selenium/node-chrome:3.12.0
hub: jenkins selenium hub 3.12.0
Docker-Selenium Image Version(s):
3.12.0
Docker Version:
18.03.1-ce
OS:
Linux 4.14.42-52.37.amzn1.x86_64 #1 SMP Tue May 22 00:41:10 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
-->
Expected Behavior -
The nodes have to connect to hub.
Actual Behavior - The nodes cannot connect to the hub because:
"14:33:16.014 INFO [SelfRegisteringRemote$1.run] - Couldn't register this node: The hub is down or not responding: Failed to connect to /xxx.xxx.xxx.xxx:4444"
On the hub logs I noticed the following error:
OFF: Grid Hub preparing to start on port 4444
2018-06-07 13:10:54.602:INFO::pool-1-thread-1 for channel id=0: Logging initialized @12793ms to org.seleniumhq.jetty9.util.log.StdErrLog
2018-06-07 13:10:54.780:INFO:osjs.Server:pool-1-thread-1 for channel id=0: jetty-9.4.7.v20170914, build timestamp: 2017-11-21T21:27:37Z, git hash: 82b8fb23f757335bb3329d540ce37a2a2615f0a8
2018-06-07 13:10:54.842:INFO:osjs.session:pool-1-thread-1 for channel id=0: DefaultSessionIdManager workerName=node0
2018-06-07 13:10:54.843:INFO:osjs.session:pool-1-thread-1 for channel id=0: No SessionScavenger set, using defaults
2018-06-07 13:10:54.849:INFO:osjs.session:pool-1-thread-1 for channel id=0: Scavenging every 660000ms
2018-06-07 13:10:54.870:INFO:osjsh.ContextHandler:pool-1-thread-1 for channel id=0: Started o.s.j.s.ServletContextHandler@54620f3b{/,null,AVAILABLE}
Jun 07, 2018 1:10:54 PM org.openqa.grid.internal.DefaultGridRegistry assignRequestToProxy
INFO: Shutting down registry.
2018-06-07 13:10:54.880:INFO:osjs.AbstractConnector:pool-1-thread-1 for channel id=0: Stopped ServerConnector@21f99f64{HTTP/1.1,[http/1.1]}{xxx.xxx.xxx.xxx:4444}
2018-06-07 13:10:54.880:INFO:osjs.session:pool-1-thread-1 for channel id=0: Stopped scavenging
2018-06-07 13:10:54.882:INFO:osjsh.ContextHandler:pool-1-thread-1 for channel id=0: Stopped o.s.j.s.ServletContextHandler@54620f3b{/,null,UNAVAILABLE}
Jun 07, 2018 1:10:54 PM org.openqa.grid.web.Hub start
SEVERE: Port 4444 is busy, please choose a free port for the hub and specify it using -port option
Jun 07, 2018 1:10:54 PM hudson.plugins.selenium.HubLauncher call
OFF: Grid Hub started on port 4444 with args: -maxSession 5 -timeout 30 -browserTimeout 0 -host xxx.xxx.xxx.xxx
I have checked if the port is already used and is not in use.
The text was updated successfully, but these errors were encountered: