Troubleshooting
Use this Go community space to get help troubleshooting any product related issues.
Welcome Guest
Log in if you have an account.
Otherwise, register for an account.
For more details, please visit the How This Site Works page.
Forums
Troubleshooting Recent Posts
Post posted February 26, 2012 by BluephlameI have a Go system, that downloads assets onto an agent, on one particular agent this process takes an increadibly long time. (>20 min for 250mb)
We recently upgraded to 2.4 to try and solve this problem, but it did not work. The interesting thing is hapens on the agent with the same resource in each environment (Testing, UAT andProduction)
The Go server displays a warning.
Job '<job name>' is not respondingJob <Job name>/RefreshEnvironment/auserver20 is currently running but has not shown any console activity in the last 13 minute(s). This job may be hung.Any ideas to make this work more efficently would be great!
- Post last edited February 9, 2012 by Kirk Vogen
I am having difficulty running multiple instances of the same pipeline simultaneously. Once one instance is running, it doesn't seem to allow another one to be triggered until the current instance has completed. I have the setting "Automatic pipeline locking" unselected, so I'd expect that it would allow simultaneous execution.
Are there other settings that impact the ability to run simultaneously?
Recent Comments (3 of 3)
- January 17, 2012 by Kirk Vogen: Thanks for the comment -- very helpful. I have two stages so far in my...
- January 16, 2012 by goley: Hi Kirk, Do you have only one stage in the pipeline? If so, then yes,...
- January 13, 2012 by Ian Bridson: Hi Kirk One constraint i can think of that could stop multiple...
We have problems use go-agent run rake task for RoR project. Can anyone help us? Please contact me if you need more information.
We installed go-server and go-agent on the same machine.
OS:
ubuntu (11.04)
Go Version:
2.4
Rails 3
We can run rake task mannualy form the go-agent pipeline/project folder "/var/lib/go-agent/pipelines/<project>
But when trigger this task from go dashboard, got the following errors:
Recent Comments (3 of 5)
- December 30, 2011 by janmejay: Hi, 'bundle install' is actually not a command. 'bundle' is a command,...
- December 28, 2011 by xyzhang: Hi, I modified /etc/default/go-agent and got the following error:...
- December 27, 2011 by janmejay: Doesn't look like it is available in PATH for job. The path you have...
I'm using Go to build projects that are hosted on Git repositories. These git repositories have submodules in.
It seems that about every minute when polling, the Go server will wipe the submodules and re-clone them, which is causing an excessive load.
Why would it be doing this? It can surely keep syncing the existing submodule clone.
For that matter, why does it even need to checkout the submodules? The changes needed to be detected are in the parent repository, as submodules checkout are essentially commit hashes stored in commits to the parent repository.
The Go server log frequently has the following log line in (about every 2 minutes):
WARN [materialUpdateThread] MaterialUpdateService:60 [Material Update] Skipping update of material GitMaterial ... which has been in-progress since [about 1 minute before]
I'm sure that a git hook initiated build is more ideal, but I'd prefer to poll if it is as efficient as it should be.
The Go server version is: 2.4.0
The git version is 1.7.4.1
Kind Regards
Andy
Recent Comments (2 of 2)
- December 21, 2011 by Andy: Hi Janmejay I've got logs for an hour of debug activity. How would you...
- December 19, 2011 by janmejay: Hi Andy, Go server performs material check every one minute, and uses...
Hello,
We would like to have a notification sent to each of our engineers when a particular stage passes for code that they submitted (basically an "all clear" message).
We're using Perforce for version control.
When we have the "Only if it contains my check-ins" checkbox checked, it doesn't work. Without it, everything works as expected.
I'm assuming this is something to do with the "My check-in aliases" field, but I can't figure out how to set it so that it works.
Here is the Materials for a recent check in I made:
Revision: 83286Modified by: jspears@CJESSE64 on 2011-11-30T15:39:57-06:00
Recent Comments
- January 17, 2012 - Thanks for the comment -- very helpful. I have two stages so far in my...
- January 16, 2012 - Hi Kirk, Do you have only one stage in the pipeline? If so, then yes,...
- January 13, 2012 - Hi Kirk One constraint i can think of that could stop multiple...
This does not send a notification when my aliases are set to jspears or jspears@CJESSE64
Am I doing something wrong?
Or, perhaps it's because of the way the user names are specified for our configuration of Perforce (i.e., user_name@CLIENT_NAME)
Any help or suggestions would be appreciated.