Community Tip - You can subscribe to a forum, label or individual post and receive email notifications when someone posts a new topic or reply. Learn more! X
I'm working with several development projects that want to start using Jenkins and our MKS Source 2009 repositories.
Since the use of Jenkins or CI tools in general is still very new here, there is reluctance in allowing our Jenkins server to connect straight to our production repository server until we get some more experience.
I've managed to negotiate a compromise where we can connect Jenkins to a "test" MKS Source Server and allow it to pull code from there. However for that to work I need a way for the changes committed to our production repository to get pushed to the test server, ideally every few minutes.
Any suggestions on a fast, stable method for sending changes from one server to the other?
Solved! Go to Solution.
Hi Mark,
my first idea would be a checkin.post trigger on your production server, that simply repeats the checkin operation on a different server.
Regarding the Jenkins environment:
We use Jenkins for a while now in our productive enviroment and there are 2 reasons make us all sleep well.
1. The user account that we use for the connection from Jenkins to MKS has only "fetch revision" and "open project" permissions; so it cannot do anything more than read. (for some project we extended this to the checkpoint permission)
2. We do not directly connect to the MKS server, but use a MKS client installed on oour Jenkins machine as a proxy ("integration point"). This is done because of a known memory leak issue in the server's API implementation. Using a client as proxy helps is this matter as it keeps our server alive while in a problem case the client dies first.
You can call PTC support to assist you in this case.
HTH Matthias
Hi Mark,
my first idea would be a checkin.post trigger on your production server, that simply repeats the checkin operation on a different server.
Regarding the Jenkins environment:
We use Jenkins for a while now in our productive enviroment and there are 2 reasons make us all sleep well.
1. The user account that we use for the connection from Jenkins to MKS has only "fetch revision" and "open project" permissions; so it cannot do anything more than read. (for some project we extended this to the checkpoint permission)
2. We do not directly connect to the MKS server, but use a MKS client installed on oour Jenkins machine as a proxy ("integration point"). This is done because of a known memory leak issue in the server's API implementation. Using a client as proxy helps is this matter as it keeps our server alive while in a problem case the client dies first.
You can call PTC support to assist you in this case.
HTH Matthias
Thanks, I wasn't aware of the API memory leak or the Integration point functionality. Connecting to a Client instead of directly to the server sounds like a great option.