I think it could be made to work with enough exploring and experimenting, but it seems to me like you're using something expensive (human time, thought, planning, diagnosis) in order to conserve something that is relatively inexpensive (hard drive capacity). Why not just buy more hard disc space, avoid the complexity and reduce the network traffic?
I think you're probably also sacrificing execution time in your desire to save disc space. In my case, execution time is the most critical aspect of the Jenkins jobs, because we can only provide feedback to users as fast as we can cycle through the jobs. I'm sure there are cases where the trade-offs you're considering are exactly the right trade-offs (for example, massive video production where you cannot afford to have more than a single copy of the workspace, a monster finite element analysis job where a single set of working files may overwhelm your capacity, etc.). Mark Waite >________________________________ > From: Eric Krystof <erkrys...@gmail.com> >To: jenkinsci-users@googlegroups.com >Sent: Friday, May 17, 2013 7:07 PM >Subject: Multiple Jenkins Instances, Single CIFS file system. > > > >Discussing an interesting scenario with my tech lead at work today. Might try >this out but was looking for any definitive remarks indicating we may be >trying to divide by zero. > > >We need to build up a few slaves since our single instance is becoming >overwhelmed. The thought was that since this is entirely local network-wise >that we possibly try this: > > >Master system with all jobs and workspaces in directory X. > > >Slaves setup to mount X as CIFS to run their jobs on. > > >We need the horsepower, not the space, so we're trying not to waste it, hence >the CIFS idea (which could add to build time, which could be acceptable). > > >So I'm wondering... is that too far off the rails? Would we need to segregate >jobs to run only on certain slaves? Does the shared mount bring along issues >of configurations being overwritten or not read properly by each Jenkins >instance? > > >Thanks for pondering. -- >You received this message because you are subscribed to the Google Groups >"Jenkins Users" group. >To unsubscribe from this group and stop receiving emails from it, send an >email to jenkinsci-users+unsubscr...@googlegroups.com. >For more options, visit https://groups.google.com/groups/opt_out. > > > > > -- You received this message because you are subscribed to the Google Groups "Jenkins Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to jenkinsci-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/groups/opt_out.
