Any update with this? I solved temporarilly this issue just creating one folder with the project code and all the configurations are a sym link to that folder, so all of them will find code in their workspace. I would prefer something in jenkins but I couldn't find how. The ref works but clones for every configuration it's a valid walkarround but I prefer to avoid it.
Thanks On Thu, Mar 29, 2012 at 1:23 PM, Sami Tikka <sjti...@gmail.com> wrote: > Open the advanced settings under the job git configuration. There you will > find a place to enter path local reference repository. That should make > things faster for you and save a lot space. > > We have build slaves with smallish SSD disks, so we're very careful about > disk space use. We have a clone off the git repo on each slave (updated > nightly) and use the reference repo feature, so most of the repo space is > shared and only what is new gets pulled over the network by clone. > > -- Sami > > Ido Ran <ido....@gmail.com> kirjoitti 27.3.2012 kello 14.02: > > I didn't try to --reference option, I will. > I also understand that the repository will be clone to each job, in my > case it's matrix job so it's even more clones. > I hope someone will find a way. > > I'll post here if I'll manage to find a solution. > > Ido > > On Tue, Mar 27, 2012 at 12:45 PM, Albert Callarisa <shark...@gmail.com>wrote: > >> I asked something similar recently but the reply didn't help me. Did you >> try with the --reference option in git? I tried but didn't help >> I'm still trying to figure out how to make it. I'm fine with a clone in >> each computer, but once is cloned I would like to just pull from the same >> place, but the workspace of the jobs is kind of random so it's fresh new >> always... >> >> >> On Tue, Mar 27, 2012 at 4:57 PM, Ido Ran <ido....@gmail.com> wrote: >> >>> Hi, >>> We use Jenkins to build our applications which are stored in git >>> repository. >>> We have several different jobs for building different configuration and >>> deployment of our applications. >>> Currently each project pull it's own clone of the git repository from >>> our central repository which takes about 7 minutes for clean build. >>> >>> We try to have single job that clone the remote repository and have all >>> the other jobs pull from that repository but we fail to do it due to >>> locking problems. >>> Is it possible and should we persevere on that approach or is it better >>> to let each job pull from the remote repository. >>> >>> Ido. >>> >> >> >> >> -- >> Albert Callarisa Roca >> > > -- Albert Callarisa Roca
