verm__ changed the topic of #buildbot to: A Software Freedom Conservancy Project | Buildbot-3.5.0 | docs: http://docs.buildbot.net/current/ | tutorial: http://docs.buildbot.net/current/tutorial | irclogs: https://libera.irclog.whitequark.org/buildbot
manx has quit [Quit: manx]
manx has joined #buildbot
mjw has quit [Killed (NickServ (GHOST command used by wielaard!~mjw_@2001:1c06:2488:1400:9e5c:8eff:fe8f:a440))]
wielaard is now known as mjw
mark_ has joined #buildbot
p12tic has quit [Quit: quit]
p12tic has joined #buildbot
aakashjain has quit [Remote host closed the connection]
vmeson has quit [Remote host closed the connection]
aakashjain has joined #buildbot
vmeson has joined #buildbot
Guest4 has joined #buildbot
<Guest4> hi there all! very beginner/silly question
<Guest4> looking to see if there's any chance there might creative way to cache package dependencies btw builds
Guest4 is now known as devo-devo-devo
<devo-devo-devo> looking to implement something like this to speed up or internal build workflow https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows
devo-devo-devo has quit [Ping timeout: 244 seconds]
vmeson has quit [Ping timeout: 252 seconds]
vmeson has joined #buildbot
aakashja_ has joined #buildbot
aakashjain has quit [Read error: Connection reset by peer]
<glogan> I use the docker containers, and either SCP/S3 the artifacts to a central source, or write them to the worker via a volume
<glogan> depending on what's needed
<glogan> For maven artifacts I just cache on the local workers. For our built tarballs I copy them to a central location, and then pull them back down when building packages
Guest4 has joined #buildbot
Guest4 has quit [Ping timeout: 244 seconds]
aakashja_ has quit [Remote host closed the connection]
aakashjain has joined #buildbot
vmeson has quit [Ping timeout: 252 seconds]
Guest4 has joined #buildbot
vmeson has joined #buildbot
mjw has quit [Killed (NickServ (GHOST command used by mark_!~mark@gnu.wildebeest.org))]
mark_ is now known as mjw
wielaard has joined #buildbot
Guest4 has quit [Quit: Ping timeout (120 seconds)]
Guest4 has joined #buildbot
Guest4 is now known as devo-devo-devo
<devo-devo-devo> thanks glogan!
<devo-devo-devo> what route might you go for python and npm dependencies? would you right those to a volume as well?
<glogan> If you're going with docker containers then yeah. In my case I actually bind-mount a local directory on the worker rather than storing things in a docker volume, but it's the same idea
<glogan> I'd probably do the same without the docker volumes tbh. Just move/copy/rsync the stuff you want somewhere safe
devo-devo-devo has quit [Quit: Ping timeout (120 seconds)]
Guest4 has joined #buildbot
Guest4 is now known as devo-devo-devo
devo-devo-devo has quit [Quit: Ping timeout (120 seconds)]
vmeson has quit [Ping timeout: 252 seconds]
vmeson has joined #buildbot
Guest4 has joined #buildbot
Guest4 is now known as devo-devo-devo
<devo-devo-devo> glogan thanks again. you wouldn't happen to have any sample or pseudo code I could take a glance at?
<glogan> sure, sec
<glogan> So, this is all ansible-ized, but it should be fairly clear to figure out
<glogan> https://github.com/opencast/opencast-project-infrastructure/blob/master/ansible-buildbot-cluster/roles/buildbot-config/templates/master.cfg#L84 is setting the latent worker(s) shared volumes (bind mounts) to the underlying storage medium/disk
<glogan> This is set of playbooks is what underpins http://ci.opencast.org/
<glogan> Our builds call https://github.com/opencast/opencast-project-infrastructure/blob/master/ansible-buildbot-cluster/roles/buildbot-config/templates/common.py#L87, which sets the relevant environment variables so that the maven dependencies end up in the right places
<glogan> among other things
devo-devo-devo has quit [Quit: Ping timeout (120 seconds)]
jcarl43 has joined #buildbot
dustinm` has quit [Quit: Leaving]
dustinm` has joined #buildbot
jcarl43 has quit [Ping timeout: 255 seconds]
vmeson has quit [Ping timeout: 272 seconds]