dronie has a bad bot

[robe]

I haven't looked at the sql table to see what bad actor is causing this:

pg-9.5: Error response from daemon: OCI runtime create failed: container_linux.go:345: starting container process caused "process_linux.go:430: container init caused \"join session keyring: create session key: disk quota exceeded\"": unknown

[strk]

Is this the reason for every recent Dronie job of PostGIS to fail per timeout after 1hour waiting time on the last matrix cell ? ie: ​https://dronie.osgeo.org/postgis/postgis/557

[strk]

The 1hour waiting time is due to a timeout. According to ​https://github.com/drone/drone/issues/554#issuecomment-58982701 such timeout can be changed via the web ui by the administrator (Regina only, at the time of writing).

I went on and tried to do it at the database level:

update repos set repo_timeout = 180 where repo_namespace = 'postgis' and repo_name = 'postgis';

[strk]

The manual tweak seem to have worked: ​https://dronie.osgeo.org/postgis/postgis/577/1/1 The above builds is been running for 1h35m now and counting. Why that agent is so slow is another question. For the drone administration UI I filed ​https://trac.osgeo.org/osgeo/ticket/2345

[strk]

this whole ticket, btw, should probably be on the "osgeo" trac, not PostGIS' (as agents are shared by all projects)

[strk]

I only see timeout problems at the moment, and I raised timeout to 3 hours, which seems mostly fine (still it takes too long to build/test postgis but we could address that in a separate ticket, maybe to build with -O2 instead of -O0)

[robe]

Milestone renamed