See also issue 743137 from a few weeks ago, which is still happening.

Looks like my job did get run, but it took 1.3 hours before it began.

android_cronet_tester shares the bots it uses with four other non-CQ trybots:
 - android_blink_rel
 - android_coverage
 - linux_android_dbg_ng
 - android_unswarmed_n5_rel

Your tryjob had to wait for capacity to become available. This is not a bug.

I'm revisiting capacity for the cronet bots on chromium.android by converting them to swarming. Doing the same here would let us move this bot onto GCE instances, alleviating the capacity bottleneck.

This is still happening.  This didn't used to happen; android_cronet_tester used to execute the job within seconds of it being requested.

#6: part of that is that there were five jobs for android_cronet_tester in ~ an hour from https://chromium-review.googlesource.com/c/615441. another part of that is that build38-b4 has been offline; I've brought it back up.

I do still intend to do what I mentioned in #4, but that's a lower priority than some other things I've been working on.

Immediate problem has been fixed (offline slave), so duping into bug 727613 for a more permanent fix.