https://uberchromegw.corp.google.com/i/chromeos/builders/veyron_minnie-cheets-chrome-pfq/builds/234

Aviv, who should this be assigned to? 

Untriaged for now, tracking issue.

The pool is now healthy. Marking fixed.

Hijacking this as a general "make pool health better" bug.

The optimal solution to this is to make servo and repair more robust. Ideally, we should be able to push out a crashing build to the lab, and see no problems other than failed tests.

As workarounds, we can balance pools automatically (Kevin is working on this), and increase the number of spare duts that we keep around.

We can also continue to improve test reporting, so that it's easy to understand when a bad build is actually bad, instead of just flake.

Although reliable repair and automatic balancing will fix
certain causes of pool health problems, it won't necessarily
fix all of them.  What data do we have about why DUTs aren't
available for testing at the time of failure?

At minimum, it would be good for the automated bug reports to
include detailed per-DUT status information.

Issue has not been modified or commented on in the last 365 days, please re-open or file a new bug if this is still an issue.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot