New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.
Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Feb 2018
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 1
Type: Bug



Sign in to add a comment

chromeos2-row3-rack1-host13 is down, and this is killing soraka-release

Project Member Reported by ejcaruso@chromium.org, Feb 27 2018

Issue description

sshing into chromeos2-row3-rack1-host13 times out, and it has failed dummy_PassServer.sanity twice. We've lost 2 runs in a row to this DUT (#920, #921) so I don't think this is just lab network flakiness anymore.

Example failure: https://luci-milo.appspot.com/buildbot/chromeos/soraka-release/921

Assigning to deputy.
 

Comment 1 by nxia@chromium.org, Feb 28 2018

locked the dut and filed b/73965859

Comment 2 by nxia@chromium.org, Feb 28 2018

Status: Fixed (was: Assigned)
re-balanced the pool
This issue appears to be more widespread than a few DUTs.

https://viceroy.corp.google.com/chromeos/provision?groups=Provision&groups=Auto+Update&breakdowns=dut_host_name&breakdowns=build&board=soraka&build_type=release&success=&devserver=chromeos%5B246%5D-devserver.*&dut=&topstreams=10&delta_window=4h&duration=8d&percentile=90&prior_alpha=0&prior_beta=0&refresh=-1&repository_behavior=DO_NOT_SKIP&type=Provision&use_precomputation=1&utc_end=1519423906

Note that due to the fact failures often take longer than successes, this will result in spikes in the failure percentage as the denominator essentially goes to zero.

If these provision failures started happening recently, I'd look for a product issue.

Sign in to add a comment