New issue
Advanced search Search tips

Issue 763122 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Sep 2017
Cc:
Components:
EstimatedDays: ----
NextAction: 2017-09-08
OS: Android
Pri: 1
Type: Bug-Regression



Sign in to add a comment

Several benchmarks failing with "shard #0 expired, not enough capacity" on Android Nexus 5X WebView Perf --device2

Project Member Reported by charliea@chromium.org, Sep 7 2017

Issue description

Example run: https://uberchromegw.corp.google.com/i/chromium.perf/builders/Android%20Nexus5X%20WebView%20Perf/builds/243 (search for "--device2")

My first thought is that the bot might just be down. However, this doesn't seem to be the case because several benchmarks are passing on the device (e.g. dramaeo.domcorequery, dromaeo.domcoretraverse, smoothness.simple_mobile_sites).

My second thought was that one particular benchmark might be taking a really long time to run, hogging all of the device's capacity and starving the other benchmarks of runtime. However, the benchmarks that succeeded aren't the first or the last ones to be scheduled. In addition, the sum of the duration of the benchmarks that didn't time out due to lack of capacity seems to be less than 20 minutes: far less than the 10 hour timeout.

Stephen, do you have any idea how to debug this? I'm very confused.
 
Summary: Several benchmarks failing with "shard #0 expired, not enough capacity" on Android Nexus 5X WebView Perf --device2 (was: Several benchmarks failing with "shard #0 expired, not enough capacity" on Android Nexus 5X Perf --device2)
There are 3 machines with 7 devices each on this bot; build164, 165, and 166. build166--device2 is down, but 165 and 164 are fine. Does that make sense?
Components: Infra>Labs
Ahhh, yes, totally didn't see that there were multiple --device2s. I'm kicking this over to labs.

Infra>Labs, could you take a look at build166-b1--device2 and see if this phone's having a problem?

Comment 4 by jo...@google.com, Sep 7 2017

Owner: jo...@chromium.org
Status: Assigned (was: Untriaged)
It was sitting at the bootloader prompt.  Continued the boot up process and did a couple of test reboots after that.

Looks good for now.
Thanks Peter! I'll reopen if something looks off in future runs.
Status: Fixed (was: Assigned)
The NextAction date has arrived: 2017-09-08

Sign in to add a comment