This might help us determine what's going on with the reboots, etc. It would at the least answer the following questions: 1. Is the rebooted tegra the same tegra as the one that started the test? 2. Is there something funny about the machine state before or after the reboot?

This logs the output of dm.getInfo() before and after the test. Contains juicy information like the tegra id, network address, uptime, memory info, and process list

Comment on attachment 645882 [details] [diff] [review] Log information about device before and after test for remote mochitest and reftest Review of attachment 645882 [details] [diff] [review]: ----------------------------------------------------------------- this one comment should apply to both files the same, it is only a suggestion, I am fine either way. ::: layout/tools/reftest/remotereftest.py @@ +441,5 @@ > reftest.stopWebServer(options) > return 1 > > reftest.stopWebServer(options) > + print dm.getInfo() we might want to consider putting this after the logcat dump, just so it is easier to find :)

Pushed (changed to work per jmaher's suggestion): https://hg.mozilla.org/integration/mozilla-inbound/rev/3926441652e8

Results of try run: https://tbpl.mozilla.org/?tree=Try&rev=bf3c53c61586 There was one reboot, nothing interesting in the log unfortunately -- as far as I can tell the device's state and identity is pretty much the same after a reboot. https://tbpl.mozilla.org/php/getParsedLog.php?id=13852292&tree=Try