tast.power.Reboot failed in CQ |
|||
Issue descriptionCould be a network adapter flake. wolf-paladin https://cros-goldeneye.corp.google.com/chromeos/healthmonitoring/buildDetails?buildbucketId=8924063449845250992 01/17 11:28:46.924 INFO | server_job:0217| ABORT ---- ---- timestamp=1547753326 localtime=Jan 17 11:28:46 Autotest client terminated unexpectedly: DUT is pingable, SSHable and did NOT restart un-expectedly. We probably lost connectivity during the test. 019/01/17 11:22:31 Started test power.Reboot 2019/01/17 11:22:31 [11:22:31.413] Rebooting DUT 2019/01/17 11:22:31 [11:22:31.423] Waiting for DUT to become unreachable 2019/01/17 11:22:35 [11:22:35.430] DUT became unreachable (as expected) 2019/01/17 11:22:35 [11:22:35.430] Reconnecting to DUT 2019/01/17 11:27:31 [11:27:31.415] Error at reboot.go:45: Failed to reconnect to DUT: context deadline exceeded 2019/01/17 11:27:31 [11:27:31.415] Stack trace: Failed to reconnect to DUT at chromiumos/tast/remote/bundles/cros/power.Reboot (reboot.go:45) at chromiumos/tast/testing.(*Test).Run.func4 (test.go:228) at chromiumos/tast/testing.runStages.func1.1 (stage.go:39) at chromiumos/tast/testing.runAndRecover.func1 (stage.go:69) at runtime.goexit (asm_amd64.s:1333) context deadline exceeded 2019/01/17 11:27:31 Completed test power.Reboot in 5m0.004s with 1 error(s)
,
Jan 17
(5 days ago)
Not sure if it's related by nyan_big failed with similar error: graphics_Gbm_SERVER_JOB FAIL: Aborting - unexpected final status message from client on chromeos4-row5-rack10-host11 graphics_Gbm_CLIENT_JOB.0 [ FAILED ] graphics_Gbm_CLIENT_JOB.0 ABORT: Autotest client terminated unexpectedly: DUT is pingable, SSHable and did NOT restart un-expectedly. We probably lost connectivity during the test. Is this a lab issue?
,
Jan 17
(5 days ago)
#2 probably not a lab issue. We have lots of DUT connectivity issues.
,
Jan 17
(5 days ago)
After some digging through logs, I think these both might just be flakes. Will revisit if this pops up again.
,
Jan 17
(5 days ago)
,
Jan 18
(5 days ago)
I don't have access to the full logs right now, but did the DUT look like it actually came back up eventually (assuming that either Autotest or Tast managed to collect some logs)? If so, it was probably network flakiness. I think I've seen similar issues with power.Reboot in the past, so if it's not actually verifying anything useful (e.g. if there are updater tests that verify earlier that the system reboots properly), then we could take the test out of the CQ. |
|||
►
Sign in to add a comment |
|||
Comment 1 by semenzato@chromium.org
, Jan 17 (5 days ago)