Skylab tasks insert TKO entries with reference_type "afe". Should be "skylab" |
||||
Issue descriptionExample task: https://chromium-swarm-dev.appspot.com/task?id=3f2997bd9dba2611 Example tko entry: MySQL [chromeos_autotest_db]> select * from tko_task_references where tko_job_idx=742; +-----+----------------+-------------+---------+------------------+ | id | reference_type | tko_job_idx | task_id | parent_task_id | +-----+----------------+-------------+---------+------------------+ | 742 | afe | 742 | NULL | 3f2997bd9dba2611 | +-----+----------------+-------------+---------+------------------+
,
Aug 6
Problem is that tko/utils.py is not able to extract swarming task id from the jobname. Swarming jobname: results/swarming-3f2997d1fbd54c11/autoserv_test Comparing the jobname with an Autotest job: https://stainless.corp.google.com/browse/chromeos-autotest-results/223383839-chromeos-test/chromeos4-row1-rack5-host15/ Autotest jobname: 223383839-chromeos-test/chromeos4-row1-rack5-host15
,
Aug 6
,
Aug 7
The following revision refers to this bug: https://chromium.googlesource.com/chromiumos/infra/lucifer/+/66eb0cd2f2d1cc99585df677a63a86855381da23 commit 66eb0cd2f2d1cc99585df677a63a86855381da23 Author: Prathmesh Prabhu <pprabhu@chromium.org> Date: Tue Aug 07 01:08:55 2018 lucifer: Fix tko/parse level Again. This was tweaked in CL:1111202, but the <hostname> sub-folder has since been removed. BUG= chromium:871426 TEST=None Change-Id: Ia8edcc20b1cfa4090ed30e770bafba8cc8f67be6 Reviewed-on: https://chromium-review.googlesource.com/1164406 Commit-Ready: Prathmesh Prabhu <pprabhu@chromium.org> Tested-by: Prathmesh Prabhu <pprabhu@chromium.org> Reviewed-by: Allen Li <ayatane@chromium.org> [modify] https://crrev.com/66eb0cd2f2d1cc99585df677a63a86855381da23/src/lucifer/cmd/lucifer/testcmd.go
,
Aug 7
#4 didn't fix staging MySQL [chromeos_autotest_db]> select * from tko_task_references where reference_type = 'skylab' order by id desc limit 5; Empty set (0.03 sec)
,
Aug 7
A task from this morning https://chromium-swarm-dev.appspot.com/task?id=3f2df32587eaf210&refresh=10 still passed in --level 3 to tko/parse. i.e. #4 is not active in staging lab.
,
Aug 7
The deployed lucifer package is chromeos-test@cros-skylab-staging-1:~$ cipd instances chromiumos/infra/lucifer Instance ID │ Timestamp │ Uploader │ Refs ───────────────────────────────────────────────────────────────────────────────────────────────────────────────────── bc8baddfbf0dbb34138b635f6793feb5ca0da7d0 │ Aug 06 18:37 PDT 2018 │ cipd-uploader@chromeos... │ prod prod-next staging latest Built by https://cros-goldeneye.corp.google.com/chromeos/healthmonitoring/buildDetails?buildbucketId=8938893088835293504 But the blamelist does not contain the CL in #4 https://crosland.corp.google.com/log/10937.0.0..10945.0.0 That CL is in a future release. https://crosland.corp.google.com/log/10945.0.0..10946.0.0 Perhaps there is a race in how the infra-go picks up landed CLs?
,
Aug 7
Allen kicked a fresh build. Wait and watch.
,
Aug 7
It works! Staging task: https://chromium-swarm-dev.appspot.com/task?id=3f2ecb3b66335d10&refresh=10&show_raw=1&wide_logs=true MySQL [chromeos_autotest_db]> select * from tko_task_references where reference_type='skylab'; +------+----------------+-------------+------------------+------------------+ | id | reference_type | tko_job_idx | task_id | parent_task_id | +------+----------------+-------------+------------------+------------------+ | 1156 | skylab | 1156 | 3f2ecb4089fe9311 | 3f2ecb22377b9d11 | | 1157 | skylab | 1157 | 3f2ecb3b66335d11 | 3f2ecb22377b9d11 | | 1158 | skylab | 1158 | 3f2ece3ea791af11 | 3f2ece2ed6adc511 | ... |
||||
►
Sign in to add a comment |
||||
Comment 1 by pprabhu@chromium.org
, Aug 6