action #44594
closed
[functional][u] test fails in sysstat - regular expression for pidstat test is incorrect
Added by szarate over 5 years ago.
Updated about 5 years ago.
Category:
Bugs in existing tests
Target version:
SUSE QA - Milestone 23
Description
Observation¶
Looking one step further into the test, we can see that waiitserial was expecting a 0, and got a "0\n"... however the test still failed... so, false negative there?
openQA test in scenario sle-15-SP1-Installer-DVD-s390x-extra_tests_in_textmode@s390x-kvm-sle12 fails in
sysstat
Reproducible¶
Fails since (at least) Build 47.1
Expected result¶
Seems to always fail on s390x since the introduction of the test module (I think)
Suggestions¶
- Crosscheck against the test on x86_64
- See why s390x is behaving differently
- Potentially rewrite the test module to use a more stable approach of parsing the serial terminal output
Further details¶
Always latest result in this scenario: latest
But pidstat | awk '{print $3}' | head | tail -n 1 should not return 0.
~ pidstat
Linux 4.18.5-1-default (localhost) 11/30/18 _x86_64_ (8 CPU)
15:04:37 UID PID %usr %system %guest %wait %CPU CPU Command
15:04:37 0 1 0.01 0.01 0.00 0.00 0.01 3 systemd
15:04:37 0 2 0.00 0.00 0.00 0.00 0.00 4 kthreadd
15:04:37 0 9 0.00 0.00 0.00 0.00 0.00 0 ksoftirqd/0
15:04:37 0 10 0.00 0.08 0.00 0.04 0.08 6 rcu_preempt
15:04:37 0 11 0.00 0.00 0.00 0.00 0.00 1 rcu_sched
15:04:37 0 12 0.00 0.00 0.00 0.00 0.00 7 rcu_bh
15:04:37 0 14 0.00 0.00 0.00 0.00 0.00 0 watchdog/0
15:04:37 0 17 0.00 0.00 0.00 0.00 0.00 1 watchdog/1
15:04:37 0 19 0.00 0.00 0.00 0.00 0.00 1 ksoftirqd/1
15:04:37 0 23 0.00 0.00 0.00 0.00 0.00 2 watchdog/2
15:04:37 0 24 0.00 0.00 0.00 0.00 0.00 2 migration/2
15:04:37 0 25 0.00 0.00 0.00 0.00 0.00 2 ksoftirqd/2
➜ ~ cat - > /tmp/run_test
pidstat | awk '{print $3}' | head | tail -n 1
➜ ~ bash -eox pipefail /tmp/run_test
+ pidstat
+ awk '{print $3}'
+ head
+ tail -n 1
14
- Subject changed from [u] test fails in sysstat - Possible false negative in wait_serial to [functional][u] test fails in sysstat - Possible false negative in wait_serial
- Target version set to Milestone 22
This is an autogenerated message for openQA integration by the openqa_review script:
This bug is still referenced in a failing openQA test: extra_tests_in_textmode
https://openqa.suse.de/tests/2347166
This is an autogenerated message for openQA integration by the openqa_review script:
This bug is still referenced in a failing openQA test: extra_tests_in_textmode
https://openqa.suse.de/tests/2375854
- Priority changed from Normal to High
Next task: Find "last good"
Bumping to "High" as used as label in currently failing tests
okurz wrote:
Next task: Find "last good"
Bumping to "High" as used as label in currently failing tests
it was introduced with 226bd65943d21e371f8308e35d7cd9010f26f5d4 on Nov 16, 2018
so I doubt there will be any last good
- Description updated (diff)
- Status changed from New to Workable
- Status changed from Workable to In Progress
- Assignee set to szarate
- Precedes action #46919: [functional][u][svirt][sporadic] auto_review:"IO::Socket::INET: connect: Connection timed out" added
- Description updated (diff)
Testing on 3 different machines (s390 included), gave the same result, however while looking well at the output, it seems that there's an extra "\n" in the test...
LC_TIME='C' pidstat |awk '{print $3}' | head | tail -n 1 | hexdump -c
0000000 0 \n
0000002
pid can vary :), didn't look too much in the test when I reported it...
The regexp is the culprit... sub { m/[1-999]/ }... pr is comming it's way
- Status changed from In Progress to Feedback
- Subject changed from [functional][u] test fails in sysstat - Possible false negative in wait_serial to [functional][u] test fails in sysstat - regular expression for pidstat test is incorrect
- Target version changed from Milestone 22 to Milestone 23
- Status changed from Feedback to Resolved
Also available in: Atom
PDF