QA (public) » openQA Project (public)

https://github.com/os-autoinst/os-autoinst/pull/2148

The PR has been merged and codecov showed 100 % coverage for baseclass.pm. So I suppose the hardest part is over.

Looks like the test for t/29-backend-generalhw.t is unstable now:

[  417s] 3: ok     1527 ms ( 0.02 usr  0.00 sys +  1.23 cusr  0.33 csys =  1.58 CPU)
[  419s] 3: [12:09:00] ./t/29-backend-amt.t ....................... 
[  419s] 3: ok 1 - backend can be created but is deprecated
[  419s] 3: ok 2 - wsman_cmdline generated
[  419s] 3: ok 3 - can call wsman
[  419s] 3: ok 4 - can call enable_solider
[  419s] 3: ok 5 - can call configure_vnc
[  419s] 3: ok 6 - can call get_power_state
[  419s] 3: ok 7 - can call is_shutdown
[  419s] 3: ok 8 - can call set_power_state
[  419s] 3: ok 9 - select_next_boot evaluates wsman command
[  419s] 3: ok 10 - can call do_start_vm
[  419s] 3: ok 11 - can call do_stop_vm
[  419s] 3: ok 12 - no (unexpected) warnings (via done_testing)
[  419s] 3: 1..12
[  419s] 3: ok     2654 ms ( 0.00 usr  0.00 sys +  2.49 cusr  0.17 csys =  2.66 CPU)
[  421s] 3:         # No tests run!
[  421s] 3: 
[  421s] 3:     #   Failed test 'No tests run for subtest "stop grabbing"'
[  421s] 3:     #   at ./t/29-backend-generalhw.t line 174.
[  421s] 3:     # Looks like you failed 1 test of 2.
[  421s] 3: 
[  421s] 3: #   Failed test 'serial grab'
[  421s] 3: #   at ./t/29-backend-generalhw.t line 177.
[  421s] 3: Can't kill('-TERM', '4096'): No such process at /home/abuild/rpmbuild/BUILD/os-autoinst-4.6.1663592796.abfa696/backend/generalhw.pm line 186
[  421s] 3: # Tests were run but no plan was declared and done_testing() was not seen.
[  421s] 3: # Looks like your test exited with 255 just after 10.
[  421s] 3: [12:09:02] ./t/29-backend-generalhw.t ................. 
…
[  421s] 3: # Subtest: serial grab
[  421s] 3:     # Subtest: capturing output
[  421s] 3:         ok 1 - serial PID assigned: 4095
[  421s] 3:         ok 2 - serial output captured
[  421s] 3:         1..2
[  421s] 3:     ok 1 - capturing output
[  421s] 3:     # Subtest: stop grabbing
[  421s] 3:         1..0
[  421s] 3:     not ok 2 - No tests run for subtest "stop grabbing"
[  421s] 3:     1..2
[  421s] 3: not ok 9 - serial grab
[  421s] 3: ok 10 - no (unexpected) warnings (via END block)
[  421s] 3: Dubious, test returned 255 (wstat 65280, 0xff00)
[  421s] 3: Failed 1/10 subtests

Not sure where this code has a race condition. I won't investigate it further until it happens another time.

Another PR for the QEMU backend: https://github.com/os-autoinst/os-autoinst/pull/2183

That should leave only pvm and ikvm left.

mkittler wrote:

Another PR for the QEMU backend: https://github.com/os-autoinst/os-autoinst/pull/2183

That should leave only pvm and ikvm left.

Yes, but we deprecated both those backends, see https://github.com/os-autoinst/os-autoinst/commit/2797e185b269cf453d09fed6350c16ea454b3c07 and https://github.com/os-autoinst/os-autoinst/commit/0918185fc4071b64a4147763726932e6679e9112 so IMHO not worth the effort to increase code coverage. How about marking them as not completely covered explicitly in the codecov config file and keep it for some months or years.

Sounds good - although it seems that our codecov config file is structured the other way around. We explicitly mention fully covered files.