Closed hishamhm closed 3 months ago
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 90.36951%. Comparing base (
223a346
) to head (a4e425c
).
Nice one!
Nice! One of the attach
test cases seems to be flaky but it isn't clear which one, this failure happened a few times in the last couple days...
One of the attach test cases seems to be flaky but it isn't clear which one, this failure happened a few times in the last couple days...
Looks ilke it's TEST 8
(the only one that has shutdown_error_log
and skip_no_debug
). When I run it locally with Valgrind, I get that error on it too. If I run this loop:
i=0; while TEST_NGINX_USE_VALGRIND=1 util/test.sh t/04-openresty/ffi/103-proxy_wasm_attach.t; do i=$[i+1]; echo $i; done
I get this failure in under 20 iterations:
t/04-openresty/ffi/103-proxy_wasm_attach.t .. 57/56
# Failed test 't/04-openresty/ffi/103-proxy_wasm_attach.t TEST 8: attach() - plan is garbage collected after execution - pattern "(?^:\[debug\] .*? wasm freeing plan)" should match a line in error.log'
# at /usr/share/perl/5.34/Test/Builder.pm line 193.
# Looks like you planned 56 tests but ran 57.
# Looks like you failed 1 test of 57 run.
t/04-openresty/ffi/103-proxy_wasm_attach.t .. Dubious, test returned 1 (wstat 256, 0x100)
All 56 subtests passed
(less 56 skipped subtests: 0 okay)
Test Summary Report
-------------------
t/04-openresty/ffi/103-proxy_wasm_attach.t (Wstat: 256 Tests: 57 Failed: 1)
Failed test: 57
Non-zero exit status: 1
Parse errors: Bad plan. You planned 56 tests but ran 57.
Files=1, Tests=57, 9 wallclock secs ( 0.02 usr 0.00 sys + 0.09 cusr 0.02 csys = 0.13 CPU)
Result: FAIL
If I add --- ONLY
to TEST 8, the failure seems to happen right away (almost every time?).
If I remove TEST 8, the loop above runs to over 200 iterations with no errors.
Flaky test fixed in #553.
See https://github.com/Kong/ngx_wasm_module/pull/530#discussion_r1635526128