Whamcloud - gitweb
LU-17919 tests: wait to resolve ENOSPC in sanity 398l 59/55359/4
authorPatrick Farrell <paf0187@gmail.com>
Fri, 7 Jun 2024 15:49:40 +0000 (11:49 -0400)
committerOleg Drokin <green@whamcloud.com>
Tue, 25 Jun 2024 03:31:25 +0000 (03:31 +0000)
Test 398l does not wait to clear up the ENOSPC it induces,
so sometimes it causes 398m to fail with ENOSPC.

Wait for deletes to resolve this.

OCI-bug-id: LFS-288

Note on test-parameters - we can't 'REPEAT' a pair of
tests, it would run 398l over and over and then 398m,
which doesn't test what we need to test.  So instead we
just create 5 sessions like this.

Test-Parameters: testlist=sanity envdefinitions=ONLY="398l 398m"
Test-Parameters: testlist=sanity envdefinitions=ONLY="398l 398m"
Test-Parameters: testlist=sanity envdefinitions=ONLY="398l 398m"
Test-Parameters: testlist=sanity envdefinitions=ONLY="398l 398m"
Test-Parameters: testlist=sanity envdefinitions=ONLY="398l 398m"
Test-Parameters: trivial
Signed-off-by: Patrick Farrell <patrick.farrell@oracle.com>
Change-Id: I2fcc1069a0304bc6edfa576331b6255289b71b98
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/55359
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Reviewed-by: Shaun Tancheff <shaun.tancheff@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
lustre/tests/sanity.sh

index b6df61d..a2117d5 100755 (executable)
@@ -28594,6 +28594,9 @@ run_test 398k "test enospc on first stripe"
 test_398l() { #  LU-13798
        wait_delete_completed
        wait_mds_ost_sync
+       # Clean up after the test
+       stack_trap wait_delete_completed
+       stack_trap wait_mds_ost_sync
 
        # 4 stripe file; we will cause out of space on OST0
        # Note the 1M stripe size and the > 1M i/o size mean this ENOSPC