Commit afe35cde6 added additional actions to test 33, but forgot to reset the image between tests. As a result, './check -nbd 33' fails because the qemu-nbd process from the first half is still occupying the port, preventing the second half from starting a new qemu-nbd process. Worse, the failure leaves a rogue qemu-nbd process behind even after the test fails, which causes knock-on failures to later tests that also want to start qemu-nbd.
Reported-by: Max Reitz <mre...@redhat.com> Signed-off-by: Eric Blake <ebl...@redhat.com> Message-Id: <20180312211156.452139-1-ebl...@redhat.com> Reviewed-by: Max Reitz <mre...@redhat.com> --- tests/qemu-iotests/033 | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/qemu-iotests/033 b/tests/qemu-iotests/033 index a1d8357331d..ee8a1338bbd 100755 --- a/tests/qemu-iotests/033 +++ b/tests/qemu-iotests/033 @@ -105,6 +105,7 @@ for align in 512 4k; do done done +_cleanup_test_img # Trigger truncate that would shrink qcow2 L1 table, which is done by # clearing one entry (8 bytes) with bdrv_co_pwrite_zeroes() -- 2.14.3