| Message ID | 20240124133538.2784097-1-saif.abrar@linux.vnet.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | tests/qemu-iotests: Bump timeout of iothreads-stream test | expand |
Am 24.01.2024 um 14:35 hat Saif Abrar geschrieben: > Current value of timeout=0.1 is sometimes too tight. > Bump timeout to 0.3 to avoid test failures. > > Signed-off-by: Saif Abrar <saif.abrar@linux.vnet.ibm.com> This is not the real problem. The real problem is that the timeout makes the test fail because we're catching the wrong TimeoutError class. What it really should do is issue another query-jobs and go back to waiting. I actually sent a patch earlier today to fix this: https://lists.gnu.org/archive/html/qemu-block/2024-01/msg00353.html Kevin
diff --git a/tests/qemu-iotests/tests/iothreads-stream b/tests/qemu-iotests/tests/iothreads-stream index 503f221f16..32e46e044e 100755 --- a/tests/qemu-iotests/tests/iothreads-stream +++ b/tests/qemu-iotests/tests/iothreads-stream @@ -63,7 +63,7 @@ with iotests.FilePath('disk1.img') as base1_path, \ finished = 0 while True: try: - ev = vm.event_wait('JOB_STATUS_CHANGE', timeout=0.1) + ev = vm.event_wait('JOB_STATUS_CHANGE', timeout=0.3) if ev is not None and ev['data']['status'] == 'null': finished += 1 # The test is done once both jobs are gone
Current value of timeout=0.1 is sometimes too tight. Bump timeout to 0.3 to avoid test failures. Signed-off-by: Saif Abrar <saif.abrar@linux.vnet.ibm.com> --- tests/qemu-iotests/tests/iothreads-stream | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)