| Message ID | 1599761867-122691-1-git-send-email-hzhou@ovn.org |
|---|---|
| State | Accepted |
| Headers | show |
| Series | [ovs-dev] ovn-controller.at: Fix test for debug/delay-nb-cfg-report. | expand |
On 9/10/20 8:17 PM, Han Zhou wrote: > In the test case for debug/delay-nb-cfg-report, it sets the delay as > 2 seconds, and firstly triggered a sync with wait timeout being 1 sec > and expects the wait timedout, and then it immediately started another > sync with wait timeout being 3 sec, expecting it not timing out. > > However, since 2 + 2 = 1 + 3, there is a good chance that the second > wait would still time out, so the test fails occasionally. This patch > fixes it by sleep another 1 sec to let the 1st delay complete before > starting the second sync and wait. > > Reported-by: Ilya Maximets <i.maximets@ovn.org> > Fixes: a8182e21484 ("ovn-controller: Unix command debug/delay-nb-cfg-report.") > Signed-off-by: Han Zhou <hzhou@ovn.org> > --- Without this patch test fails on Travis more frequently than succeeds. With this patch tests survived more than 1000 iterations on Travis without failure. Tested-by: Ilya Maximets <i.maximets@ovn.org> > tests/ovn-controller.at | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/tests/ovn-controller.at b/tests/ovn-controller.at > index 7ed5216..d806134 100644 > --- a/tests/ovn-controller.at > +++ b/tests/ovn-controller.at > @@ -402,6 +402,9 @@ AT_CHECK([ovn-appctl -t ovn-controller debug/delay-nb-cfg-report 2], [0], > > AT_FAIL_IF([ovn-nbctl --timeout=1 --wait=hv sync]) > > +# sleep another 1 sec so that the original 2 sec sleep is done in ovn-controller > +sleep 1 > + > AT_CHECK([ovn-nbctl --timeout=3 --wait=hv sync]) > > AT_CHECK([ovn-appctl -t ovn-controller debug/delay-nb-cfg-report 0], [0], >
On Thu, Sep 10, 2020 at 12:48 PM Ilya Maximets <i.maximets@ovn.org> wrote: > > On 9/10/20 8:17 PM, Han Zhou wrote: > > In the test case for debug/delay-nb-cfg-report, it sets the delay as > > 2 seconds, and firstly triggered a sync with wait timeout being 1 sec > > and expects the wait timedout, and then it immediately started another > > sync with wait timeout being 3 sec, expecting it not timing out. > > > > However, since 2 + 2 = 1 + 3, there is a good chance that the second > > wait would still time out, so the test fails occasionally. This patch > > fixes it by sleep another 1 sec to let the 1st delay complete before > > starting the second sync and wait. > > > > Reported-by: Ilya Maximets <i.maximets@ovn.org> > > Fixes: a8182e21484 ("ovn-controller: Unix command debug/delay-nb-cfg-report.") > > Signed-off-by: Han Zhou <hzhou@ovn.org> > > --- > > Without this patch test fails on Travis more frequently than succeeds. > With this patch tests survived more than 1000 iterations on Travis without > failure. > > Tested-by: Ilya Maximets <i.maximets@ovn.org> > Thanks Ilya. I applied the fix. > > tests/ovn-controller.at | 3 +++ > > 1 file changed, 3 insertions(+) > > > > diff --git a/tests/ovn-controller.at b/tests/ovn-controller.at > > index 7ed5216..d806134 100644 > > --- a/tests/ovn-controller.at > > +++ b/tests/ovn-controller.at > > @@ -402,6 +402,9 @@ AT_CHECK([ovn-appctl -t ovn-controller debug/delay-nb-cfg-report 2], [0], > > > > AT_FAIL_IF([ovn-nbctl --timeout=1 --wait=hv sync]) > > > > +# sleep another 1 sec so that the original 2 sec sleep is done in ovn-controller > > +sleep 1 > > + > > AT_CHECK([ovn-nbctl --timeout=3 --wait=hv sync]) > > > > AT_CHECK([ovn-appctl -t ovn-controller debug/delay-nb-cfg-report 0], [0], > > >
diff --git a/tests/ovn-controller.at b/tests/ovn-controller.at index 7ed5216..d806134 100644 --- a/tests/ovn-controller.at +++ b/tests/ovn-controller.at @@ -402,6 +402,9 @@ AT_CHECK([ovn-appctl -t ovn-controller debug/delay-nb-cfg-report 2], [0], AT_FAIL_IF([ovn-nbctl --timeout=1 --wait=hv sync]) +# sleep another 1 sec so that the original 2 sec sleep is done in ovn-controller +sleep 1 + AT_CHECK([ovn-nbctl --timeout=3 --wait=hv sync]) AT_CHECK([ovn-appctl -t ovn-controller debug/delay-nb-cfg-report 0], [0],
In the test case for debug/delay-nb-cfg-report, it sets the delay as 2 seconds, and firstly triggered a sync with wait timeout being 1 sec and expects the wait timedout, and then it immediately started another sync with wait timeout being 3 sec, expecting it not timing out. However, since 2 + 2 = 1 + 3, there is a good chance that the second wait would still time out, so the test fails occasionally. This patch fixes it by sleep another 1 sec to let the 1st delay complete before starting the second sync and wait. Reported-by: Ilya Maximets <i.maximets@ovn.org> Fixes: a8182e21484 ("ovn-controller: Unix command debug/delay-nb-cfg-report.") Signed-off-by: Han Zhou <hzhou@ovn.org> --- tests/ovn-controller.at | 3 +++ 1 file changed, 3 insertions(+)