Martin Sivák has submitted this change and it was merged. Change subject: Make sure VDSM had enough time to start before failing ...................................................................... Make sure VDSM had enough time to start before failing This improves things on two fronts: - Using After in the systemd unit instructs the init system to make sure the hosted engine service is started after the vdsm process, eliminating one race condition. - Moving the initial ServiceNotUp handling to the _run_agent method allows us to try multiple times before we fail. Change-Id: Ie23b110b3018c88656b2119dbeb24d431e922bdf Bug-Url: https://bugzilla.redhat.com/show_bug.cgi?id=1267511 Signed-off-by: Martin Sivak <msivak(a)redhat.com&gt; --- M initscripts/ovirt-ha-agent.service M ovirt_hosted_engine_ha/agent/agent.py M ovirt_hosted_engine_ha/agent/hosted_engine.py 3 files changed, 11 insertions(+), 11 deletions(-) Approvals: Martin Sivák: Verified; Looks good to me, approved; Passed CI tests Simone Tiraboschi: Looks good to me, but someone else must approve -- To view, visit https://gerrit.ovirt.org/46831 To unsubscribe, visit https://gerrit.ovirt.org/settings Gerrit-MessageType: merged Gerrit-Change-Id: Ie23b110b3018c88656b2119dbeb24d431e922bdf Gerrit-PatchSet: 2 Gerrit-Project: ovirt-hosted-engine-ha Gerrit-Branch: master Gerrit-Owner: Martin Sivák <msivak(a)redhat.com&gt; Gerrit-Reviewer: Jenkins CI Gerrit-Reviewer: Martin Sivák <msivak(a)redhat.com&gt; Gerrit-Reviewer: Simone Tiraboschi <stirabos(a)redhat.com&gt; Gerrit-Reviewer: automation(a)ovirt.org