Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 922795

Summary:	systemd-vdsmd reconfigure should not start vdsmd
Product:	[Retired] oVirt	Reporter:	Alon Bar-Lev <alonbl>
Component:	vdsm	Assignee:	Yaniv Bronhaim <ybronhei>
Status:	CLOSED DUPLICATE	QA Contact:	Haim <hateya>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	unspecified	CC:	abaron, acathrow, alonbl, bazulay, danken, iheim, jkt, mgoldboi, yeylon
Target Milestone:	---
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:	infra
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2013-09-02 12:06:51 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Alon Bar-Lev 2013-03-18 14:33:38 UTC

Description of problem:
# /lib/systemd/systemd-vdsmd reconfigure

vdsm is started.

fedora-18
vdsm-4.10.3-7.fc18.x86_64

How reproducible:
# systemctl stop vdsmd
# /lib/systemd/systemd-vdsmd reconfigure
# sleep 10
# systemctl status vdsmd
  
Actual results:
vdsm up

Expected results:
vdsm down

Additional info:
I almost sure this was not the case a while ago.

Comment 1 Yaniv Bronhaim 2013-08-27 18:24:49 UTC

Over systemd vdsm service is depended on libvirtd service, so when libvirtd is restarted it starts also vdsm. After forcing reconfiguration, we restart libvirtd service.

Each service restart of any of the services that mentioned under vdsmd.service Requires (multipathd.servicem libvirtd.service, time-sync.target, iscsid.service, rpcbind.service, supervdsmd.service) leads to vdsmd start.

Is it wrong that way? over initctl it doesn't happen, not sure about upstat, but in vdsmd init script we just take care of starting the required services.

Comment 2 Alon Bar-Lev 2013-08-27 18:41:01 UTC

(In reply to Yaniv Bronhaim from comment #1)
> Is it wrong that way? over initctl it doesn't happen, not sure about upstat,
> but in vdsmd init script we just take care of starting the required services.

Well yes... this is unexpected.

If we have service B[up] and C[down] that depend on service A.

What is expected when restarting A?

For some extend it is logical:
1. stop B
2. stop A
3. start A
4. start B (as it was up when process began).

However it does not make any sense that C is effected, as it imposes no conflict to the process of restarting A.

Comment 3 Yaniv Bronhaim 2013-08-27 19:26:50 UTC

The logic is to send sigterm to the required services and then service start, so that's how it works.
I'll try to find way to avoid starting stopped services on systemd dependency mechanism

Comment 4 Yaniv Bronhaim 2013-09-02 12:06:51 UTC


*** This bug has been marked as a duplicate of bug 1003575 ***