On Nov 27, 2007, at 10:20 AM, HIDEO YAMAUCHI wrote:
Hi,
in this case the cluster can't stop the resource because the resource
isnt valid according to the RA (since you didnt add a parameter)
we tried to stop it and the RA reported failure.
without stonith enabled there is nothing more the cluster can do
In this case an operator does not have the choice except KILL.
i guess you could repair the resource definition and then use
crm_resource -C
that should probably work
In GUI, is not the check of the required parameter necessary to
prevent such a mistake?
quite possibly
/me is utterly ignorant of the gui and prefers it that way
http://www.gossamer-threads.com/lists/linuxha/dev/43184
Regards,
Hideo Yamauchi.
--- Andrew Beekhof <[EMAIL PROTECTED]> wrote:
On Nov 27, 2007, at 8:11 AM, HIDEO YAMAUCHI wrote:
Hi,
I tested the addition of the resource from GUI.
First cib.xml does not have the resource.
I used the latest development edition.(Heartbeat-Dev-68a5c0c53078)
I operated it in order of next.
1)I am connected to the DC node in GUI.
2)I add a FileSystem resource without appointing a parameter.
3)The FileSystem resource is displayed with unmanaged in GUI.
4)I stop service of Heartbeat.
But, Heartbeat does not stop.
The last following log appeared.
----------------------------------------------------------------
crmd[4029]: 2007/11/27_16:00:26 info: do_shutdown: Terminating the
tengine
crmd[4029]: 2007/11/27_16:00:26 info: stop_subsystem: Sent -TERM to
tengine: [4032]
crmd[4029]: 2007/11/27_16:00:26 info: do_shutdown: Waiting for
subsystems to exit
crmd[4029]: 2007/11/27_16:00:26 info: do_shutdown: All subsystems
stopped, continuing
crmd[4029]: 2007/11/27_16:00:26 info: crmdManagedChildDied: Process
tengine:[4032] exited (signal=0,
exitcode=0)
crmd[4029]: 2007/11/27_16:00:26 info: crmdManagedChildDied: Process
pengine:[4033] exited (signal=0,
exitcode=0)
crmd[4029]: 2007/11/27_16:00:26 info: do_shutdown: All subsystems
stopped, continuing
crmd[4029]: 2007/11/27_16:00:26 ERROR: verify_stopped: Resource
resource_ was active at shutdown. You
may ignore this error if it is unmanaged.
crmd[4029]: 2007/11/27_16:00:26 info: process_client_disconnect:
Received HUP from tengine:[-1]
crmd[4029]: 2007/11/27_16:00:26 info: process_client_disconnect:
Received HUP from pengine:[-1]
crmd[4029]: 2007/11/27_16:00:26 ERROR: verify_stopped: Resource
resource_ was active at shutdown. You
may ignore this error if it is unmanaged.
crmd[4029]: 2007/11/27_16:05:06 ERROR: verify_stopped: Resource
resource_ was active at shutdown. You
may ignore this error if it is unmanaged.
----------------------------------------------------------------
In the case of unmanaged, is there a right procedure to stop?
in this case the cluster can't stop the resource because the resource
isnt valid according to the RA (since you didnt add a parameter)
we tried to stop it and the RA reported failure.
without stonith enabled there is nothing more the cluster can do
Or is this bug?
I attach a collection of hb_report result.
Regards,
Hideo Yamauchi.
<report_Dev.tar.gz>_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
