Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: Linux-HA: Users

attribute or variable to detect failed resource

 

 

Linux-HA users RSS feed   Index | Next | Previous | View Threaded


smahesh at alcatel-lucent

Sep 14, 2009, 2:31 AM

Post #1 of 4 (507 views)
Permalink
attribute or variable to detect failed resource

Hi group,


I have lsb (/etc/init.d/* service scripts) resources in my two node cluster. Is there a variable or attribute like 'pingd' or '#uname' that I can use to check if the lsb resource has failed?

<rsc_location id="service_running" rsc="res1">
<rule id="service_running_rule" score="-INFINITY" Boolean_op="or">
<expression id="service_running_rule_expr_undefined" attribute="SVC_FAILED" operation="not_defined"/>
<expression id=" service_running_rule_expr_zero" attribute="SVC_FAILED" operation="gt" value="0"/>
</rule>
</rsc_location>


Regards,
Mahesh
_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems


dejanmm at fastmail

Sep 14, 2009, 4:12 AM

Post #2 of 4 (470 views)
Permalink
Re: attribute or variable to detect failed resource [In reply to]

Hi,

On Mon, Sep 14, 2009 at 04:31:26AM -0500, MAHESH, SIDDACHETTY M (SIDDACHETTY M) wrote:
> Hi group,
>
>
> I have lsb (/etc/init.d/* service scripts) resources in my two node cluster. Is there a variable or attribute like 'pingd' or '#uname' that I can use to check if the lsb resource has failed?

No. Why would you want that? That's the basic cluster
functionality, to deal with resources which failed.

Thanks,

Dejan

> <rsc_location id="service_running" rsc="res1">
> <rule id="service_running_rule" score="-INFINITY" Boolean_op="or">
> <expression id="service_running_rule_expr_undefined" attribute="SVC_FAILED" operation="not_defined"/>
> <expression id=" service_running_rule_expr_zero" attribute="SVC_FAILED" operation="gt" value="0"/>
> </rule>
> </rsc_location>
>
>
> Regards,
> Mahesh
> _______________________________________________
> Linux-HA mailing list
> Linux-HA [at] lists
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems


smahesh at alcatel-lucent

Sep 14, 2009, 11:20 PM

Post #3 of 4 (451 views)
Permalink
Re: attribute or variable to detect failed resource [In reply to]

Hi,

I have a two node cluster. I have two resources - VIP (virtual IP) and APP (res1). The APP resource has to be co-located with VIP resource. Since VIP is the primary resource, I want to move the VIP+APP to the other node in the cluster when APP fails.

Is there a clean way to do this? I was hoping that there would be a property/attribute that tells me that APP is down/failed state and VIP can have location constraints based on this attribute.

Currently, if APP fails, it is down on the current node and does not migrate to the other node because of the VIP and APP co-location constraint.

Regards,
Mahesh



-----Original Message-----
From: linux-ha-bounces [at] lists [mailto:linux-ha-bounces [at] lists] On Behalf Of Dejan Muhamedagic
Sent: Monday, September 14, 2009 4:43 PM
To: General Linux-HA mailing list
Subject: Re: [Linux-HA] attribute or variable to detect failed resource

Hi,

On Mon, Sep 14, 2009 at 04:31:26AM -0500, MAHESH, SIDDACHETTY M (SIDDACHETTY M) wrote:
> Hi group,
>
>
> I have lsb (/etc/init.d/* service scripts) resources in my two node cluster. Is there a variable or attribute like 'pingd' or '#uname' that I can use to check if the lsb resource has failed?

No. Why would you want that? That's the basic cluster
functionality, to deal with resources which failed.

Thanks,

Dejan

> <rsc_location id="service_running" rsc="res1">
> <rule id="service_running_rule" score="-INFINITY" Boolean_op="or">
> <expression id="service_running_rule_expr_undefined" attribute="SVC_FAILED" operation="not_defined"/>
> <expression id=" service_running_rule_expr_zero" attribute="SVC_FAILED" operation="gt" value="0"/>
> </rule>
> </rsc_location>
>
>
> Regards,
> Mahesh
> _______________________________________________
> Linux-HA mailing list
> Linux-HA [at] lists
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems


dejanmm at fastmail

Sep 15, 2009, 3:43 AM

Post #4 of 4 (448 views)
Permalink
Re: attribute or variable to detect failed resource [In reply to]

Hi,

On Tue, Sep 15, 2009 at 01:20:19AM -0500, MAHESH, SIDDACHETTY M (SIDDACHETTY M) wrote:
> Hi,
>
> I have a two node cluster. I have two resources - VIP
> (virtual IP) and APP (res1). The APP resource has to be
> co-located with VIP resource. Since VIP is the primary
> resource, I want to move the VIP+APP to the other node in the
> cluster when APP fails.
>
> Is there a clean way to do this? I was hoping that there
> would be a property/attribute that tells me that APP is
> down/failed state and VIP can have location constraints based
> on this attribute.
>
> Currently, if APP fails, it is down on the current node and
> does not migrate to the other node because of the VIP and APP
> co-location constraint.

I'm really not sure what you're talking about. If a resource
fails, then the cluster (CRM) should try to do something about
it, i.e. either restart it or try to start it on another node.
You can control that, i.e. restart/migrate decision using
resource-stickiness and migrate-threshold. Take a look at the
pacemaker documentation.

Thanks,

Dejan

> Regards,
> Mahesh
>
>
>
> -----Original Message-----
> From: linux-ha-bounces [at] lists [mailto:linux-ha-bounces [at] lists] On Behalf Of Dejan Muhamedagic
> Sent: Monday, September 14, 2009 4:43 PM
> To: General Linux-HA mailing list
> Subject: Re: [Linux-HA] attribute or variable to detect failed resource
>
> Hi,
>
> On Mon, Sep 14, 2009 at 04:31:26AM -0500, MAHESH, SIDDACHETTY M (SIDDACHETTY M) wrote:
> > Hi group,
> >
> >
> > I have lsb (/etc/init.d/* service scripts) resources in my two node cluster. Is there a variable or attribute like 'pingd' or '#uname' that I can use to check if the lsb resource has failed?
>
> No. Why would you want that? That's the basic cluster
> functionality, to deal with resources which failed.
>
> Thanks,
>
> Dejan
>
> > <rsc_location id="service_running" rsc="res1">
> > <rule id="service_running_rule" score="-INFINITY" Boolean_op="or">
> > <expression id="service_running_rule_expr_undefined" attribute="SVC_FAILED" operation="not_defined"/>
> > <expression id=" service_running_rule_expr_zero" attribute="SVC_FAILED" operation="gt" value="0"/>
> > </rule>
> > </rsc_location>
> >
> >
> > Regards,
> > Mahesh
> > _______________________________________________
> > Linux-HA mailing list
> > Linux-HA [at] lists
> > http://lists.linux-ha.org/mailman/listinfo/linux-ha
> > See also: http://linux-ha.org/ReportingProblems
> _______________________________________________
> Linux-HA mailing list
> Linux-HA [at] lists
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
> _______________________________________________
> Linux-HA mailing list
> Linux-HA [at] lists
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Linux-HA users RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.