Hi,

On Mon, Nov 26, 2007 at 08:04:17AM -0800, Jason Snyder wrote:
> Ah, its cibadmin, not cibadm.  That explains a lot ;-)
> 
> Here is the config that I am having trouble with mounting the drbd partition 
> (it comes up primary/secondary correctly according to /proc/drbd and I can 
> manually mount it with `mount -o noatime /dev/drbd0 /ha/apache`):
> 
>  <cib admin_epoch="0" have_quorum="true" ignore_dtd="false" num_peers="2" 
> cib_feature_revision="1.3" ccm_transition="2" generated="true" 
> dc_uuid="9934e766-a79a-42f2-915a-3134009b3793" epoch="102" num_updates="3" 
> cib-last-written="Sun Nov 25 19:17:24 2007">
>    <configuration>
>      <crm_config>
>        <cluster_property_set id="cib-bootstrap-options">
>          <attributes>
>            <nvpair name="last-lrm-refresh" 
> id="cib-bootstrap-options-last-lrm-refresh" value="1196046991"/>
>          </attributes>
>        </cluster_property_set>
>      </crm_config>
>      <nodes>
>        <node id="9934e766-a79a-42f2-915a-3134009b3793" uname="rmirror1" 
> type="normal"/>
>        <node id="54f1e969-c41b-45c9-b17f-08fd5a1ecfd4" uname="rmirror0" 
> type="normal"/>
>      </nodes>
>      <resources>
>        <group ordered="true" collocated="true" id="apache_group">
>          <primitive class="ocf" type="IPaddr2" provider="heartbeat" 
> id="apache_IpAddr1">
>            <instance_attributes id="apache_IpAddr1_instance_attrs">
>              <attributes>
>                <nvpair name="target_role" id="apache_IpAddr1_target_role" 
> value="started"/>
>                <nvpair id="2c77185e-f341-4736-81f3-84b0982b5432" name="ip" 
> value="192.168.1.6"/>
>                <nvpair id="baa8e9b6-99d2-47c1-adeb-598830249605" name="nic" 
> value="eth0:1"/>
>              </attributes>
>            </instance_attributes>
>          </primitive>
>          <instance_attributes id="apache_group_instance_attrs">
>            <attributes>
>              <nvpair id="apache_group_target_role" name="target_role" 
> value="started"/>
>            </attributes>
>          </instance_attributes>
>          <primitive class="ocf" type="apache" provider="heartbeat" 
> id="apache_web">
>            <instance_attributes id="apache_web_instance_attrs">
>              <attributes>
>                <nvpair name="target_role" id="apache_web_target_role" 
> value="started"/>
>                <nvpair name="configfile" 
> id="014e7881-1df7-460a-bc82-df40f1b9defe" value="/etc/apache2/apache2.conf"/>
>              </attributes>
>            </instance_attributes>
>          </primitive>
>        </group>
>        <master_slave id="ms-drbd0">
>          <meta_attributes id="ma-ms-drbd0">
>            <attributes>
>              <nvpair id="ma-ms-drbd0-1" name="clone_max" value="2"/>
>              <nvpair id="ma-ms-drbd0-2" name="clone_node_max" value="1"/>
>              <nvpair id="ma-ms-drbd0-3" name="master_max" value="1"/>
>              <nvpair id="ma-ms-drbd0-4" name="master_node_max" value="1"/>
>              <nvpair id="ma-ms-drbd0-5" name="notify" value="yes"/>
>              <nvpair id="ma-ms-drbd0-6" name="globally_unique" value="false"/>
>              <nvpair id="ma-ms-drbd0-7" name="target_role" value="stopped"/>
>            </attributes>
>          </meta_attributes>
>          <primitive class="ocf" provider="heartbeat" type="drbd" id="drbd0">
>            <instance_attributes id="ia-drbd0">
>              <attributes>
>                <nvpair id="ia-drbd0-1" name="drbd_resource" value="drbd0"/>
>                <nvpair name="target_role" id="drbd0:0_target_role" 
> value="started"/>
>              </attributes>
>            </instance_attributes>
>          </primitive>
>          <instance_attributes id="ms-drbd0_instance_attrs">
>            <attributes/>
>          </instance_attributes>
>        </master_slave>
>        <primitive class="ocf" provider="heartbeat" type="Filesystem" id="fs0">
>          <meta_attributes id="ma-fs0">
>            <attributes>
>              <nvpair name="target_role" id="ma-fs0-1" value="started"/>
>            </attributes>
>          </meta_attributes>
>          <instance_attributes id="ia-fs0">
>            <attributes>
>              <nvpair id="ia-fs0-1" name="fstype" value="reiserfs"/>
>              <nvpair id="ia-fs0-2" name="directory" value="/ha/apache"/>
>              <nvpair id="ia-fs0-3" name="device" value="/dev/drbd0"/>
>            </attributes>
>          </instance_attributes>
>        </primitive>
>      </resources>
>      <constraints>
>        <rsc_order id="drbd0_before_fs0" from="fs0" action="start" 
> to="ms-drbd0" to_action="promote"/>
>        <rsc_colocation id="fs0_on_drbd0" to="ms-drbd0" to_role="master" 
> from="fs0" score="infinity"/>
>      </constraints>
>    </configuration>
>  </cib>
> 
> 
> Logs are pretty huge and I am having trouble zeroing in on what
> you need.  Any tips on grabbing something useful to see?

Look for ERROR: and WARN: strings. The best would be to just grab
all logs, compress, and attach them. If your version has
hb_report installed, you can use that.

Thanks,

Dejan

> 
> Thanks.
> 
> On Monday 26 November 2007, Andrew Beekhof wrote:
> > configuration? logs?
> > kinda hard to help without either of those two things...
> > 
> > On Nov 25, 2007, at 7:06 PM, Jason Snyder wrote:
> > 
> > > Software: Ubuntu 7.10 Gusty Gibbon (virtual), Heartbeat 2.1.2 (using  
> > > CRM), DRBD 8.03 (all are versions that come with Ubuntu 7.10), and  
> > > VMWare 1.0.3 hosting the virtual servers.
> > >
> > > What I have working so far:
> > > 1. DRBD is compiled in and configured.  Can use either manual  
> > > control or the CRM to control it.
> > > 2. Virtual IP (on eth0:1) and a web server in a group that seem to  
> > > follow each other from one node to the next successfully under CRM  
> > > control.
> > >
> > > What doesn't seem to work: Mounting of file-systems with CRM.  Every  
> > > method I have tried to add it to the cib.xml file has lead to a  
> > > resource with an exclamation mark and a status of unmanaged resource  
> > > in haclient.py.  Even following the examples on the Linux-HA.org  
> > > site exactly fails.  (I also noticed that options on the website  
> > > don't even come close to matching up with what haclient.py offers  
> > > and even seemingly basic operations seemed to get fudged beyond  
> > > usability with this interface with seemingly no recourse but to  
> > > manually edit the files while heartbeat is offline and delete *.sig  
> > > before starting back up, which is rather vexing if you forget a step  
> > > and then all of your changes are gone when you restart heartbeat,  
> > > unless of course I am doing something terribly wrong in my methods.)
> > 
> > cibadmin -Q > tmp.xml
> > vi tmp.xml
> > cibadmin -R -x tmp.xml
> > 
> > no need to stop the cluster or delete .sig files
> > 
> > >
> > >
> > > What I am ultimately after is the ability to have an active/active  
> > > 64-bit dual server configuration (with each server running different  
> > > services [not clustered services]) that supports monitoring and  
> > > recovery of local services and automatic failover for high  
> > > avalability.  The only part I have left in this scheme to make work  
> > > is the actual resource manager and at that it seems getting it to  
> > > mount a file system and manage it properly is where I am falling  
> > > short on this scheme.  Also I need to do more thorough testing once  
> > > I have it all setup to make sure it really does what it is supposed  
> > > to.
> > >
> > > _______________________________________________
> > > 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

Reply via email to