[Pacemaker] [Question]View order of the node of crm_mon

Andrew Beekhof andrew at beekhof.net
Wed Feb 29 06:56:02 UTC 2012 

On Wed, Feb 29, 2012 at 2:30 PM,  <nozawat at gmail.com> wrote:
> Hi Andrew,
>
>  Thank you for your patch.
>
>  After I applied a patch, and testing, the problem seemed to be solved.

Ok, I'll apply to 1.1 and we can backport it.

>  BTW, is crm_trace a function from pacemaker-1.1 ?

Correct.

>  Because an error had occurred in build, crm_trace commented out and testing.
>
> Regards,
> Tomo
>
> On Wed, 29 Feb 2012 10:13:59 +1100
> Andrew Beekhof <andrew at beekhof.net> wrote:
>
>> Could you try the following patch:
>>
>> diff --git a/lib/pengine/unpack.c b/lib/pengine/unpack.c
>> index 8890850..4fcd224 100644
>> --- a/lib/pengine/unpack.c
>> +++ b/lib/pengine/unpack.c
>> @@ -191,6 +191,15 @@ unpack_config(xmlNode * config, pe_working_set_t
>> * data_set)
>>      return TRUE;
>>  }
>>
>> +static gint sort_nodes_uname(gconstpointer a, gconstpointer b)
>> +{
>> +    const node_t *na = a;
>> +    const node_t *nb = b;
>> +
>> +    return strcmp(na->details->uname, nb->details->uname);
>> +}
>> +
>> +
>>  gboolean
>>  unpack_nodes(xmlNode * xml_nodes, pe_working_set_t * data_set)
>>  {
>> @@ -286,7 +295,7 @@ unpack_nodes(xmlNode * xml_nodes, pe_working_set_t
>> * data_set)
>>              unpack_instance_attributes(data_set->input, xml_obj,
>> XML_TAG_UTILIZATION, NULL,
>>
>> new_node->details->utilization, NULL, FALSE, data_set->now);
>>
>> -            data_set->nodes = g_list_append(data_set->nodes, new_node);
>> +            data_set->nodes = g_list_insert_sorted(data_set->nodes,
>> new_node, sort_nodes_uname);
>>              crm_trace("Done with node %s", crm_element_value(xml_obj,
>> XML_ATTR_UNAME));
>>          }
>>      }
>>
>>
>> On Wed, Feb 29, 2012 at 9:42 AM, Andrew Beekhof <andrew at beekhof.net> wrote:
>> > On Tue, Feb 28, 2012 at 2:23 PM, ?<nozawat at gmail.com> wrote:
>> >> Hi Andrew,
>> >>
>> >>> > I thought whether the configuration files could describe the condition of the sort.
>> >>>
>> >>> That seems like overkill.
>> >> I don't need me to there either.
>> >> Order of the output becomes unsettled now because it use g_hash_table.
>> >
>> > No, its a list:
>> >
>> > unpack.c
>> > ? ? ? ? ? ?data_set->nodes = g_list_append(data_set->nodes, new_node);
>> >
>> > crm_mon.c
>> > ? ?for (gIter = data_set->nodes; gIter != NULL; gIter = gIter->next) {
>> >
>> >
>> >> I think that it becomes the order of host name, and there should be unity.
>> >>
>> >> Regards,
>> >> Tomo
>> >>
>> >> On Tue, 28 Feb 2012 13:54:38 +1100
>> >> Andrew Beekhof <andrew at beekhof.net> wrote:
>> >>
>> >>> 2012/2/28 ?<nozawat at gmail.com>:
>> >>> > Hi Andrew,
>> >>> >
>> >>> >> >> Basically the order comes from how they are listed in the xml.
>> >>> >> >> We could sort it if you like.
>> >>> >> > Can I set the configuration file and a parameter?
>> >>> >>
>> >>> >> I'm confused. Can you rephrase this?
>> >>> > I thought whether the configuration files could describe the condition of the sort.
>> >>>
>> >>> That seems like overkill.
>> >>>
>> >>> > For example, I thought whether conditions like the asc order of key information or a desc order could be described.
>> >>> >
>> >>> > Regards,
>> >>> > Tomo
>> >>> >
>> >>> > On Tue, 28 Feb 2012 12:36:00 +1100
>> >>> > Andrew Beekhof <andrew at beekhof.net> wrote:
>> >>> >
>> >>> >> 2012/2/28 ?<nozawat at gmail.com>:
>> >>> >> > Hi Andrew,
>> >>> >> >
>> >>> >> >> Basically the order comes from how they are listed in the xml.
>> >>> >> >> We could sort it if you like.
>> >>> >> > Can I set the configuration file and a parameter?
>> >>> >>
>> >>> >> I'm confused. Can you rephrase this?
>> >>> >>
>> >>> >> > Or do I edit xml file?
>> >>> >>
>> >>> >> That would also work.
>> >>> >>
>> >>> >> >
>> >>> >> > Regards,
>> >>> >> > Tomo
>> >>> >> >
>> >>> >> > On Mon, 27 Feb 2012 23:01:56 +1100
>> >>> >> > Andrew Beekhof <andrew at beekhof.net> wrote:
>> >>> >> >
>> >>> >> >> On Wed, Feb 22, 2012 at 3:47 PM, nozawat <nozawat at gmail.com> wrote:
>> >>> >> >> > Hi
>> >>> >> >> >
>> >>> >> >> > My environment is as follows.
>> >>> >> >> > Pacemaker-1.0.11+Heartbeat-3.0.5
>> >>> >> >> > I constitute a cluster in node-A and node-B.
>> >>> >> >> >
>> >>> >> >> > There is not the file in the following directories.
>> >>> >> >> > /var/lib/heartbeat
>> >>> >> >> > /var/lib/heartbeat/crm
>> >>> >> >> >
>> >>> >> >> > Therefore indication of crm_mon is as follows when I start 2 nodes at
>> >>> >> >> > the same time.
>> >>> >> >> > ------ crm_mon -----
>> >>> >> >> > ============
>> >>> >> >> > Last updated: XXX XXX XX XX:XX:XX XXXX
>> >>> >> >> > Stack: Heartbeat
>> >>> >> >> > Current DC: node-B (841b433a-70f1-483f-a80f-790f2026fd9d) - partition
>> >>> >> >> > with quorum
>> >>> >> >> > Version: 1.0.11-1554a83db0d3c3e546cfd3aaff6af1184f79ee87
>> >>> >> >> > 2 Nodes configured, unknown expected votes
>> >>> >> >> > 0 Resources configured.
>> >>> >> >> > ============
>> >>> >> >> >
>> >>> >> >> > Online: [ node-B node-A ]
>> >>> >> >> > --------------------------
>> >>> >> >> > View order may be reversed with other machines.
>> >>> >> >> > The indication order seems to be based on information of cib.
>> >>> >> >> >
>> >>> >> >> > Therefore it is a question.
>> >>> >> >> > What is the condition of the view order of crm_mon?
>> >>> >> >>
>> >>> >> >> Basically the order comes from how they are listed in the xml.
>> >>> >> >> We could sort it if you like.
>> >>> >> >>
>> >>> >> >> >
>> >>> >> >> > Regards,
>> >>> >> >> > Tomo
>> >>> >> >> >
>> >>> >> >> > _______________________________________________
>> >>> >> >> > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> >>> >> >> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>> >>> >> >> >
>> >>> >> >> > Project Home: http://www.clusterlabs.org
>> >>> >> >> > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> >>> >> >> > Bugs: http://bugs.clusterlabs.org
>> >>> >> >>
>> >>> >> >> _______________________________________________
>> >>> >> >> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> >>> >> >> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>> >>> >> >>
>> >>> >> >> Project Home: http://www.clusterlabs.org
>> >>> >> >> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> >>> >> >> Bugs: http://bugs.clusterlabs.org
>> >>> >> >
>> >>> >> > _______________________________________________
>> >>> >> > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> >>> >> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>> >>> >> >
>> >>> >> > Project Home: http://www.clusterlabs.org
>> >>> >> > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> >>> >> > Bugs: http://bugs.clusterlabs.org
>> >>> >>
>> >>> >> _______________________________________________
>> >>> >> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> >>> >> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>> >>> >>
>> >>> >> Project Home: http://www.clusterlabs.org
>> >>> >> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> >>> >> Bugs: http://bugs.clusterlabs.org
>> >>> >
>> >>> > _______________________________________________
>> >>> > Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> >>> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>> >>> >
>> >>> > Project Home: http://www.clusterlabs.org
>> >>> > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> >>> > Bugs: http://bugs.clusterlabs.org
>> >>>
>> >>> _______________________________________________
>> >>> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> >>> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>> >>>
>> >>> Project Home: http://www.clusterlabs.org
>> >>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> >>> Bugs: http://bugs.clusterlabs.org
>> >>
>> >>
>> >> _______________________________________________
>> >> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> >> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>> >>
>> >> Project Home: http://www.clusterlabs.org
>> >> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> >> Bugs: http://bugs.clusterlabs.org
>>
>> _______________________________________________
>> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>>
>> Project Home: http://www.clusterlabs.org
>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> Bugs: http://bugs.clusterlabs.org
>
>
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org

More information about the Pacemaker mailing list