[Linux-cluster] Ricci doesn't work

[Chip Burke]

Here we goŠ

node1:
ricci-0.16.2-55.el6.x86_64


node2:
ricci-0.16.2-55.el6.x86_64

[root at xanadunode2 ~]# service ricci stop
Shutting down ricci:                                       [  OK  ]
[root at xanadunode2 ~]# service ricci start
Starting ricci:                                            [  OK  ]
[root at xanadunode2 ~]# ccs -d -h localhost --getconf
***Sending to ricci server:
<ricci function="process_batch" async="false" version="1.0"><batch><module
name="cluster"><request API_version="1.0"><function_call
name="get_cluster.conf"></function_call></request></module></batch></ricci>
***Sending End
***Received from ricci server
<?xml version="1.0"?>
<ricci authenticated="false" success="5" version="1.0"/>

***Receive End
localhost password:
***Sending to ricci server:
<ricci function="authenticate" password="XXXXXX" version="1.0"/>
***Sending End
***Received from ricci server
<?xml version="1.0"?>
<Timeout_reached_without_valid_XML_request/>

***Receive End
Error: no ricci tag in ricci response

Of course I fudged the password with XXXs for the list.


Thanks!
________________________________________
Chip Burke







On 8/16/12 12:30 PM, "Chris Feist" <cfeist at redhat.com> wrote:

>On 08/15/12 16:49, Chip Burke wrote:
>> There is nothing in messages or secure on either node1 or 2 at that
>>time.
>
>Ok, there's something going on with the ricci authentication on that
>node.  Can 
>you give me the output of 'rpm -q ricci' as well as do a
>'/etc/init.d/ricci
>restart'.
>
>Then on the node that is running ricci, try this command:
>ccs -d -h localhost --getconf
>
>(it should ask your for a password, and enter the ricci password)
>
>Thanks,
>Chris
>
>> ________________________________________
>> Chip Burke
>>
>>
>>
>>
>>
>>
>>
>> On 8/15/12 4:50 PM, "Chris Feist" <cfeist at redhat.com> wrote:
>>
>>> On 08/15/12 15:08, Chip Burke wrote:
>>>> modcluster-0.16.2-18.el6.x86_64
>>>>
>>>> And
>>>>
>>>> [root at xanadunode1 ~]# ccs -d -h xanadunode2 --getconf
>>>> ***Sending to ricci server:
>>>> <ricci function="process_batch" async="false"
>>>> version="1.0"><batch><module
>>>> name="cluster"><request API_version="1.0"><function_call
>>>>
>>>> 
>>>>name="get_cluster.conf"></function_call></request></module></batch></ri
>>>>cc
>>>> i>
>>>> ***Sending End
>>>> ***Received from ricci server
>>>> <?xml version="1.0"?>
>>>> <ricci authenticated="false" success="5" version="1.0"/>
>>>>
>>>> ***Receive End
>>>> xanadunode2 password:
>>>> ***Sending to ricci server:
>>>> <ricci function="authenticate" password="XXXXXX" version="1.0"/>
>>>> ***Sending End
>>>> ***Received from ricci server
>>>> <?xml version="1.0"?>
>>>> <Timeout_reached_without_valid_XML_request/>
>>>>
>>>> ***Receive End
>>>> Error: no ricci tag in ricci response
>>>
>>> Thanks, can you also provide what was in /var/log/messages and
>>> /var/log/secure
>>> when those errors occurred?
>>>
>>>>
>>>> Thanks!
>>>>
>>>> ________________________________________
>>>> Chip Burke
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On 8/15/12 3:56 PM, "Chris Feist" <cfeist at redhat.com> wrote:
>>>>
>>>>> On 08/14/12 20:34, Chip Burke wrote:
>>>>>> [root at xanadunode1 ~]# ccs -h xanadunode2 --getconf
>>>>>>
>>>>>> This gives me similar results. It sits and spins for a few minutes
>>>>>>and
>>>>>> then fails with:
>>>>>>
>>>>>> Error: no ricci tag in ricci response
>>>>>>
>>>>>> ccs -f /etc/cluster/cluster.conf -h xanadunode2 --setconf
>>>>>
>>>>> Can you send me the output of 'rpm -q modcluster' and
>>>>> 'ccs -d -h xanadunode2 --getconf'
>>>>>
>>>>> Thanks,
>>>>> Chris
>>>>>
>>>>>>
>>>>>> This locks up everything going to GFS2 mounts. Two nodes recovered,
>>>>>> the
>>>>>> other didn't, required a fence_node. GFS2 showed this before the
>>>>>> fence.
>>>>>>
>>>>>> cd: /datastore/lvol0: Input/output error
>>>>>>
>>>>>> Along with the error Error: no ricci tag in ricci response
>>>>>>
>>>>>> ________________________________________
>>>>>> Chip Burke
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 8/14/12 3:12 PM, "Chris Feist" <cfeist at redhat.com> wrote:
>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Can you try using ccs to get the current configuration of that
>>>>>>>node:
>>>>>>> ccs -h <host name> --getconf
>>>>>>>
>>>>>>>
>>>>>>> As well as use ccs to try and set the conf on that node?
>>>>>>> ccs -f <cluster.conf file> -h <host name> --setconf
>>>>>>>
>>>>>>> This should let us narrow down whether it's an issue with ricci or
>>>>>>> luci.
>>>>>>>
>>>>>>> Thanks!
>>>>>>> Chris
>>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Linux-cluster mailing list
>>>>>> Linux-cluster at redhat.com
>>>>>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Linux-cluster mailing list
>>>>> Linux-cluster at redhat.com
>>>>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>>>
>>>>
>>>> --
>>>> Linux-cluster mailing list
>>>> Linux-cluster at redhat.com
>>>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>>>
>>>
>>>
>>> --
>>> Linux-cluster mailing list
>>> Linux-cluster at redhat.com
>>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>
>>
>> --
>> Linux-cluster mailing list
>> Linux-cluster at redhat.com
>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>
>
>
>--
>Linux-cluster mailing list
>Linux-cluster at redhat.com
>https://www.redhat.com/mailman/listinfo/linux-cluster