Ulrich,

Hi!

having a problem, I wonder what " corosync-cfgtool -s" does check actually:
I see on all nodes and all rings "status  = ring 0 active with no faults", but 
the nodes seem unable to comminicate somehow.

For UDPU/UDP without RRP it will always display "ring 0 active with no faults". With RRP it will will display error when one of the ring is failed.

For KNet it will display much better information:
# corosync-cfgtool -s
Printing link status.
Local node ID 5
LINK ID 0
        addr    = IP
        status:
                nodeid  5:      localhost
                nodeid  6:      disconnected


I there a kind of "corosync node ping" that actually checks that the local node 
can communicate with another given node using the corosync (TOTEM) protocol?

As Klaus wrote for non Knet setup corosync-quorumtool is probably closest.


The problem is that the cluster with three nodes says (h16 is the local node):
Node List:
   * Node h18: UNCLEAN (offline)
   * Node h19: UNCLEAN (offline)
   * Online: [h16 ]

In syslog I get a message like this every few (consensus?)  seconds:
Nov 20 07:46:59 h16 corosync[6240]:   [TOTEM ] A new membership 
(172.20.16.16:36752) was formed. Members
Nov 20 07:46:59 h16 corosync[6240]:   [CPG   ] downlist left_list: 0 received
Nov 20 07:46:59 h16 corosync[6240]:   [QUORUM] Members[1]: 116
Nov 20 07:46:59 h16 corosync[6240]:   [MAIN  ] Completed service 
synchronization, ready to provide service.
Nov 20 07:47:05 h16 corosync[6240]:   [TOTEM ] A new membership 
(172.20.16.16:36756) was formed. Members
Nov 20 07:47:05 h16 corosync[6240]:   [CPG   ] downlist left_list: 0 received
Nov 20 07:47:05 h16 corosync[6240]:   [QUORUM] Members[1]: 116
Nov 20 07:47:05 h16 corosync[6240]:   [MAIN  ] Completed service 
synchronization, ready to provide service.

It shows there is just one node with id 116. This is probably firewall problem.

Regards,
  Honza


The cluster is brand new and has absolutely no resources set up.
<configuration>
   <crm_config>
     <cluster_property_set id="cib-bootstrap-options">
       <nvpair id="cib-bootstrap-options-have-watchdog" name="have-watchdog" 
value="false"/>
       <nvpair id="cib-bootstrap-options-dc-version" name="dc-version" 
value="2.0.4+20200616.2deceaa3a-3.3.1-2.0.4+20200616.2deceaa3a"/>
       <nvpair id="cib-bootstrap-options-cluster-infrastructure" 
name="cluster-infrastructure" value="corosync"/>
       <nvpair id="cib-bootstrap-options-cluster-name" name="cluster-name" 
value="cluster1"/>
     </cluster_property_set>
   </crm_config>
...nodes...
   <resources/>
   <constraints/>
</configuration>

Regards,
Ulrich


_______________________________________________
Manage your subscription:
https://lists.clusterlabs.org/mailman/listinfo/users

ClusterLabs home: https://www.clusterlabs.org/


_______________________________________________
Manage your subscription:
https://lists.clusterlabs.org/mailman/listinfo/users

ClusterLabs home: https://www.clusterlabs.org/

Reply via email to