[ClusterLabs] Corosync unable to start after rpm generated but when we compiled from source code is working as expected

Jan Friesse jfriesse at redhat.com
Wed Oct 2 02:13:38 EDT 2019


> Hi Team,
> We have taken corosync source code from below git repository and compile and installed able to start corosync service as expected.
> https://github.com/corosync/corosync/tree/v1.4.8

Any reason to use 1.4.8? If you really want to stay on unsupported 
Corosync 1 branch, then I would recommend 1.4.10 (last released 1.x) -

> when we are try to create as RPM and installed on the node , after that we are unable to start the corosync service with below error.
> RPM installed version : corosync-1.4.8-1.x86_64
> While Starting the corosync server it got failed
> # corosync
> corosync [MAIN  ] Corosync Cluster Engine ('UNKNOWN'): started and ready to provide service.

Please compile released tarball. Git archive of tag version wasn't 
working as expected.

> corosync [MAIN  ] Corosync built-in features: nss
> corosync [MAIN  ] Corosync Executive couldn't open configuration object database component.

This means that corosync was not able to load objdb plugin. This is 
first plugin which is loaded so it means corosync plugins (files with 
lcrso extension) are ether not installed at all or corosync tries to 
find them on other place.


> corosync [MAIN  ] Corosync Cluster Engine exiting with status 12 at main.c:1811.
> Please check and let us know if I missing any object library required for the same.
> Thanks and Regards,
> S Sathish S
> _______________________________________________
> Manage your subscription:
> https://lists.clusterlabs.org/mailman/listinfo/users
> ClusterLabs home: https://www.clusterlabs.org/

More information about the Users mailing list