-
Bug
-
Resolution: Unresolved
-
Normal
-
None
-
rhel-10.1
-
None
-
No
-
Moderate
-
rhel-ha
-
2
-
False
-
False
-
-
None
-
None
-
None
-
None
-
Unspecified
-
Unspecified
-
Unspecified
-
None
What were you trying to do that didn't work?
Set a long cluster name (either with 'pcs cluster setup' or 'pcs cluster rename'), force needs to be used, however the error message suggests that the name can be safely override if GFS2 is not used - which is this case.
This is because corosync can't parse line, that is too long.
Please provide the package NVR for which the bug is seen:
found in pcs-0.12.0-5.el10
How reproducible is this bug?:
always
Steps to reproduce
[root@hvirt-049 ~]# LONG_NAME=$(tr -dc A-Za-z0-9 < /dev/urandom | head -c 493) [root@hvirt-049 ~]# pcs cluster setup $LONG_NAME hvirt-049 hvirt-247 --force No addresses specified for host 'hvirt-049', using 'hvirt-049' No addresses specified for host 'hvirt-247', using 'hvirt-247' Warning: Chosen cluster name 'pK5wTKova6SdIhkiVOBeWcTMsG5kv4QVFORUybqNUToIIU39GqYj4SlAC9lfX8c592tb7LcPOuZL2oScyhIyZfvDi20gwcgaI3wACZxI8l83kIS7cGgQ1MfqhqUrl8gIhDFwBYGtCiw3ALLAT8z4xlmloNYafwJyRsic84cG04OJfgxeD1StwBXwVODLmnhyVK3eIFav5edqzg2xBWsAG39ZyvQwEoaSTx7EwMy2Aj8gHS2wg3G4WWmt7dtIirN268JMgJqOPjNdQ5phGLBaYSjAQmqWKgTlraHZudd8oYrers0oLM2Uo125nCv06hrS2zzPUlQ8Qyke80SxJjNFUKbexzbq8uPYKV5Xpsf0CpJxYsQE0aPicNJ8JNL5btwuP0tklRL999Fg1uOg1f5JzmFgsliRPrXFeTj1i2obTszCy8YmUrNO5oFaj8UANFuFHjpIfXWEMDG1sddrueEjBThC2T57DyEZotJ51Kkh6fQRXPUi1tak' will prevent mounting GFS2 volumes in the cluster, use at most 32 of a-z A-Z 0-9 _- characters; you may safely override this if you do not intend to use GFS2 Destroying cluster on hosts: 'hvirt-049', 'hvirt-247'... hvirt-247: Successfully destroyed cluster hvirt-049: Successfully destroyed cluster Requesting remove 'pcsd settings' from 'hvirt-049', 'hvirt-247' hvirt-049: successful removal of the file 'pcsd settings' hvirt-247: successful removal of the file 'pcsd settings' Sending 'corosync authkey', 'pacemaker authkey' to 'hvirt-049', 'hvirt-247' hvirt-049: successful distribution of the file 'corosync authkey' hvirt-049: successful distribution of the file 'pacemaker authkey' hvirt-247: successful distribution of the file 'corosync authkey' hvirt-247: successful distribution of the file 'pacemaker authkey' Sending 'corosync.conf' to 'hvirt-049', 'hvirt-247' hvirt-049: successful distribution of the file 'corosync.conf' hvirt-247: successful distribution of the file 'corosync.conf' Cluster has been successfully set up. [root@hvirt-049 ~]# pcs cluster start --all --wait hvirt-049: Error connecting to hvirt-049 - (HTTP error: 400) hvirt-247: Error connecting to hvirt-247 - (HTTP error: 400) Error: unable to start all nodes hvirt-049: Error connecting to hvirt-049 - (HTTP error: 400) hvirt-247: Error connecting to hvirt-247 - (HTTP error: 400) [root@hvirt-049 ~]# systemctl status corosync {...} Jun 16 11:06:51 hvirt-049.fast.eng.rdu2.dc.redhat.com corosync[678680]: parser error: /etc/corosync/corosync.conf:3: Line too long
Expected results
Validation of maximal length of the cluster name, that corosync can parse, is in place. This number seems to be 493 characters (together with the spaces and "cluster_name" text in the config, it is 512 characters for the line).
Actual results
Validation is not in place, cluster can't be started.