1. Support
  2. Knowledge base
  3. 100018774

CVM node cannot join cluster after upgrading.

Article: 100018774
Last Published: 2022-01-19
Ratings: 0 0
Product(s): InfoScale & Storage Foundation

Problem

CVM node cannot join cluster after upgrading.

 

Error Message

VxVM vxio V-5-3-0 get_nid_master inconsistent protocol versions. Self proto version is 70, while incoming proto version is 60. Reconfig will fail.

Solution

Cluster protocol version is different between the CVM master (running version 60) and the joining node (running version 70). To check the cluster protocol version, run the following command:

# cat /etc/vx/volboot

volboot 3.1 0.1 60
hostid host1
end
###############################################################
###############################################################
###############################################################
###############################################################
###############################################################
###############################################################
###############################################################
###########################
#

Cluster protocol version is last entry in the first line. If it's different between the CVM nodes, the cluster protocol needs to be updated to the latest version.
 
To list the protocol versions supported by the CVM master, run:

master# vxdctl support
Support information:
 vxconfigd_vrsn:  23
 dg_minimum:      10
 dg_maximum:      140
 kernel:          17
 protocol_minimum:60
 protocol_maximum:70
 protocol_current:60

There are two ways to upgrade the protocol version depending on the version running on the CVM master

1. If the CVM master is running lower cluster protocol version, run the following command to upgrade all the nodes in the cluster:

master# vxdctl upgrade

2. If the CVM master is running higher cluster protocol version and the joining node needs to be updated, use the following procedure:

slave# vxdctl stop
slave# mv /etc/vx/volboot /etc/vx/volboot.<date>.save
slave# vxconfigd -m disable -x syslog
slave# vxdctl init
slave# vxdctl enable

Check if the cluster protocol version has been updated in /etc/vx/volboot file.

slave# grep volboot /etc/vx/volboot
volboot 3.1 0.1 70

Now the CVM node should be able to join the cluster.
 

 

Was this content helpful?

Article Languages