You will see the "2605-014 The table is unreadable" in the syslog or messages log file.

To resolve this problem, please execute the following steps::

(1) [on either node] As a precaution, save the current policy:

 sampolicy -s <filename>.xml

(2) [on the problem node] As another precaution, perform backup on directory /var/ct.

(3) [on either node] As root, execute the following command to prevent unintended reboot:

 chrsrc -c IBM.PeerNode CritRsrcProtMethod=5

(4) [on either node] As root, turn off the automation temporarily:

 samctrl -M T

(5) [on either node] As root, shutdown the domain by force, so that all the managed resources will be kept running:

 stoprpdomain -f <domain name>

(6) [on problematic node] As root, execute the following command to remove RSCT configuration:

 /usr/sbin/rsct/install/bin/recfgct -s

(7) [on all nodes] As root, execute the following command to prepare a node to be added to the domain:

 preprpnode <node list>

NOTE: The "preprpnode" command has to be executed on ALL nodes because the "recfgct" command has removed all RSCT configuration, including the public keys needed for communication among the nodes.

(8) [on one of the good nodes] As root, execute the following command to start the domain and copy the domain configuration information to the problematic node:

 startrpdomain -L <domain name>

(9) [on the same good node] Check the domain status until it goes online successfully:

 lsrpdomain

(10) [on either node] As root, re-enable TSAMP automation:

 samctrl -M F

(11) [on either node] As root, re-enable critical resource protection:

 chrsrc -c IBM.PeerNode CritRsrcProtMethod=3