book
Article ID: 311756
calendar_today
Updated On:
Issue/Introduction
Symptoms:
While attempting to configure/reset the vRA Postgres Database you get an error similar to:
"Could not configure replica for node: <cafe.node.760065698.31327> on address: <primary fqdn>, reason is: I/O error on GET request for \"https://<primary fqdn>:5480/config/cluster/replica?noLock=false&reset=true&baseBackup=true&masterNodeHost=<secondary fqdn>\": null; nested exception is org.apache.http.client.ClientProtocolException"
postgresql.csv error in pg_hba.conf file
'2017-11-18 19:39:30.865 UTC,,,23971,,5a108c72.5da3,2,,2017-11-18 19:39:30 UTC,,0,LOG,F0000,"authentication option not in name=value format: e.",,,,,"
line 105 of configuration file ""/var/vmware/vpostgres/current/pgdata/pg_hba.conf""",,,,""'
shell-ui-app service not running -> UNREGISTERED
Environment
VMware vRealize Automation 6.2
VMware vRealize Automation 7.x
VMware vRealize Automation 7.1.x
VMware vRealize Automation 6.x
VMware vRealize Automation 7.3.x
VMware vRealize Operations Management Pack for vRealize Automation 2.0.x
VMware vRealize Automation Desktop 6.2.x
VMware vRealize Automation Desktop 6.2
VMware vRealize Automation 7.2.x
VMware vRealize Automation 6.2.x
VMware vRealize Automation 7.0.x
Cause
Error may occur when one of the nodes becomes disconnected for a long period of time. Error may occur when both nodes are in replica or primary node at the same time.
Resolution
1. Take snap shots of vRA VA appliances and postgres database backup of primary node.
2. Following steps are to be performed only on replica/secondary database server
a. Stop postgres service: sevice vpostgres stop
b. Synchronize replica database with primary database:
vcac-config -e cluster-config-replica --master-node-host <fqdn of master postgres> --reset
Additional Information
Impact/Risks:
possible loss of data. Be sure to take a snapshot and a backup of the postgres database before attempting this procedure.