I was updating a Sub's name in ccmadmin -> System -> Server and mistyped the name. I quickly corrected it, but it was too late :-(
Doing a "run sql select * from processnode" on the pub shows the correct value, but the sub is showing the wrong value. Because of this, DB Replication is borked.
This is just my dev cluster, so I'm planning on trashing the sub and rebuilding it so that it gets a fresh copy of the DB from the pub. But before I do that, any suggestions as to how to correct this incorrect value?
- a "run sql update processnode ..." doesn't work.
- I've tried "utils dbreplication stop" on all nodes then "utils dbreplication reset all" on the pub. No joy :-(
I'd seen that support forums document. I'd tried the dropadmindb, but it didn't seem to work. The only thing that did work was rebuilding the Sub. Not my favoured approach, but as it was just my dev system, it was the easiest way of getting the cluster working again.
I just ran into a similar issue in my lab environment, I was updating the publisher IP address and moving it around when my ProcessNode table got messed up on a subscriber server.
DB Replication was completely busted on the subscriber (CLI output from the sub below):
admin:utils dbreplication runtimestate
Runtime state cannot be performed on a cluster with a single active node; aborting operation
admin:run sql select name,nodeid from ProcessNode name nodeid ================== ====== EnterpriseWideData 1 192.168.0.21 2 192.168.0.20 3
The real publisher IP was 192.168.0.19 not 192.168.0.21.
I can confirm (at least for version 9.1.2) that the ProcessNode table was updated on the subscriber when I performed the following process (using this doc):
1. utils dbreplication stop all (Only on the publisher) 2. utils dbreplication dropadmindb (First on all the subscribers one by one then the publisher) 3. utils dbreplication reset all ( Only on the publisher )
After about 20 minutes replication was back up and the ProcessNode table was correct.
admin:run sql select name,nodeid from ProcessNode name nodeid ================== ====== EnterpriseWideData 1 192.168.0.19 2 192.168.0.20 3
Are you getting this error “Installer User Interface Mode Not Supported. The installer cannot run in this UI mode. To specify the interface mode, use the -i command-line option, followed by the UI mode identifier. The value UI mode identifiers...
The below trick might come handy when you have to add a new node to a cluster but you don't have or is unsure of the security password for the publisher. This procedure has been around for ages.
1) Login into the CLI of the Publisher.