Cluster (1 PUB + 4 SUBs) running on VM with version 18.104.22.16800-2
VMware Installation: 2 vCPU Intel(R) Xeon(R) CPU E5-2643 0 @ 3.30GHz, disk 1: 80Gbytes, disk 2: 80Gbytes, 6144Mbytes RAM
When trying to perform a manual backup, in the "The following components are registered with Disaster Recovery System" pane, only PUB info appears. Nothing related with my other 4 SUBs shows up as depicted in the attached image.
I already checked connectivity, db replication, restarted DRF Master and Local services on PUB, and DRF Local in all SUBs. Still nothing.
I think it is supposed to backup all servers not just Publisher.
Pedro Miguel Marques
After some more digging I think the problem might be related to certificates between PUB and the SUBs.
Through CLI with the platform/drf/trace/drfLA0* on one of the SUBs I managed to pull this info.
So, it seems I have to dig a little further on how to solve this certificates issue.
2013-12-17 22:52:18,826 INFO [drfLocalRegMonitorThread] - drfLocalWorker.isLAConnected(): Unable to contact server. Master or Local Agent could be down...sleeping for 10 seconds
2013-12-17 22:52:20,994 DEBUG [main] - drfNetServerClient: Reconnect: Creating socket from Host: BNF51MTA1CS001 port: 4040
2013-12-17 22:52:20,994 DEBUG [main] - drfNetServerClient.Reconnect: detected isSecureMode is : true
2013-12-17 22:52:20,994 DEBUG [main] - drfNetSSLManager: createSSLSocketFactory: /usr/local/platform/.security/ipsec/trust-certs/ipsec-trust.keystore and/or /usr/local/platform/.security/ipsec/keys/ipsec-trust.passphrase does not exists
2013-12-17 22:52:20,994 DEBUG [main] - drfNetServerClient.Reconnect: Unable to connect to Host: BNF51MTA1CS001
2013-12-17 22:52:20,994 DEBUG [main] - drfNetServerClient:Reconnect, Unable to connect to host: [BNF51MTA1CS001], message: Unable to create SSLSocketFactory, cause: null
2013-12-17 22:52:20,994 DEBUG [main] - drfNetServerClient.closeSocket, closing connection to: [BNF51MTA1CS001]
2013-12-17 22:52:20,994 DEBUG [main] - drfNetServerClient.sleepRandom: sleeping for: 9 seconds
Thanks for all the help
It is confirmed. It was related to the certificates.
with help from this good doc
and tips from this forum I managed to perform a full backup.
Just to let you know that I didn't need to perform all the steps stated on the aforementioned doc.
I simply regenerated the ipsec-trust .pem file with the Publisher name on the Publisher and waited for its replication throughout the Subscribers.
And then restarted DRF Master and DRF Local on PUB, and DRF Local on all SUBs.
Pedro Miguel Marques