Re: Create DR for two cluster failover with Arbitrator but showed "Download Pending" status

jianli · ‎02-10-2020

Two individual vManage cluster setup on two DC, using 19.2.1, with arbitrator to setup DR.

We add all of the controller with vpn 512 ip, when create DR, we input all 6 vBond VPN 512 IPs, we can complete DR process, but after that DR status showed " DownloadPending".

I checked on arbitrator it showed two cluster are down.

Please see below logs of arbitrator:

192.168.1.x, 192.168.2.x are cluster ip, can't reach to arbitrator, arbitrator can only reach to two cluster vpn 0 dedicate IP: 172.16.155.xx/27

abtt:~$ tail -f /var/log/nms/vmanage-server.log
31-Jan-2020 15:59:11,343 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Majority ConsulMembers are DOWN. Can't perform switchover
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Data Center personality: arbitrator, member IP: 172.31.155.52, status:UP
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Data Center personality: primary, member IP: 192.168.1.3, status:DOWN
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Data Center personality: secondary, member IP: 192.168.2.2, status:DOWN
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Data Center personality: primary, member IP: 192.168.1.1, status:DOWN
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Data Center personality: secondary, member IP: 192.168.2.3, status:DOWN
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Data Center personality: primary, member IP: 192.168.1.2, status:DOWN
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || Data Center personality: secondary, member IP: 192.168.2.1, status:DOWN
31-Jan-2020 16:04:11,345 ICT INFO [abtt] [DataCenterHealthManager] (vManage-akka.actor.default-dispatcher-4) || All nodes of Primary are DOWN!, checking to see if majority ConsulMembers

Does anybody have any idea about this?

Thanks!

RohitRaj03827 · ‎02-13-2020

Hi jianli,

-> Cluster nodes must be part of same DC due to low latency requirements. Single cluster does not fulfill DR requirements, need for automatic fail over.

-> primary nodes, secondary nodes cluster interface should be under VPN 0 also an Arbitrator interfaces which connects to Active node and secondary nodes that is also under VPN 0.

Regards,

Rohit Raj

Regards,
Rohit Raj