We are planning to implement MongoDB as a stretch cluster which includes Primary secondary in DC 1, Secondary secondary in DC 2, Arbiter in cloud(DC 3) where DC 1 and DC 2 are on premise servers and DC 3 in cloud
My questions are :
1) Should we need to keep same OS in AWS as our On-premise server or we can keep Amazon Linux also ?
In our case it is RHEL 7.2 in On premise servers
2) If No, will there be a performance impact if we use different OS between replica nodes ?
3) If yes, Do we need to do patch upgrades as frequent as we do in On Prem servers ?
4) How the failover works here in case of write concern as Majority ?
5) What are the other trade-offs in this approach if any ?
Asked by vignesh
(25 rep)
Oct 31, 2018, 05:14 PM
Last activity: Jun 9, 2021, 03:04 AM
Last activity: Jun 9, 2021, 03:04 AM