I am after some clarification. From what I read, based on corosync limitations around bonding and latency. Plus issues around multiple IP's per host. An simple network design for a three node cluster would be as follows. Note that PVECLSTR1 and PVEPROD1 are the same node. My understanding is that using one hostname for multiple IP's are problematic and corosync has problems with bonded networks in some scenarios. Keeping backup and production off the corosync network and switch will address potential latency issues.
Would this starting point eliminate most of the pitfalls of a basic cluster? If backups were interferring with production, they could be moved to their own network (or network throttling used).
I think my main question would be, will two hostnames on a node cause any issues? It seems to be the only solution to work around the common issues I have read, but I do not know what issues it will create.
Would this starting point eliminate most of the pitfalls of a basic cluster? If backups were interferring with production, they could be moved to their own network (or network throttling used).
I think my main question would be, will two hostnames on a node cause any issues? It seems to be the only solution to work around the common issues I have read, but I do not know what issues it will create.
HTML:
Switch Hostname IP # of Nic's Use (management / cluster)
Switch1 PVECLSTR1 192.168.0.1 1 Corosync / SSH mgmt/ Web interface
Switch1 PVECLSTR2 192.168.0.2 1 Corosync / SSH mgmt/ Web interface
Switch1 PVECLSTR3 192.168.0.3 1 Corosync / SSH mgmt/ Web interface
Switch Hostname IP # of Nic's Use (production network)
Switch2/3 PVEPROD1 192.168.1.1 1 VM's, backup etc
Switch2/3 PVEPROD2 192.168.1.2 1 VM's, backup etc
Switch2/3 PVEPROD3 192.168.1.3 1 VM's, backup etc
Last edited: