Replies posted by JeroenTielen
Looks like HA Reservation is not turned on. When a node/host goes down (lcm or failure) then machines are migrated on best effort. When there is not enough memory available then machine are turn off. When the node comes back online those machines are turned on again.
From the best practices guis: https://portal.nutanix.com/page/documents/solutions/details?targetId=BP-2071-AHV-Networking:ahv-networking-best-practices-checklist.htmlI quote: Don't mix NIC models from different vendors in the same bond. Don't mix NICs of different speeds in the same bond. You should use leaf/spine network setup to mitigate switch failures. When doing switch maintenance do 1 switch at the time.
Hello @DenisF A single user VM with multiple TCP streams could use up to 20 Gbps of bandwidth in an AHV node with two 10 GB adapters. so no need to agg. 4 ports. It is recommended to enable LACP fallback on the switch used to connect the AHV nodes. Cisco Nexus: no lacp suspend-individual Cisco Catalyst: no port-channel standalone-disable In the AHV host CLI and on most switches, the default OVS LACP speed configuration is slow, or 30 seconds. This value—which is independent of the switch timer setting—determines how frequently the AHV host requests LACPDUs from the connected physical switch. The fast setting (1 second) requests LACPDUs from the connected physical switch every second, which helps you detect interface failures more quickly. Failure to receive three LACPDUs—in other words, after 3 seconds with the fast setting—shuts down the link in the bond. Nutanix recommends setting lacp-time to fast on the AHV host and physical switch to decrease link failure detection tim
Hi @Sanal, is the new node connected the same way as the other production nodes and are you using LCAP?If discovery is not working (ipv6 issues) or LACP is configured on the switch ports I always re-foundation the new node with the same versions as the cluster. During foundation make sure you set the correct ip address for AHV and CVM and make sure LACP is configured (as needed). Also make sure the automatic cluster creation is not selected. When foundation is ready you can expand the cluster manually by entering the CVM ip.
Yes with full nutanix and ahv you need two clusters. 1 in each dc. You can configure synchronious replication to keep this in sync and do the same as with you vmware metro availability. i’ve created a blog post about this: https://next.nutanix.com/community-blog-154/convert-your-cluster-to-ahv-and-configure-leap-for-disaster-recovery-40970
If you turn off HA and do LCM with not much memory available, then some guest VMs will shut down oog they can't be migrated to another node. Best practice is to turn it on. When there is too much memory used and ha cant be turned on, the cluster is undersized.
What I always do with wintertime change (so last sunday and not the summertime change).Disable the scheduled reboot on saterday or friday; On monday open golden image (do some magic like gpupdate or full virus scan); Save the golden image and deploy it to the vda's; enable reboo schedule. Had lots of problems in the past with the wintertime change on SBC environments. P.s. Why using pvs and not mcs?
If the virtual machine is stored on a storage container with Replication Factor 2 then there are 2 copies of the data stored. If the host goes down with a virtual machine with affinity rules for that specific host, then the virtual machine stays down. You have to remove the affinity rule to get the machine started again. Best practice is to set affinity rule to at least 2 hosts so the vm will migrate when lcm or maintenance is done.
Login to the community
Login with your account
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.