because the iSCSI failover thread is already very long I want to start a new one ...
I'm testing iSCSI autofailover with two DSS demo servers and I think it's a really good solution ! I tested it with SQL DB, streaming, copying files ... while doing the failover without any interruption (maybe video stucks for one second ...), great! Then I "killed" the primary (power off), powered on again, sync, failback, also without troubles. Next test is Hyper-V VHD on the iSCSI disk ...
But: I can't find any "best practice" whitepaper, e.g. how to do cluster-safe patching, upgrading, NIC layout ...
- What is the right patching practice? First patch the secondary, then primary ...? Are all coming patches autofailover-safe (running the cluster with different versions)?
- How to configure 2 DSS with e.g. 6 GBit-NICs (example: 2 virtual IPs for iSCSI MPIO, 2 bonded for quick replication, 1 management, 1 NAS ...)?
- Can I add more volumes to running cluster?
Best Practice for DSS Autofailover
! I tested it with SQL DB, streaming, copying files ... while doing the failover without any interruption (maybe video stucks for one second ...), great! Then I "killed" the primary (power off), powered on again, sync, failback, also without troubles. Next test is Hyper-V VHD on the iSCSI disk ...
Reply With Quote