0

I'm having a situation with Hyper-V that I really need some help with. We recently added a 4th physical host to our Hyper-V cluster. It was working fine until this last weekend when some things got rebooted, now one of the VLANs isn't reachable. The main culprit is a SQL server that is clustered, and running on VLAN 3. When I ping the IP address within VLAN 3, I can reach it from any of the others. Let's call it IP 1. When I ping from the AD machine inside VLAN 3, it's reachable. When I connect to the SQL Server on IP 1 from one of the two cluster members, I can connect to the database.

When I go to one of the servers in VLAN 2 that needs to talk to it and try to connect, I can't. I moved one of the IIS Servers that's on the same host from VLAN 1 into VLAN 3 and re-iped it, and when I did it was able to connect and return data from the SQL Cluster, I then moved it into VLAN 2 and re-iped again it would not connect. I've tried moving that same IIS Server onto one of the previous hosts, and am having resource issues where the VM won't start.

I'd love to hear some suggestions on other things I can test at this point. I don't have physical access to the hosts, but I have contact with those who do. We've checked a few different things and just can not figure out what might be happening or the next steps. I don't have SCVMM running, and am hesitant to activate it and potentially cause a production issue as I understand I would have to add the whole cluster at one time. I've thought about rebuilding the SQL clusters in VLAN 3, but they validate, and show that things should be working, and when moving into that VLAN everything works.

I've also verified the DNS and Subnet masking and both are accurate and should work. Additionally, other machines in VLAN 3 are reachable, such as the AD server that is there. Some production machines are in the same VLAN 3 as well and working fine.

Any thoughts?

0

Most likely the DB server's route to VLAN 2 subnet is incorrect. You need to study the DB server routing table to see how it routes packets to VLAN 2 and check if it matches the network layout.

|improve this answer|||||
  • All of the routing points out to the gateway, and I see that the ip address matches on a /32. The interesting thing that this led me to, is that the primary database in the cluster's IP can't be found through using arp. In fact, all the database IP including the main cluster name point at the IP from arp of the main database. This doesn't sound like it would be correct? – George Stafford Dec 21 '17 at 22:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.