12-09-2015 04:34 PM
I do have three node WSFC window cluster. All the nodes are virtual machine. Each node has one SQL server 2012 instance installed and Availability group has been configured. Top of this AG Listener has been created.
Customer is facing time out issue randomly. We checked and found that when we do NSLOOKUP it doesn’t give consistent result. Sometime listing two IPs which is correct but some time list only one IP which is obviously not correct. Even Cluster network name list only one IP which was supposed to list two IPs.
My question here is, why it happens. Do we need to do any specific configuration/entry in Infoblox grid or anything else need to take care? I am a core database administrator so doesn't have much idea on DNS part. Can anybody help here?
12-30-2015 05:18 AM
From my understanding, you have configured two IPs for a Devices in Infoblox. When you do a Nslookup for the FQDN from the devices or any PCs you are getting both the IPs and some times you are getting only one.
We need to confirm whether the Infoblox DNS server is giving out the right information to the device i.e. both the IPs to the device or not. When you face this issue, do the following testing which is not going make any downtime on your server.
1. Try to resolve the FQDN from the CLI of the Authoritative DNS server of that domain.
2. If you get the correct answer, please take a traffic capture from the devcie while doing a nslookup and check which server is giving out the information.
3. Also, try poining out the query towards the Authoritative DNS server from your PC which is having this issue and see whats happening.