11-05-2014 04:12 PM
Hi All,
I searched around for this and couldnt find anything that led me to resolution. I have 6 VMs that have a 'client name' in the policy view window (not 'new client' selection window) and in the activity monitor window, that is NOT the same as their hostname in the new client selection window (like when browsing the VMhosts for guests to add).
I think this could have something to do with reverse DNS, I have had some issues with that before, but none of the hostnames listed that are erroring out - are present on the reverse records of either DNS server.
I also have 2 that are listed as vmhostname 169.254.x.x addresses, and I have no idea why on that either. I can't figure out exactly which VM's are causing this issue becuase they dont have the same name listed when I go to the VM selection window.
I have had off and on issues like this and I want to get it all squashed and working properly. It looks like backing up via vmware display name could possibly get around this (we are very good about display names matching hostname) but I am not sure of what issues this could present.
Solved! Go to Solution.
11-07-2014 09:15 AM
Its pulling that information from vCenter, so I would guess the vmx file must have that information in it. Maybe the SQL server has 2 NIC's and the one has an IP and the other not. The order of binding might be incorrect so the 169 is listed instead of the correct one. Look at vcenter and the configs of the vm's.
NBU does not generate the info.
11-05-2014 08:02 PM
Hi,
Check the info below for an explanation of the 2 methods. I suspect your vm has more than one IP listed which is resulting in the inconsistencies.
http://www.symantec.com/docs/TECH147535
This is a brief description of how NBU validates VM's for backing up using VStorage backups.
11-06-2014 02:02 PM
Hi Raain,
Its possible they had a 169 address for a while but right now the servers in question only have a single address. I renamed the fi_cntl folder to fi_cntl_old and it recreated it but still keeps the old names on there. Do you know how I can sort of 'force a recheck' or something where it will re-check the machines names and update them? I have tried removing them from the policy and re-adding, restarting services, etc. with no luck.
Do you know if there are any gotchas or things to look out for when switching to backing up via display name after backing up via hostname for so long?
Thank you
Nate
11-06-2014 08:50 PM
Hi,
If you removed the content from (or renamed) the folder it would generate a fresh copy by talking to vcenter. So then it means its getting that info from vCenter
Or, you've removed the folder from the wrong backup host (not the same as listed in the policy), and you're looking at the old data.
11-07-2014 08:47 AM
Interesting. I did rename the folder (i only have one backup host, the master server) and it was recreated, but I have 2 SQL servers that still seem to want to use a 169.x as VM hostname, but they do not have more than 1 IP address listed. I appreciate the comments!
11-07-2014 09:15 AM
Its pulling that information from vCenter, so I would guess the vmx file must have that information in it. Maybe the SQL server has 2 NIC's and the one has an IP and the other not. The order of binding might be incorrect so the 169 is listed instead of the correct one. Look at vcenter and the configs of the vm's.
NBU does not generate the info.