cancel
Showing results for 
Search instead for 
Did you mean: 

IP instead of computer name

Solution_Provid
Level 3

At the “Browse and select Virtual Machine” dialog box, some machines show the IP in the “VM Host Name” column instead of the computer name, which shows correctly on the “Display Name” column.  I was able to “fix” a couple of them. One of them was missing the reverse lookup entry on DNS, so after creating it and refreshing, the machine displayed properly. The other one had a second IP address defined on the NIC, which was the result of a previous test. After deleting the second address and refreshing the condition was also resolved.

However I still have a problem with some VMs that have actually a second NIC configured with a local address (192.168.1.10) and it seems like NetBackup discovers the IP in a certain order that results in some of these machines to show like this in NetBackup’s console.

I wouldn’t feel comfortable creating reverse lookup entries on the Domain DNS pointing to local addresses, and I can tell that this is only occurring in some and not all of the machines that have more than 1 IP address, so it may have to do with the order in which they were created? Either way, I was hoping that there may be some sort of a technique to get these names to match, but rather than experimenting, I wanted to see if anyone has a suggestion on how to deal with this situation.

6 REPLIES 6

MilesVScott
Level 6
Certified

Not sure if this is the issue, but you can prioritise the order that your NICs are accessed by DNS at least in Windows. Here are the steps to do so:

From the "Network Connection"s portion under Control Pannel Select the Advanced Dropdown and choose "Advanced Setting" In the first box highlight the corresponding NIC and using the arrows on the right changed its ordering.

As I said not %100 on this but it may be what your looking to for. Let me know if it works!

Solution_Provid
Level 3

That was a good idea Miles! Thank you very much for your response! I tried your suggestion in the Windows Virtual Machine, and moved the public NIC to the top, then refreshed NetBackup’s console multiple times, even closed the console and refresh angain, but the host name still shows as an IP instead of the computer name.  Since this machine is running FlashBackup-Windows, there is no client service to stop and start. It must be something else then…

V4
Level 6
Partner Accredited

are host file hardening is playing any role here ?

 

Solution_Provid
Level 3

Hosts file hardening? That is a very interesting thought! I checked the C:\Windows\system32\drivers\etc\hosts of the VM and the contents has Microsoft’s standard header and the only line not preceded with a # symbol is 127.0.0.1 localhost. Then I checked the same file on the vCenter server and it shows one extra line containing ::1 localhost

Apparently according to the OS they all seem to have the default hosts file and all name resolution is provided by the Windows Active Directory-Integrated DNS and this seems to work fine in other VMs with more than 1 NIC or IP address, and my assumption is mainly because there is a reverse lookup entry found for the secondary IPs discovered by NetBackup in vCenter.

So I’m not sure on what to do with the units that have internal local addresses like 192.168.1.1 that some VMs have as a secondary IP with no Gateway, DNS nor WINS entries defined, and some of them will have the same secondary local address, which is why I want to refrain from creating reverse lookup entries just for NetBackup to be able to display it in the console so that it can be added to the intended policy and then maybe remove the reverse lookup entry, unless is the last resort or a Symantec’s approved technique to get those types of systems added to the FlashBackup-Windows Policy.

Just for additional information, I am currently backing up some of them using the IP address and the backups succeed and all. It is just that it is complicated to document this phenomenon especially at the time to restore or cross referencing the existing computer names with the protected systems, especially in the situation where more than one machine may have the same secondary local IP.

Is there anything you recommend checking to see if the hosts file hardening is interfering in this scenario?

MilesVScott
Level 6
Certified

Out of curiosity, why is there a second NIC configured? If there a need for it or is that just a standard in your company? If there isnt a specific reason for a local IP you may want to consider disabling it. Dosent really get to the bottom of the issue, but it would solve the problem your having. Just food for thought!

Solution_Provid
Level 3

Thank you for your comment. Excellent question! And I agree 100% that the absolute preference is one IP and one NIC per Computer. However in our environment we have several exceptions, for example Microsoft cluster’s heartbeat connection uses it to learn about the partner’s state in a separate NIC often in a “Private” network, we also have SharePoint logic that depends on the IP address being called; also some MQ Connections ride in the same physical NIC yet they carry a separate queue in additional IP address, and also Microsoft Network Load Balance servers. Real-time and Off-band replication and they never cease to surprise me with a new flavor of configuration. But I’m totally with you, I’d prefer not to have multi-homed units on the network, and I also need to learn how to live with them as well.

As you mentioned, I was able to resolve a handful of them by removing the secondary IP as it was there either as a test that had been finished and no longer needed or the result of decommissions and/or upgrades so it was easy to contact the unit owner and quickly identify that is safe to remove the secondary IP address.

In some others, it was resolved because the secondary address seems to be used in the application’s logic and it is totally safe in those cases to create a Domain-Integrated reverse lookup entry if it doesn’t exist already, as it is intended to resolve to the same computer name anyway.

So the remaining group of units that have these internal local addresses that may be participating in an NLB or connect with devices like cameras or card readers that reside in these ranges, it may be not as easy to contact the unit owner and then come up with a plan for a change and so on.

Symantec’s documentation specifically mentions that this list gets populated by the reverse lookup DNS record. However it does not specifically talk about how to deal with those pesky secondary IP addresses in local private ranges.

Your comment is very valid because there may be a way to obtain the same results using alternate methods, and this will be the most desirable solution. Yet meanwhile the backup operators must either live with the IP address restoration points, or learn a work around or something to somehow overcome the challenge.