-1

Our salesforces are using their laptop either on the LAN or on the road. They connect to the office with hamachi (gateway mode).

When outside, laptops could ping office servers using the virtual IP given by hamachi (10.40...) but not using their windows name. Windows names are resolved to local IP (192.168...).

There is no problem to ping and connect to printers and linux based NAS, so I think it's WINS related.

Laptops are rebooted before attempting to connect form outside. Does this flush wins cache ?

I think laptop sends a wins query that is answered by one of the computers on LAN which give the name table it has (with 192.167... IPs).

How to solve this issue ?

I've tried to flush wins cache using both netbt and ipconfig /flushdns, with no luck.

1 Answer 1

0

Why on earth are you using WINS in a modern environment (and one with Linux in it!), and Hamachi in a professional one? You are asking for disaster.

My advice to you would be to dispose of WINS post-haste (even before Hamachi). Do not rely on it. Use DNS, with authenticated dynamic updates.

To make your environment coherent, set your DNS suffixes appropriately in DHCP so that there isn't a distinction between a "windows name" (really you mean netbios name) and a DNS name.

Following this, do not multihome all your servers onto the VPN, if that's what you're doing (I'm not sure from your question). Instead, route your LAN at the VPN hosts' default gateway, and route your VPN at the LAN hosts' default gateway. Then, when hosts get the correct IP for a host from DNS (which they can cache unto eternity if they like), they will be able to communicate with it.

3
  • By WINS, I mean Windows 7 default name resolution, I must be wrong. Can you please tell me wich protocol is used by default to resolve names on Windows 7 ?
    – bokan
    Oct 10, 2013 at 8:03
  • Typically, it will use DNS preferentially, along its search path of default suffixes, and failing that, NBNS/WINS (which is a legacy protocol). If you are using NBNS at all, it is because DNS isn't working, basically. Check out what's happening using nslookup on one of the nonworking clients, and possibly traceroute too. Oct 10, 2013 at 8:42
  • And don't forget about LLMNR.
    – joeqwerty
    Oct 10, 2013 at 15:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .