RESOLVED: on for a few hours then guess what

[jono659]

Why would the machines on the opposite network become suddenly not accessible. Tunnels all fine connection still up, somebody enlighten me for gods sake.

XP VPN can be on for days never a problem, why do there seem to be so many issues with this software.

Last edited by jono659 on Mar 24 04 7:29 am, edited 2 times in total.

[jono659]

No takers again then

[erwin]

Hi there

Im not sure of your understanding when it comes to netbios enumeration and how MS networks share the availabilty of network/computer resources etc, but there are several reasons why machines would show up as unavailable after once being available for connection etc.

Since netbios enumeration (update of netbios records) happens up to once every 10 minutes across the VPN, (incremental up to ten minutes once the network is well established), have you confirmed that these machines are still physically available on the remote LAN you are trying to join?
i.e are the machines still on line (have you tried retesting, pinging ) their availability or confirming their availability from someone at the remote network.

Have you tried to reach the machine via UNC IP address eg (\\192.168.4.1) as opposed to Computer name (which is really a Netbios record) from Run in the start menu.

Another point to consider is how is the browsing configured on the remote LAN, is there a master browser set to handle the identifying of network resources and resource lists on the LAN or are machines left to hold broadcast elections for their browser status. You can read all about Netbios and the way that Browsers handle resources on the MS website.

You just need to be aware that the WinGate VPN does not control how MS networking and Netbios handle resource displaying i.e showing shares/machines available etc.

For more of an overview on the role Netbios plays in WinGate VPN you can read about it on our website

http://support.qbik.com/index.php?_a=kn ... ubcat&_i=2
Under VPN

Regards
Erwin

[jono659]

they do in fact have elections, the advantage I have with my vpn is that I have 2 seperate dsl modems controlling 2 different subnets on two seperate static ip's, this being the case I have the server and the client both in my office even though they tunnell through the internet to connect. When the machines became unavailable, the client (2 machines in the network) was connected and fine the tunnells were active but the host and its network (6 machines) were unavailable. Looking at the host the local machines to it were all on line the tunnell to the client was up and active but the two machines were unavailable, this occured after about 7 hours. will take a closer look today. Thank you for your concise help.

JonO

[jono659]

KK,

Have disabled "computer browser" on all machines excpet the client and server. I think the problem may have been that on the server side there are 2 machines with wireless connections (one bridged with a normal nic card) I have a feeling that when the wireless card only machine won the election the problems would start. Have done the work will let you know what happens.

Regards

JonO

[jono659]

KK it stayed up 3 hours, any other ideas?

Hi there

Im not sure of your understanding when it comes to netbios enumeration and how MS networks share the availabilty of network/computer resources etc, but there are several reasons why machines would show up as unavailable after once being available for connection etc.

Since netbios enumeration (update of netbios records) happens up to once every 10 minutes across the VPN, (incremental up to ten minutes once the network is well established), have you confirmed that these machines are still physically available on the remote LAN you are trying to join?

Yes the computers on the remote side are all available

i.e are the machines still on line (have you tried retesting, pinging ) their availability or confirming their availability from someone at the remote network.

The remote network is in my office thru a seperate dsl line

Have you tried to reach the machine via UNC IP address eg (\\192.168.4.1) as opposed to Computer name (which is really a Netbios record) from Run in the start menu.

Not available via ping or name

Another point to consider is how is the browsing configured on the remote LAN, is there a master browser set to handle the identifying of network resources and resource lists on the LAN or are machines left to hold broadcast elections for their browser status. You can read all about Netbios and the way that Browsers handle resources on the MS website.

server and client are the only ones in the broadcast elections on their relative networks and are both showing as the master browser on their own side

You just need to be aware that the WinGate VPN does not control how MS networking and Netbios handle resource displaying i.e showing shares/machines available etc.

For more of an overview on the role Netbios plays in WinGate VPN you can read about it on our website

http://support.qbik.com/index.php?_a=kn ... ubcat&_i=2
Under VPN

Regards
Erwin

Thanks JonO

[jono659]

Couple of other things, first I had DHCP on in the vpn but have assigned each machine an IP manually, I have now disabled the DHCP.

In the knowledgebase article i noticed this

If all client machines behind the WinGate VPN server are going to be participating in the WinGate VPN, then they should all have the internal Ip of the WinGate server set as their default gateway.

Does this mean the server as at the moment i have them all pointed at the DSL Router


Thanks

JonO

[MattP]

Yes, you'll need to have them all going through the internal NIC on the server, from there the server should pass the connection through its external NIC to the router. It sounds like you only have one NIC on the WinGate server?

[Pascal]

You only need the default gateway pointed to the server if you are not using one of the other two methods. (RIP client OR static routes)

I believe you're using the RIP client, correct ?

[jono659]

Yes I am using the rip 2 client, also and here is a strange one, the machines on the opposite vpns were all not accessible when I left work, but on arriving home I connected the home machines to the server all machines on both networks are available to this machine and have stayed up all night! However the two original nets were still not available to each other back at work when I arrived this morning.

Does that help or confuse.

JonO

[jono659]

Found this in event viewer is this the problem?

"The master browser has received a server announcement from the computer XEON that believes that it is the master browser for the domain on transport NetBT_Tcpip_{17FAC9F5-7678-4E44-B5EE. The master browser is stopping or an election is being forced."

If so how can I rectify, knock out one of the two from the election?

JonO

Ok on reflection this cant be the case as if one of the client/server went down the local network would loose connectivity with it's local network. I am going to try setting Xeon to master in the registry to IsDomainMaster but leave Babybird to poll in elections in case Xeon becomes physically unavailable

[jono659]

Think I have isolated this problem, its seems to originate from one particular machine (Babybird) which is also the only participating client/server to suffer from the Application Exiting problem. the other 2 networks are solid and have been up now for 4 days without a problem so will have to find out what is conflicting in this machine. I thought that the problem discussed here http://forums.qbik.com/viewtopic.php?t= ... ab07e9a0c9 would have given support a clue, but no-one seems to want to touch it. I am sure if I can get to the bottom of this I will resolve my not accessible problem.

JonO

[jono659]

De /Re installed wingate VPN

Did XP System File Check


100% no application exiting and no unavailables.

Thank $%%^&^


JonO