[Help] Question on Erroneus Mob View & Connection

1.>

The Phantom Mob in Port Lux appears as a Hound in my client.

any idea as to what caused it? client file? server file?

2.>

i can login using my global ip (internal & serverlist txt files) when i am in
the same pc as the server. but if i use the same client on a networked (or other external) pc, i can login but cannot join the channel.

any ideas? its weird considering that vmware is separate from main os, regardless of where i am connecting from. i wonder what sequence of events happen during clicking of the channel, because if i knew, then id be able to figure out the answer why 2 identical clients but positioned on a different pc
produce different results.

1. I get this too with the chinese client. Mummies are broken too as are skeleton warriors. The client files look ok as i have compared against an unmodded client and personally i think it is a mob id mismatch in the hexed cabal.exe or cabal.exe we are using. We don't know how to make our own and i can't work it out as i can't find unmodified versions of the exes so i can compare so this is not likely to change anytime soon.

2. Not weird at all really, in fact i would expect it. You can't use a WAN IP (global) behind a router or behind NAT. The machine the server runs on probably isn't which is why it works. You can only use LAN ips behind NAT.

that was enlightening, thank you very much.

so global connection will never be achieved if nat + local centos ips are used. (which is what i used - you hit me bullseye ;) )

one final barrage of questions, (hopefully this will really help a lot of forumers)

what do you suggest i use to allow global access?

bridge or host only.

i am using a single static ip to connect to the net.

must i use it for (global - main ip) centos then make (local ip) windows my network? or its not necessary)

can i set bridge ip as internal (but fake (made up ip since i only have a single ip)) ? (unless if i use lan ip then i have upto 255 ^^)

does ip when networking need to be in increments? (will it conflict if ip in physical and virtual adapters are the same)

did you manually configure ips (etho file), or set it before installation (centos option)
or use the "setup" command via terminal

how does the "vmware bridge protocol" switch work, must it be checked? (what happens if it is unchecked)

can single static ip be used for bridge ? (or only dynamic)

can net access be achieved using host only? (or only a lan file connection) (can this be used to allow global access?)

*** don't get me wrong, i read a lot, but a lot of those documents didn't make sense, and so i figured that simple words from a good teacher or someone with experience would make the big difference. ***

thank you in advance for whatever response you may give this matter ^^

The short answer here is that i really don't know. I have not yet tried external access as there are still more important problems to tackle like the login bug and the fact the client is a complete mess and i'm spending my time there at the moment. The setup of the server end i much like a lot of other server code i have run so that isn't the problem, what is a problem is that we have vmware in the middle of this and that complicates things in terms of external access.

From a server point of view external clients can't "see" lan ips and local lan players can't "see" WAN ips behind NAT so configuring the server ip so the server config ip entries have to be set to values that the right ip is always handed back to whoever is connecting. The client (player) is going to need to get either the wan ip or the lan ip back from the server depending on where they connect from and the server should always get either the lan ip or localhost (127.0.0.1). Figuring out which .ini entries get handed back at each stage of login/channel select/move to world takes a bit of experimentation unless you have a manual and various people are still at that stage.

The vmware bridge option pretty much sets up a new "fake" lan and bridges your actual computer to the virtual one as a direct mapping. Without bridging you have NAT which is largely a big fudge to get around the fact that only one ip address can capture a port for sending or listening at once and you need to translate several connections to one ip to multiple ips on different ports based on criteria. The second one has more ways it can go wrong which is why bridging is a better option.

I'm planning on doing some external testing this weekend so i should have more of an idea on exactly how to configure it for that afterwards which will be a much better time to write a guide than now i think. I will be updating my database guide too as people are still struggling with bad db dumps :/