I'm working on setting up a new server for sending email and I'm trying to figure out how to configure postfix to send email from 3 different domains. Ideally the nobody user on the server should be able to email using *@domain1.com, *@domain2.com, *@domain3.com just by changing the from field in the email that is being sent.
Incoming email is handled by google apps and port 25 (incoming) on the server is blocked.
I'm setting up a postfix mailserver and I've run acrossed "Address masquerading".
The description from the documentation says
Address masquerading is a method to hide hosts inside a domain behind their mail gateway, and to make it appear as if the mail comes from the gateway itself, instead of from individual machines.
What does this actually mean and what does the mail server do with masquerading?
I'm running into a bizarre issue with a router on my local university. There is one wireless router near the classroom where I have most of my classes that has a dhcp server running that gives out the self-assigned ip address that it's currently using for itself. (This router is broadcasting the same SSID as the rest of the campus routers) I use my laptop in class frequently since I do all my course work on it.
When my laptop is running Windows I just lose network connectivity for a minute or two whenever windows jumps to the signal from the bad router. Windows seems to realize something went horribly wrong and jumps back to the previous signal immediately. While annoying this is tolerable.
However when I'm running OSX the wireless driver stays locked on the signal from the bad router. I've been able to shake it from the signal after deleting the wireless network from my auto-join list, reconnecting and forcing a new ip address from the advanced network options. Sometimes this doesn't work and the only working solution is to find an ethernet jack and get a good ip address. When I finally get a good ip address from ANY source my wireless self-assigned ip issues disappear until the next time.
Asking the campus tech people to fix the router is pointless from my experiences with them. They still haven't fix many many other issues. Hacking into the router's configuration page is a bad idea because I might get expelled and I only have one semester to go before I graduate.
How can I prevent Windows and OSX from ever connecting to this router or at the very least get OSX to jump to a different signal when this happens?
I'm wondering what would be the best vpn solution for a linux server with windows, linux, and osx clients. I'm currently using openvpn but I find it difficult to set up and the windows client has given me more headaches than manually setting up a mailserver for the first time.
I have specific requirements for the clients. Some clients need to send their traffic through the vpn (laptops). Others only need to be able to be seen on the vpn but shouldn't be sending their traffic through the vpn (like the web server).
I would like to be able to use the Windows and OSX built in vpn clients if possible.
Are there other options available other than openvpn or am I doing something horribly wrong with openvpn?
I'm working on a website that could eventually have a very large database. The website will be an art community not until deviantART. I'll be running a lot of queries for analyzing how the site is used. I expect some of the queries to kill the database because I don't know how to optimize SQL queries (yet). I'm starting out with a small linux VPS and would move to physical servers in the future.
Which database would scale the best given this situation?