Hello, currently there is very handy post-login scripting already implemented but it naturally executes only when successful login happens. I would like to have another two script callbacks which runs: 1. immediately after client connects to server - dovecot should run script with connection IP and wait for result and then reject/accept connection with return message accordingly. 2. after failed login - dovecot should exec script with connection IP, login params without waiting for result The main reason behind this idea is to use some logic, database (redis...) or indicators like GeoIP, p0f as bruteforce prevention. It might be also usable for IP based ACLs. AFAIK the only way to create something similar is to proxy whole authentication to some external process, which is not desirable when dovecot manages user database. SH
On 27.09.2018 10:45, Stanislav Humpl?k wrote:> Hello, > > currently there is very handy post-login scripting already implemented > but it naturally executes only when successful login happens. I would > like to have another two script callbacks which runs: > > 1. immediately after client connects to server - dovecot should run > script with connection IP and wait for result and then reject/accept > connection with return message accordingly. > 2. after failed login - dovecot should exec script with connection IP, > login params without waiting for result >https://wiki.dovecot.org/Authentication/Policy Aki
Possibly Parallel Threads
- SV: Looking for a guide to collect all e-mail from the ISP mail server
- Another Fedora decision
- Is it possible to execute Objective-C code via LLVM JIT?
- [OpenCL] Question about pre-linking passes required to build OpenCL program
- Is it possible to execute Objective-C code via LLVM JIT?