Definitely sounds that network failure had something to do with it. Anyway, since we cant take time back (as they say), lets see what we can do to fix current problem.
Few questions:
1) Does it happen to every user?
2) Is everyone using same vpn client version?
3) Is there any user at all that can connect?
4) If yes to question 3, then do they get correct OM popl IP?
5) Is this a cluster? If yes, see if you can do a quick failover and test
6) Can you run -> fw tab -t om_assigned_ips -s and provide the output?
Cheers,
Andy