> 4 !root telnets to a unit at one time.
Why is that? It seems a completely arbitrary number, and we often
have to perform contortions to get into a PM whose administrative
telnet limit has been reached. Can you at least put an idle timeout
on the !root telnet session? And maybe even implement TCP keepalives
for those sessions? Right now it's a management nightmare with no
obvious justification.
And speaking of arbitrary numbers, why does the RADIUS-settable
Idle-Limit only range between 2 minutes and 4 hours? Four hours looks
suspiciously like (60*4 = 240, which is close to the max value of a
byte), but the 2 minutes lower limit makes no sense. Specifically,
I'd like to be able to set an Idle-Limit of 0 for users that shouldn't
idle out. Both Ascend Maxen and USR NETservers can do this, and
frankly it's the only reason I'd ever set any idle timeout other than
the port's default, which is also broken in 3.3 and up :-(
-- Robert