Re: (ASCEND) "MPP - no NULL msg timeout"

To: Kevin A Smith <kevin@ascend.com>
Subject: Re: (ASCEND) "MPP - no NULL msg timeout"
From: Phillip Vandry <vandry@mlink.net>
Date: Tue, 15 Sep 1998 15:10:28 -0400 (EDT)
cc: ascend-users@bungi.com
In-reply-to: Your message of "Tue, 15 Sep 1998 11:49:47 EDT." <Version.32.19980915114742.00e1fdd0@spanky.ascend.com>
Sender: owner-ascend-users@max.bungi.com

> In one way that's true, BUT LQM is passed all the way up the (PPP) stack
> where MP+/NULL is only passed to the MP+ engine. LQM packets are
> queued/dequeued like any other PPP frame, so have been known to
> clear "stuck" queues in the past - for want of a better explanation.

You mean like "clean the plumbing"? :-) Sounds like an odd thing to need.

In fact, I have found that LQM packets often get stuck in queues BEHIND
data packets on loaded links and thus get delayed, causing spurious
disconnects when the link is too loaded and they are delayed too long.
Of course they are supposed to be higher priority than data packets
so that they can still pass with ease and not be missed on a loaded
link, but this is not what I have found in practice.

Conslusion: sounds like MP+ has the better solution, since it does
the job of checking the L1&L2 connectivity.

-Phil
++ Ascend Users Mailing List ++
To unsubscribe:	send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd:	<http://www.nealis.net/ascend/faq>

Prev by Date: Re: (ASCEND) "MPP - no NULL msg timeout"
Next by Date: Re: (ASCEND) "MPP - no NULL msg timeout"
Prev by thread: Re: (ASCEND) "MPP - no NULL msg timeout"
Next by thread: (ASCEND) Multi-ip NAT problems
Index(es):
- Main
- Thread