HP3000-L Archives

July 2001, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Bruce Toback <[log in to unmask]>
Reply To:
Bruce Toback <[log in to unmask]>
Date:
Tue, 10 Jul 2001 15:37:44 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (69 lines)
Tracy Johnson writes:

>Sometime afterwards, the HP calls with another
>solution to the dilemma.  A Beta Patch that changes
>the way UDP behaves.
>
>Seems UDP is single-threaded on the HPe3000.

Hmm... this doesn't make any sense. UDP is a connectionless protocol,
meaning that unlike TCP, the lower levels of the networking stack don't
know or care about responses to UDP packets. So there's nothing to
"thread" at the network level.

>The
>Spoolers are SNMP, which by some process that I
>can only call magic, works under UDP.

Network communications use a hierarchy of protocols, each of which
specifies some part of the addressing and routing for the message. A
process that wants to query a node via SNMP creates a UDP packet with the
SNMP message in it, then sends the packet out. Incidentally, the spoolers
use SNMP only for printer status, not for printing.

>The beta patch that HP is providing, will either
>do one of two things:
>
>1)  UDP will become multi-threaded, like TCP and
>start spawning separate processes or,
>
>2)  UDP will become tweaked so that the time-out
>for it will not be so long.

This has to refer to the spooler process specifically, not to UDP in
general.

>(Note there is nothing in NMMGR that affects how
>long UDP waits for a time out.  So there is nothing
>that can be self-corrected there.)

Again, that's because there's no such thing as a timeout for a UDP
packet. Each packet is complete in and of itself, so once the packet is
sent, there's no reply to wait for. (The process that sent it might be
waiting for a reply, but that's not the business of the network code.)

>It is assumed something on the network has changed
>for this gateway, but it is not yet determined
>what it is.  The circuit and router is maintained
>by a vendor who denies any changes.

Are they suddenly blocking SNMP? That wouldn't surprise me; SNMP
generally has no business going across the Internet except to networks
known in advance.

-- Bruce


--------------------------------------------------------------------------
Bruce Toback    Tel: (602) 996-8601| My candle burns at both ends;
OPT, Inc.            (800) 858-4507| It will not last the night;
11801 N. Tatum Blvd. Ste. 142      | But ah, my foes, and oh, my friends -
Phoenix AZ 85028                   | It gives a lovely light.
btoback AT optc.com                |     -- Edna St. Vincent Millay
Mail sent to [log in to unmask] will be inspected for a
fee of US$250. Mailing to said address constitutes agreement to
pay, including collection costs.

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2