Number: 1472

Date: 19-Jun-84 17':29':15

Submitter: Sannella.PA

Source: dmrussell.pa

Subject: ExternalReceiverAbort message from printer should cause break

Assigned To: 

Attn: vanMelle

Status: Open

In/By: 

Problem Type: Bug

Impact: Moderate

Difficulty: Moderate

Frequency: Everytime

Priority: Perhaps

System: Communications

Subsystem: PUP Protocols

Machine: 

Disk: 

Lisp Version: 

Source Files: 

Microcode Version: 

Memory Size: 

File Server: 

Server Software Version: 

Disposition: 

Description: '
Sender': Sannella.PA'
Date':  5 Jun 84 16':16':30 PDT (Tuesday)'
Subject': "receiver timed out" messages from Spruce printers'
To': vanMelle.pa'
cc': LispSupport.pa'
From': LispSupport.pa'
Reply-To': LispSupport.pa'
'
Is this the same as the old problem with file servers?'
'
----------------------------------------------------------------'
Date': Tue, 5 Jun 84 08':33 EDT'
From': dmrussell.pa'
Subject': Printing in a foreign land'
To': LispSupport'
cc': dmrussell'
'
Much to my chagrin, I''ve discovered that trying to print files from Interlisp when you don''t live in the PARC environment is a hard job. '
...'
'
b) I keep getting "receiver timed out" messages from ordinary Spruce printers; at least from Spruce printers that are one hop away.  However, Interlisp lies to me (or at least misleads me) when it says "{CASK}<DMR>SHEM;103 => {LPT}"  It may be sent, but it never gets there.  Misleading msg, methinks.  Besides that, why does anyone call it {LPT}?  Talk about archaic holdovers.  Grump, grump.  (Or maybe it stands for "Laser Printer."  Right.)'
'
----------------------------------------------------------------'
Date': Tue, 5 Jun 84 09':13 EDT'
From': dmrussell.pa'
Subject': More on Printing'
To': LispSupport'
cc': dmrussell'
'
Just to confuse things, I got the "Receiver timed out message", but still got my hardcopy.'
'
-- DMR -- '
'
----------------------------------------------------------------'
'
Date':  8 Jun 84 22':13 PDT'
From': vanMelle.pa'
Subject': Re': "receiver timed out" messages from Spruce printers'
In-reply-to': LispSupport.pa''s message of 5 Jun 84 16':16':30 PDT (Tuesday)'
To': LispSupport.pa'
cc': vanMelle.pa'
'
What''s the "old problem with file servers"?'
'
He doesn''t say where he gets the "receiver timed out" messages from.  Might be related to the problem he described talking to Wabash servers, on which I interacted with him some (final answer not yet known).  In the course of that I did observe that there definitely IS a bug or misfeature in the Empress code such that if the printing host sends back an "ExternalReceiverAbort", Empress prints the error message accompanying the abort and simply returns NIL.  This leads the guy who copied file to {LPT} to lie about its success.  ExternalReceiverAbort conventionally means there is something wrong with what you''re trying to send such that trying again later won''t help.  I''ve never seen it around here in real life.  A break rather than quietly returning NIL seems to be a more appropriate response.'
'
	Bill'
'
-----'
'
Sender': Sannella.PA'
Date': 11 Jun 84 14':54':39 PDT (Monday)'
Subject': Re': "receiver timed out" messages from Spruce printers'
In-reply-to': vanMelle''s message of 8 Jun 84 22':13 PDT'
To': vanMelle'
cc': LispSupport'
From': LispSupport.pa'
Reply-To': LispSupport.pa'
'
What I meant by "the old problem with file servers" was AR 721 (Writing to NS file server gives many "Filesever':filing#nn not responding" msgs).  I had guessed that the problem was that it was printing "receiver timed out" messages inappropriately.'
'
Do you think that the cause of his problem is the "ExternalReceiverAbort" bug you described?'
'
-----'
'
Date': 11 Jun 84 15':15 PDT'
From': vanMelle.pa'
Subject': Re': "receiver timed out" messages from Spruce printers'
In-reply-to': LispSupport.pa''s message of 11 Jun 84 14':54':39 PDT (Monday)'
To': LispSupport.pa'
cc': vanMelle.pa'
'
The problem is completely unrelated to ar 721, since Spruce servers are not NS servers, and have nothing in common with them.'
'
The "ExternalReceiverAbort" bug I described explains why it would be possible that ''Interlisp lies to me ... when it says "{CASK}<DMR>SHEM;103 => {LPT}"''.  However, it doesn''t explain at all why the server is sending the abort in the first place.'
'
	Bill'


Workaround: 

Test Case: 

Edit-By: Masinter

Edit-Date: 17-Aug-84 14':02':25