GNU bug report logs - #65602
30.0.50; eshell-test/{elisp,subcommand}-reset-in-pipeline fails intermittently on macOS

Previous Next

Full log

View this message in rfc822 format

From: Jim Porter <jporterbugs <at> gmail.com>
To: Stefan Kangas <stefankangas <at> gmail.com>
Cc: 65602 <at> debbugs.gnu.org
Subject: bug#65602: 30.0.50; eshell-test/{elisp,subcommand}-reset-in-pipeline fails intermittently on macOS
Date: Thu, 14 Sep 2023 17:59:16 -0700

On 9/14/2023 12:33 PM, Jim Porter wrote:
> On 9/14/2023 12:14 PM, Stefan Kangas wrote:
>> I have sent the logs to you off-list.
> 
> Thanks for the logs. They're pretty large, so I'll have to take some 
> extra time to look them over, but I'm beginning to wonder if there's a 
> bug in process.c. In the logs for em-unix-tests, I see messages for the 
> start of the "echo" process and its sentinel, but nothing for the output 
> filter, even though we *should* be getting output. I'll keep thinking 
> about this, and possibly add some further instrumentation to narrow this 
> down...

I pushed some minor changes to the logging (cadd3326625), and a 
mostly-unrelated fix to Eshell's iterative evaluation that I noticed 
from perusing the logs (one day I should really just rip that out and 
replace it with generator.el...).

In particular, I added logging for the actual command-line args that we 
pass to external commands. That should help to narrow down whether we're 
calling the subprocess correctly, or if there's some issue where the 
expected output gets lost on the way somewhere...

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.