andys@ulysses.att.com (05/17/89)
Earlier today, I reported a bug in rmail. I have narrowed down the problem to the vicinity of where rmail-nuke-pinhead-header is called. That function, and the code preceding it and following it, get looped through ad-infinitum, if the Unix-style header (From ) contains a time-zone specification of the form +nnnn. For your debugging pleasure, I attach a piece of mail which has been munged to stimulate the bug. The only change from a real piece of mail was to change the time zone for the *sender* only from EDT to +0200 the three places it appears. --- Andy Sherman/AT&T Bell Laboratories/Murray Hill, NJ *NEW ADDRESS* AUDIBLE: (201) 582-5928 *NEW PHONE* READABLE: andys@ulysses.ATT.COM or att!ulysses!andys *NEW EMAIL* The views and opinions are my own. Who else would want them? *OLD DISCLAIMER* -----Mail guaranteed to cause neurological impairment follows------------------ From arpa!rice.edu!sun-managers-relay Tue May 16 14:53:57 +0200 1989 Received: by rice.edu (AA25150); Tue, 16 May 89 13:57:08 CDT Sender: sun-managers-relay@rice.edu Received: from radio.astro.utoronto.ca by rice.edu (AA25123); Tue, 16 May 89 13:53:01 CDT Received: by radio.astro.utoronto.ca id AA00434; Tue, 16 May 89 14:53:57 +0200 Date: Tue, 16 May 89 14:53:57 +0200 From: Brian Glendenning <brian@radio.astro.utoronto.ca> Message-Id: <8905161853.AA00434@radio.astro.utoronto.ca> To: sun-managers@rice.edu Cc: slezak@lll-lcc.llnl.gov In-Reply-To: Tom Slezak's message of Tue, 16 May 89 08:35:25 PDT <8905161535.AA14273@lll-lcc.llnl.gov> Subject: Mystery crashes - (disk related crashes solved) Status: RO OK - I've found at least one source of crashes (thanks largely to kensmith@cs.buffalo.edu who kindly adb'd the core and told me what he found, which allowed me to track it down). When we upgraded to 4.0.1 from 3.4 we left (restored) our old root and /usr on disk to facilitate restoring local modifications. In our old system the /dev had entries for xy2, which we don't have on this system (and which we haven't had for a long time under 3.x either). Whenever anything (find, du, ls, rm) tried to access the first "nonexistent" directory entry (xy2a), the system crashed. This seems overly violent to me, but it least it's easily fixed. There's at least one satisfied reader of sun-managers out here :-) Brian -- Brian Glendenning - Radio astronomy, University of Toronto brian@radio.astro.utoronto.ca uunet!utai!radio!brian glendenn@utorphys.bitnet