The 5.9.5 release of fetchmail is now available at the usual locations,
including <URL:http://www.tuxedo.org/~esr/fetchmail>.
The source archive is available at:
<URL:http://www.tuxedo.org/~esr/fetchmail/fetchmail-5.9.5.tar.gz>
Here are the release notes:
fetchmail-5.9.5 (Thu Nov 8 14:14:35 EST 2001), 21162 lines:
* Changed the logging logic along lines suggested by Jan Klaverstijn,
* fetchmailconf looks first in the directory it's running from to find
fetchmail.
* Make sure we get a success status correctly from open_smtp_sink()
and open_bsmtp_sink().
* Matthias Andree's env.c patch to refuse service when QMAILINJECT is defined.
* Immediately abort if a non-empty QMAILINJECT environment variable is
found. If it is set and contains f or i, qmail-inject or qmail's
sendmail `compatibility' wrapper will rewrite From: or Message-ID:
headers, respectively. En passant, fix the bug that program_name was
not filled in before used when the user's ID had no PW entry, leading
to (null) or crash when printing the error message. Patch by Matthias
Andree.
* NextStep and OpenStep port patch from Eric Sunshine.
* Block signals during SockConnect() so we don't get a socket descriptor
leak if we're hit by an alarm signal during connect(2).
* Set queryname even when server is inactive; avoids a core-dump bug in
the fetchids code.
There are 396 people on fetchmail-friends and 627 on fetchmail-announce.
The important fixes in this version are the ones to (1) plug a socket
descriptor leak on timeout, and (2) get a success status correctly
when using the MDA option.
By popular demand, diffs from the previous release have been omitted.
--
<a href="http://www.tuxedo.org/~esr/">Eric S. Raymond</a>
"America is at that awkward stage. It's too late to work within the system,
but too early to shoot the bastards."
-- Claire Wolfe