summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2013-10-06Bug 1289: Clarify host list processing failuresTodd Lyons
Enhance explanation of +ignore_unknown.
2013-10-05Add exiqsumm fix to ChangeLogTodd Lyons
2013-10-05Fix exiqsumm output for single queue item.Todd Lyons
Patch from Richard Hall, with one minor addition to prevent uninitialized value error during output.
2013-10-04Bug 1392: Change status text when no dmarc recordWolfgang Breyha
2013-10-03Fix DKIM variable name in documentation.Ted Cooper
2013-10-03Print tarball names in debug mode tooTodd Lyons
2013-10-03Filter xfpt needs new macro tooTodd Lyons
2013-10-03Finish the xfpt fix.Phil Pennock
The readers of the version expect no surrounding whitespace; putting the content in-place requires macro expansion, which requires that .literal mode be `layout` or `off`, but either of those will escape the XML angle brackets. The easiest fix is to generate the XML as part of the macro definition. Moved local_params rules out of Makefile and into GenLocalParams
2013-10-03Bug 1389: Initialize and clear variablesTodd Lyons
2013-10-03Merge branch 'master' of ssh://git.exim.org/home/git/eximTodd Lyons
2013-10-03Unbreak HTML build for RC candidates.Phil Pennock
The HTML build now uses the website repo, which extracts the version number from the XML (generated from the .xfpt). Meanwhile, commit 2aee48d6 made the version number in the build process dynamic, taking the value from the release script (via an environ variable). This change fixes the invocation to pass the version _without_ an RC suffix to the XML build, letting HTML generation happen.
2013-10-03Mention Redis lookup in NewStuffTodd Lyons
2013-10-03Setting LC_ALL with make overrode this sort checkexim-4_82_RC2Todd Lyons
2013-10-02Fix docbook paths for xml generationTodd Lyons
2013-10-01Bugzilla 1217: Experimental Redis lookupTodd Lyons
Add want_experimental() test in the script to create the lookups Makefile to ease detection of requested Experimental features, and simplify the #ifdef guards in the redis.c.
2013-09-30Do not use MSG_NOSIGNAL on send() in ${udpsend}Jeremy Harris
Some non-Posix systems don't define it. Anyway, Exim ignores SIGPIPE for most purposes.
2013-09-30Merge branch 'patch-3' of https://github.com/bes-internal/exim into ↵Todd Lyons
master_dmarc_doc
2013-09-30Added documentation of features, acknowledgementsTodd Lyons
2013-09-30Add ratelimit doc addition to ChangeLogTodd Lyons
2013-09-30Typo re-fix in ratelimit.pl docTodd Lyons
2013-09-30Merge branch 'patch-1' of https://github.com/bes-internal/exim into ↵Todd Lyons
master-bes-ratelimit.pl
2013-09-30Merge branch 'master_tpda'Todd Lyons
2013-09-30Adjust test build link paths.Todd Lyons
Details at: http://comments.gmane.org/gmane.mail.exim.user/91154 Add ignore for a logfile from test run.
2013-09-30Bug 1031: Experimental TPDAJeremy Harris
Remove whitespace
2013-09-30Remove whitespaceTodd Lyons
2013-09-30fix dmarc_status condition documentationbes-internal
2013-09-30Fix dovecot with empty 334 challenge.Phil Pennock
Thomas Morper reported, with 4.82RC1, that he saw "334 NULL" as the challenge when using AUTH PLAIN to Dovecot when the client does not send an initial response. I could replicate. This was caused by commit 3f1df0e3 on 2012-11-19 (PP/13 of 4.82); I was too cautious in the robustness fixes; the clue came in this line of debug output: 76430 dovecot: warning: ignoring trailing tab This change removes that check, and documents in a comment that this input is acceptable protocol-wise, and why. With this fix: AUTH PLAIN 334 AGZyZWRlcmljAGh1bXB0eS1kdW1wdHk= 235 Authentication succeeded
2013-09-28small typobes-internal
2013-09-28Correct spelling of dmarc_disable_verifyJeremy Harris
2013-09-28Add cscope.files build targetJeremy Harris
2013-09-28Fix non-IPv6 compile on v6-less systemJeremy Harris
2013-09-28ratelimit.pl: format usage section, print debug to STDERR, add examplesbes-internal
2013-09-27Document hexquote & udpsend in NewStuff/ChangeLogPhil Pennock
Tony documented his new features in 7a5698fa; this change simply adds them to the list of things that have changed.
2013-09-26Clarify CL: the CVE security fix already in 4.80.1Phil Pennock
On re-reading the text for 4.82, it read as though there were a new security fix which might require an upgrade. Clarified that this fix has already been released (in 4.80.1).
2013-09-24Additions to the NewStuff summary fileTodd Lyons
2013-09-23Add documentation for new $authenticated_fail_idTodd Lyons
2013-09-22Bug 1287 - Fix tls_require_certTodd Lyons
2013-09-22Prevent TLS rebinding in LDAP connectionsTodd Lyons
Bugzilla 1375
2013-09-22Add expansion $authenticated_fail_idTodd Lyons
2013-09-15ChangeLogJeremy Harris
2013-09-15Fix numeric comparisons for 64b. Bug 1385Jeremy Harris
2013-09-04ChangeLog is in chronological order within each release.Phil Pennock
numbers are sequential, namespaced by users.
2013-09-04Merge branch 'master' of ssh://git.exim.org/home/git/eximTodd Lyons
2013-09-04Make sender/recipient search case-insensitiveTodd Lyons
2013-09-04tls_dhparam size constraint suggestions.Phil Pennock
Between NSS and Debian patching of older Exim releases, there's a narrow range of values likely to interoperate well. Document this.
2013-09-01GnuTLS website movesPhil Pennock
2013-07-31Fix segfault in stdio with non-SMTP MIME ACL.Phil Pennock
When injecting a message locally in non-SMTP mode, and with MIME ACLs configured, if the ACL rejected the message, Exim would try to `fprintf(NULL, "%s", the_message)`. This fixes that. Most ACLs are plumbed in SMTP-only and looking through the others in receive.c, they all appear to be safely guarded, so it was just this one that slipped through. Crash report and assistance tracking down the root cause from Warren Baker.
2013-07-22Fix debug output in ${acl }Jeremy Harris
2013-07-21More tidying of ACL-config skipJeremy Harris
2013-07-21Remove ACL-config skip so that ${acl } expansions work from queue-runs.Jeremy Harris
Previously we skipped parsing the ACL section when not needed. Now it is potentially needed in all cases. The skip was ~5% faster than a full parse so probably not a large part of the exim process startup. Fix up testsuite output files affected by the removal and add a regression test.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-26 04:17:35 +0000