| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2012-01-13 | kill delivery_backend_lookup() proto as it is already in smtpd.h | Charles Longeau | |
| committed on behalf of gilles@ ok gilles@ chl@ | |||
| 2012-01-12 | we should never use ramqueue_first_envelope() as a mean to determine the | Gilles Chehade | |
| first schedulable envelope otherwise we end up with nsched wrong ... heck, let's kill ramqueue_first_envelope() and be done with it ;) | |||
| 2012-01-12 | The status field in the envelope is confusing. Its only purpose is to | Eric Faurot | |
| notify the runner of what happened with an envelope that has been scheduled. It is not part of the state of the envelope, and it is not even dumped. So it should only be set by mta/mda, checked by runner to decide what to do with the envelope, and ignored everywhere else. ok gilles@ | |||
| 2012-01-12 | the ramqueue filling at startup was busted, it would load envelopes one by | Gilles Chehade | |
| one instead of doing it efficiently. fix runner_timeout() and ramqueue_load() so that at startup smtpd fills the ramqueue as long as there's no schedulable envelope in it, interrupts filling if there is and resume once it's scheduled. bug spotted by Nathanael Rensel, bug fix by me w/ help from eric@ tested by eric@ and I | |||
| 2012-01-12 | now that we no longer do a stateful iteration on schedule queue, we need | Gilles Chehade | |
| to make ramqueue_next_envelope() aware of pauses in mda/mta. while at it kill the pausing of bounces since they are reinjected in smtp and end up paused by the mda/mta cases. fixes an infinite loop observed by eric@ when pausing deliveries and trying to interrupt smtpd while it attempts to fetch next envelope :) tested by eric@ and I, ok eric@ | |||
| 2012-01-12 | use mbox backend for mbox delivery. | Eric Faurot | |
| ok gilles@ | |||
| 2012-01-12 | fgets(3) returns a pointer, so compare against NULL, not integer 0. | Antoine Jacoutot | |
| from Michael W. Bombardieri ok miod@ | |||
| 2012-01-12 | Remove dead code for config reloading for now. It is not functionnal | Eric Faurot | |
| and confusing. ok gilles@ | |||
| 2012-01-12 | Add a fsqueue_envelope_dump_atomic() function that writes an envelope | Eric Faurot | |
| to incoming/envelope.tmp before moving it to its final destination. This allows to make sure that envelopes created directly in the queue (such as bounces) can't let the queue in an inconsistent state if the program stops in the middle of an envelope dump. This also allow to simplify qwalk, since we don't put any temporary contents anywhere in "walkable" queue dirs. ok gilles@ | |||
| 2012-01-12 | remove envelope_get_errormsg() and move envelope_set_errormsg() | Eric Faurot | |
| to envelope.c ok gilles@ | |||
| 2012-01-12 | alter line contents before writing it, rather than after... | Eric Faurot | |
| ok gilles@ | |||
| 2012-01-12 | use absolute path as fts may chdir() during traversal. | Eric Faurot | |
| ok gilles@ | |||
| 2012-01-12 | Mention the information source used for the ASN lookup. | Stuart Henderson | |
| 2012-01-12 | Zap unused var in asn_lookup; from Michael W. Bombardieri | Stuart Henderson | |
| 2012-01-11 | fix flags writing | Eric Faurot | |
| ok gilles@ | |||
| 2012-01-11 | remove stateful iteration from ramqueue, if we ever need to reintroduce it | Gilles Chehade | |
| we'll do it, but it isn't used and causes potential bugs idea by Nathanael Rensel, diff by me, ok eric@ | |||
| 2012-01-11 | in ramqueue_schedule(), do not reorder ramqueue envelopes when forcing a | Gilles Chehade | |
| reschedule of the entire queue as that can lead to an infinite loop. bug spotted and bugfix by Nathanael Rensen, thanks ! ok eric@ | |||
| 2012-01-11 | queue_fsqueue_ascii.c, you're not welcome anymore | Gilles Chehade | |
| 2012-01-11 | implement an envelope_ascii API that's not tied to a specific queue_backend | Gilles Chehade | |
| simplify queue_fsqueue | |||
| 2012-01-11 | remove bogus call | Eric Faurot | |
| ok gilles@ | |||
| 2012-01-11 | Try to parse hostnames as IP addresses before resolving. This allows | Eric Faurot | |
| relays to be given as IP address in the config file. ok gilles@ | |||
| 2012-01-11 | enable back CA support, just don't verify client ... | Gilles Chehade | |
| tested @ home | |||
| 2012-01-11 | Simplify runner/queue by getting rid of Q_PURGE. Instead, let smtpd | Eric Faurot | |
| periodically clear the purge/ directory. At init time, the fsqueue backend simply moves the existing incoming/ dir in purge/ to discard aborted sessions. ok gilles@ chl@ | |||
| 2012-01-11 | Finally remove the queue_message_update() function which ended up | Eric Faurot | |
| being only called by bounce sessions, so most of the code there was actually useless. The envelope is directly deleted or updated at the relevant place. ok gilles@ | |||
| 2012-01-11 | Improve error reporting. Most errors during hostname lookup are | Eric Faurot | |
| now correctly reported as temporary failures. from Nathanael Rensen, tweaks by me. | |||
| 2012-01-11 | remove dead prototype | Eric Faurot | |
| from Nathanael Rensen ok gilles@ | |||
| 2012-01-11 | fix typo | Eric Faurot | |
| from Nathanael Rensen ok gilles@ | |||
| 2012-01-09 | Rename whatis.{db,index} back to mandoc.{db,index} | Ingo Schwarze | |
| such that makewhatis(8) called from pkg_delete(1) does not fail trying to open old-style whatis.db files as btree(3). Suggested by espie@ deraadt@ millert@ | |||
| 2012-01-07 | sloppy Ingo, whatis.db + whatis.index | Marc Espie | |
| fix PkgCheck to use the proper API. | |||
| 2012-01-06 | typo in log_peer_warnx message | Stuart Henderson | |
| 2012-01-06 | mkdir needs -m to set the mode | Sebastian Reitenbach | |
| OK yasuoka@ | |||
| 2012-01-05 | Enable the new apropos(1), whatis(1), and makewhatis(8). | Ingo Schwarze | |
| Unlink the old apropos(1), whatis(1), and makewhatis(8) from the build. Call the new tools from pkg_create(1) and pkg_add(1). "Please enable it now." deraadt@ | |||
| 2012-01-05 | update to nginx-1.0.11 | Robert Nagy | |
| 2011-12-31 | Remove the EXTENSIONS ifdef danse; it might have made sense in the past | Antoine Jacoutot | |
| but user(8) behaviour has become OS specific in time so there's no sense keeping this. We are OpenBSD and we use BSD EXTENSIONS unconditionally. ok miod@ millert@ deraadt@ | |||
| 2011-12-30 | Fix a warning. | Antoine Jacoutot | |
| inputs from and ok miod@ | |||
| 2011-12-28 | Print PPP-Max-Payload headers (for RFC 4638 PPPoe), part of a larger diff | Stuart Henderson | |
| from Matt Dainty. ok haesbaert@ | |||
| 2011-12-28 | s/minum/minimum/ in comment | Stuart Henderson | |
| 2011-12-28 | only add the rdomain for hostname lookups, when we have a hostname to lookup. | Peter Hessler | |
| noticed by zepard at gmail OK henning@ | |||
| 2011-12-27 | Q_BOUNCE is not used anymore | Eric Faurot | |
| ok gilles@ | |||
| 2011-12-27 | Instead of using a separate "bounce" queue, create the bounce envelope | Eric Faurot | |
| directly as an envelope of the bounced message, just like "regular" envelopes. ok gilles@ | |||
| 2011-12-27 | Follow adduser(8) behaviour (as well as what the pkg tools and our | Antoine Jacoutot | |
| installer do) and do not append the user to its primary group name in /etc/group. ok miod@ | |||
| 2011-12-25 | Missing architecture on the .Dt line; found with mandocdb(8); ok jmc@. | Ingo Schwarze | |
| 2011-12-24 | formatting errors, found using freebsd's "igor"; | Jason McIntyre | |
| 2011-12-24 | -p requires an argument; from Stefan Pettersson | Jason McIntyre | |
| ok krw | |||
| 2011-12-23 | some spelling fixes, found using freebsd's "igor" document verifier; | Jason McIntyre | |
| 2011-12-23 | On envelope creation, get the message id in a saner way than the | Eric Faurot | |
| current confusing hack. ok gilles@ | |||
| 2011-12-22 | Add fqueue_envelope_path() function to create path to envelopes. | Eric Faurot | |
| ok chl@ gilles@ | |||
| 2011-12-21 | disable temporarily CA support, it prevents some remote hosts from | Charles Longeau | |
| establishing a ssl session. will be investigated and reenabled soon asked and commited on behalf of gilles@ ok gilles@ | |||
| 2011-12-19 | fix/improve envelope_validate(): | Eric Faurot | |
| - return an informative error string if the envelope is invalid. - take the envelope id as a parameter and make sure it matches. - do not expect the errorline to start with an SMTP response code, as this is not always the case: a temporary failure with mda would cause the envelope to be marked as corrupted. Instead, just make sure that all string fields are actual strings to prevent overflows later. ok gilles@ chl@ | |||
| 2011-12-18 | Sync comments with latest cleanup changes | Charles Longeau | |
| ok eric@ | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |