Age | Commit message (Expand) | Author |
2009-03-15 | since we are going to share code in smtp sessions and mta sessions, we need | Gilles Chehade |
2009-03-15 | save 4 bytes per message by moving the datafp field of struct message to | Gilles Chehade |
2009-03-11 | fix log_info(), while other processes know about message uid, the smtp | Gilles Chehade |
2009-03-10 | implement basic logging, needs more work; ok gilles@ | Jacek Masiulaniec |
2009-03-08 | shrink struct message a bit by removing a couple fields which are no longer | Gilles Chehade |
2009-02-22 | put repeated inet_ntop calls into new func ss_to_text, which uses | Jacek Masiulaniec |
2009-02-19 | don't let libevent buffer long lines forever; ok gilles@ | Jacek Masiulaniec |
2009-02-18 | previous commit was incomplete, this fixes ssmtp | Gilles Chehade |
2009-02-18 | smtp_accept() now requests from lka that it performs the hostname lookup | Gilles Chehade |
2009-02-17 | after a message has been commited to queue, do not incondtionnally clear | Gilles Chehade |
2009-02-13 | Ensure ss is never dereferenced when NULL. Can't happen with current code, | Jacek Masiulaniec |
2009-01-30 | when decreasing ssl related counters, make sure the session was flagged as | Gilles Chehade |
2009-01-30 | improve statistics for smtp process. not only collect the current sessions | Gilles Chehade |
2009-01-30 | clear the F_EVLOCK flag earlier to prevent the error event handler from | Gilles Chehade |
2009-01-30 | bump descriptors limit to the max and set the maximum connections count to | Gilles Chehade |
2009-01-30 | fix a very annoying events masking issue which would cause a fatal() to be | Gilles Chehade |
2009-01-29 | Implement "smtpctl show stats"; ok gilles@ | Jacek Masiulaniec |
2009-01-29 | missing prototype for session_respond() in mta, move session_respond()'s | Gilles Chehade |
2009-01-29 | - remove debug message | Gilles Chehade |
2009-01-29 | Received header line was incomplete for mail submitted through the enqueuer | Gilles Chehade |
2009-01-28 | reuse recipient_to_path; ok gilles@ | Jacek Masiulaniec |
2009-01-28 | Make races between queue and runner impossible by implementing the policy: | Jacek Masiulaniec |
2009-01-28 | add a struct path to struct message so that we can keep track of the RCPT | Gilles Chehade |
2009-01-12 | dot escaping, as required by rfc; ok gilles@ | Jacek Masiulaniec |
2009-01-04 | aliases/forwards expansion was not done correctly and a race could | Gilles Chehade |
2009-01-01 | remove unnecessary includes; ok gilles@ | Jacek Masiulaniec |
2008-12-22 | cleanup | Jacek Masiulaniec |
2008-12-22 | To reset state, it's enough to set s->s_state = S_HELO, rcptcount | Jacek Masiulaniec |
2008-12-21 | fix session flags resetting; ok gilles@ | Jacek Masiulaniec |
2008-12-21 | - missing prototype + smtp.c was misusing session_auth_pickup() | Gilles Chehade |
2008-12-21 | - AUTH PLAIN may receive credentials as a parameter to AUTH or on a | Gilles Chehade |
2008-12-20 | - import first bricks of SMTP AUTH support. currently only AUTH PLAIN is | Gilles Chehade |
2008-12-18 | Introduce safe_fclose, which tries to push file to the disk as | Jacek Masiulaniec |
2008-12-18 | Declarations for functions used only in smtp_session.c were moved | Jacek Masiulaniec |
2008-12-18 | Check fwrite return code at DATA stage. | Jacek Masiulaniec |
2008-12-18 | Cleanup /incoming before handling each MAIL FROM. | Jacek Masiulaniec |
2008-12-13 | IMSG_* namespace cleanup. | Jacek Masiulaniec |
2008-12-07 | Disable EV_READ when sending IMSG_PARENT_AUTHENTICATE. | Jacek Masiulaniec |
2008-12-07 | Replace evbuffer_add_printf calls with wrapper function, session_respond, | Jacek Masiulaniec |
2008-12-07 | Don't check / reset s->s_msg.datafp where its state is obviously known. | Jacek Masiulaniec |
2008-12-06 | In session_destroy, use "if (s->s_state >= S_MAIL)", and not | Jacek Masiulaniec |
2008-12-06 | Don't include <err.h> where log.c API must be used. | Jacek Masiulaniec |
2008-12-06 | NULL-ify s_msg.datafp upon fclose unconditionally. | Jacek Masiulaniec |
2008-12-06 | evbuffer_readline already strips <CRLF> so that callers don't have to. | Jacek Masiulaniec |
2008-12-05 | - cosmethic, no functionnal change | Gilles Chehade |
2008-12-04 | - in session_read(), set EV_WRITE if we are going to send a "transaction | Gilles Chehade |
2008-12-04 | - when in state S_DATACONTENT, do not disable EV_READ if the last line we | Gilles Chehade |
2008-12-04 | - when doing the session timeout lookup, do not remove the last session | Gilles Chehade |
2008-12-04 | - fix event masking for DATA and make DATA look more like MAIL and RCPT | Gilles Chehade |
2008-12-03 | - fix event masking issues in smtp process which could lead to a fatal() if | Gilles Chehade |