387677 - "Check for new message every x minutes" stopped working for POP3

Status: RESOLVED FIXED

(MailNews Core :: Networking: POP, defect)

Description

[Robert Kaiser]

For POP3 servers, the "Check for new message every x minutes" setting in account manager has stopped working.

German newsgroup regulars report this has happened between builds from 200705030320 CEST and 200705040251 CEST, leaving a regression range of http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2007-05-02+18%3A20%3A00&maxdate=2007-05-03+19%3A15%3A00&cvsroot=%2Fcvsroot

Suspecting the change needs to be in mailnews/ leads to http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=all&branch=HEAD&branchtype=match&dir=mozilla%2Fmailnews&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2007-05-02+18%3A20%3A00&maxdate=2007-05-03+19%3A15%3A00&cvsroot=%2Fcvsroot as the changes that can have caused this.

From glancing on those checkins, I'd suspect bug 379070 as the culprit (once again).

Comment 1

[Jürgen 'JiHa' Harter]

For those of you understanding german, above mentioned newsgroup thread is http://groups.google.de/group/de.comm.software.mozilla.nightly-builds/browse_frm/thread/ce9dfbf9b34e340c/609abcf53666b894?tvc=1#609abcf53666b894

Some additional information from there:
It seems as if mail is checked once and therefore new mail was fetched no further checks are done.

'sudo tail -f /var/log/messages' (gentoo; SeaMonkey-nightly; "Check for new message every '2' minutes"; fetching mail from 127.0.0.1) shows:

Jun 13 05:47:13 t900 xinetd[5006]: START: pop-3 pid=20480 from=127.0.0.1
Jun 13 05:47:13 t900 xinetd[5006]: EXIT: pop-3 status=0 pid=20480 duration=0(sec)
Jun 13 05:49:13 t900 xinetd[5006]: START: pop-3 pid=20483 from=127.0.0.1
Jun 13 05:49:13 t900 xinetd[5006]: EXIT: pop-3 status=0 pid=20483 duration=0(sec)
Jun 13 05:50:01 t900 cron[20486]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons )
Jun 13 05:51:13 t900 xinetd[5006]: START: pop-3 pid=20498 from=127.0.0.1
Jun 13 05:51:13 t900 xinetd[5006]: EXIT: pop-3 status=0 pid=20498 duration=0(sec) 

SM checks for new messages every 2 minutes what is correct.

Then writing a message to yourself:
hafi@t900 ~ $ mail hafi
Subject: t
t
.
Cc:
hafi@t900 ~ $ date
Wed Jun 13 05:51:27 CEST 2007 

This mail is shown in the log:
Jun 13 05:51:26 t900 postfix/pickup[15382]: 37F72A68613: uid=1000 from=<hafi>
Jun 13 05:51:26 t900 postfix/cleanup[20508]: 37F72A68613: message-id=<[e-mail]>
Jun 13 05:51:26 t900 postfix/qmgr[12143]: 37F72A68613: from=<[e-mail]>, size=290, nrcpt=1 (queue active)
Jun 13 05:51:26 t900 postfix/local[20510]: 37F72A68613: to=<[e-mail]>, orig_to=<hafi>, relay=local,
delay=0.06, delays=0.04/0.01/0/0.01, dsn=2.0.0, status=sent (delivered to mailbox)
Jun 13 05:51:26 t900 postfix/qmgr[12143]: 37F72A68613: removed 

SM checks and fetches it:
Jun 13 05:53:13 t900 xinetd[5006]: START: pop-3 pid=20513 from=127.0.0.1
Jun 13 05:53:13 t900 xinetd[5006]: EXIT: pop-3 status=0 pid=20513 duration=0(sec)
Jun 13 06:00:01 t900 cron[20521]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons )
Jun 13 06:00:01 t900 cron[20523]: (root) CMD (rm -f /var/spool/cron/lastrun/cron.hourly) 

From then on, no further checks are done.
Next check should have been at 05:55:13.

Comment 2

[Andrew Schultz]

Attachment: patch

I'm seeing this, but the broken code seems from a later date than the regression window stated in comment 0.  I'm seeing a regression from bug 33451 (a la bug 355008 comment 11) from part that landed on 4/24.

Comment 3

[David :Bienvenu]

Is that the right patch? I can't review dom patches, for one thing; for another, that doesn't seem like it could have anything to do with the file spec or string changes.

Comment 4

[Andrew Schultz]

Attachment: patch

this is the right patch

Comment 5

[David :Bienvenu]

Comment on attachment 272656 [details] [diff] [review]
patch

thx, that looks much better :-) thx for the patch!

Comment 6

[Andrew Schultz]

FIXED

Comment 7

[Sven Grull]

Could it be that this checkin broke "check for new messages at startup"? 
I just switched from SM 1.5 (trunk) to SM 2.0 (trunk build 20070727) and SeaMonkey won't check for new messages (POP3) at startup anymore although the pref is still set.

Comment 8

[Caio Tiago Oliveira (asrail)]

No, because it still working here.