Result is: [ ] - after several test messages to a list
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Systers-mailman |
Fix Released
|
Critical
|
Andy Grover | ||
Development |
Fix Released
|
Critical
|
Andy Grover |
Bug Description
Verification by Jen:
1 - Postfix is definitely accepting the message (/var/log/mail.log)
2 - There is no problem sending outgoing messages from sheinvents to members on a specific list:
/usr/local/
from you here you can do a:
mail <address at blah>
3 - This is definitely not a network latency issue
4 - Tail of logs/info shows when new post is received:
Dec 04 15:09:56 2008 (11322) Result is: []
Dec 04 15:09:56 2008 (11322) msg['Subject'] = /Will this get logged?/
Dec 04 15:09:56 2008 (11322) post_addr_
Dec 04 15:09:56 2008 (11322) footer-text = /
To subscribe to this conversation, send email to <email address hidden> or visit <http://
To contribute to this conversation, use your mailer's reply-all or reply-group command or send your message to <email address hidden>
To start a new conversation, send email to <email address hidden>
To unsubscribe entirely from dlist-test, send email to <email address hidden> with subject unsubscribe./
Dec 04 15:09:56 2008 (11322) Result is: []
5 - Mailman queue timestamps are being incremented for outgoing posts
qfiles/dlist, in, out, archive
6 - When a member is added to the list, the first post from the new user appears as in logs/info as:
Dec 04 15:28:21 2008 (28177) DlistUtils: Making SQL query 'SELECT COUNT(*) FROM subscriber WHERE <email address hidden>''
Dec 04 15:28:22 2008 (28177) DlistUtils: query succeeded
Dec 04 15:28:22 2008 (28177) DlistUtils: Making SQL query 'INSERT INTO subscriber (mailman_key, preference, deleted, format, suppress) VALUES ('<email address hidden>', 1, FALSE, 1, 0)'
Dec 04 15:28:22 2008 (28177) DlistUtils: query succeeded
Dec 04 15:28:22 2008 (28177) Result of query is /None/
Additionally, I received a bounced message when trying to post:
This is the mail system at host mango.buunabet.com.
I'm sorry to have to inform you that your message could not
be delivered to one or more recipients. It's attached below.
For further assistance, please send mail to postmaster.
If you do so, please include this problem report. You can
delete your own text from the attached returned message.
<email address hidden>: host sheinvents.
5.1.1 <email address hidden>: Recipient address rejected: User
unknown in local recipient table (in reply to RCPT TO command)
Reporting-MTA: dns; mango.buunabet.com
X-Postfix-Queue-ID: C069512206
X-Postfix-Sender: rfc822; <email address hidden>
Arrival-Date: Thu, 4 Dec 2008 15:30:05 -0800 (PST)
Final-Recipient: rfc822; <email address hidden>
Action: failed
Status: 5.1.1
Remote-MTA: dns; sheinvents.com
Diagnostic-Code: smtp; 550 5.1.1 <email address hidden>: Recipient
address rejected: User unknown in local recipient table
7. Fixed a permissions problem with Handlers/Dlist.pyc - changed from 660 to 664 and I received the post to my newly subscribed address.
8. Bounced message although post seems to have gone through:
Your message was rejected because it was sent to an invalid address. If you want to add a message to an existing conversation,send it to <email address hidden>, replacing 'conversation' with the name of the existing conversation. If you want to start a new conversation on dlist-test, send your message to <email address hidden>
Subject:
Re: [Dlist-test] Please let me know if anyone receives
From:
Jennifer Redman <email address hidden>
Date:
Thu, 04 Dec 2008 16:27:07 -0800
CC:
<email address hidden>
Return-Path:
<email address hidden>
X-Original-To:
<email address hidden>
Delivered-To:
<email address hidden>
Received:
from mango.buunabet.com (mango.buunabet.com [198.145.64.178]) by sheinvents.com (Postfix) with ESMTP id 9DA2510210116 for <email address hidden>; Thu, 4 Dec 2008 16:27:10 -0800 (PST)
Received:
by mango.buunabet.com (Postfix, from userid 58) id D31C61226B; Thu, 4 Dec 2008 16:27:09 -0800 (PST)
X-Spam-
SpamAssassin 3.2.5 (2008-06-10) on mango.buunabet.com
X-Spam-Status:
No, score=-4.3 required=5.0 tests=ALL_
Received:
from pineapple.
Message-ID:
<email address hidden>
Reply-To:
<email address hidden>
User-Agent:
Thunderbird 2.0.0.18 (Macintosh/
MIME-Version:
1.0
References:
<email address hidden>
In-Reply-To:
<email address hidden>
Content-Type:
multipart/
Wonder if I can reply to this message
#####
This bug is based on this email from Robin:
I have been testing things with robin-test for the last couple of days. I used my work email (<email address hidden>) and my home email (<email address hidden>) as the two members and things worked well (not like kathy-test which wasn't sending any mail). Kathy joined with 2 email addresses (hotmail and yahoo) today. I asked a syster to join this list and send mail to it, to test out the garblings. The mail went to both of kathy's emails but to neither of mine. I have since tried sending from each email separately to make sure they both are impacted, and it's now been 4 hours since the last one was sent, so I'm pretty sure that it's not coming through (I'll send mail in the morning if it's really just slowness, but the first one was sent at 11am this morning, so this seems waaay too long.)
My guess is that this is related to the problem with kathy-test where the recipient list retrieved from the database was nil. I have no idea what is going on, but this seems like a serious problem, and certainly impacts anything we do about testing sent and received mail. If folks can try to track this down while I am gone, this would be a big help. There is a message in the info log that represents the set of recipients for the message -- here is an example for a message sent out today (Tuesday) -- and those are the recipients who actually got this message
'Dec 02 12:41:14 2008 (11322) Result is: ['<email address hidden>', '<email address hidden>']
I have a guess that this has to do with the code that does the "suppress", which is whether you get all the messages or just the first in each thread. My guess is that it is either being written or read incorrectly, but that's just a guess.
Yes, you can solve it by creating a new list, but at some point, I believe it will go back into this mode again, and I wouldn't trust the tests you've run because once we fix it, the fix might mess other things up.
Changed in systers: | |
assignee: | nobody → andrew-grover |
importance: | Undecided → Critical |
milestone: | none → codefreeze |
status: | New → Confirmed |
Changed in systers: | |
assignee: | andrew-grover → andy-groveronline |
description: | updated |
description: | updated |
description: | updated |
description: | updated |
Changed in systers: | |
status: | Confirmed → Fix Released |
committed in r36. Please retest.