I have a list owner who runs a list designed to be subscribed to from text-messaging cell phones (list traffic consists only of a daily haiku the phone, the user replies to it, and then... nada. Seems to go into a
- very neat). People can use the web interface to sign up, but the confirmations are a problem. The subscription confirmation arrives at
bit bucket. I've confirmed that the ID cookie does make it across just fine.
A little later, the list owner receives a copy of the message below. At first I thought there was some discrepancy between the Return-Path: and the From: headers, but when I send a regular email from the phone to myself, they both come through fine, and match. I'm stumped.
The list owner *really* wants to turn off confirmations altogether, but I don't see any way to do this. Is there one? If confirmations absolutely can't be disabled, any ideas on why cell phone subscriptions fail, or what can be done about it?
Thanks, Scot
*From: *mailman-bounces@tinywords.com *Date: *May 2, 2005 9:51:59 PM PDT *To: *Haiku-owner@tinywords.com *Subject: Uncaught bounce notification * The attached message was received as a bounce, but either the bounce format was not recognized, or no member addresses could be extracted from it. This mailing list has been configured to send all unrecognized bounce messages to the list administrator(s).
For more information see: http://tinywords.com/mailman/admin/haiku_tinywords.com/bounce
From: *5102925586@vtext.com *Date: *May 2, 2005 9:51:54 PM PDT *To: *Haiku-bounces@tinywords.com *Subject: confirm f9bc6456b89ec9c10360aaa140411e154f0ff60b
Want to subscribe to tinywords? Please reply to this message
FWIW, I found the command-line tools add_members and remove_members,
which allowed me to write a simple shell script to manage subscribes
and unsubscribes by drawing from files. So we'll have PHP forms for
subbing and unsubbing, which will write to new_subs.txt and
del_subs.txt. Then we just run this shell script once per hour, and get
around the confirmations problem with cell phones (though I'm surprised
-- aren't there any other mailman lists out there with people subbing
their cell phones? you'd think there would be a better way...)
#!/bin/sh # Add new subscribers to mailing list from text file # Then remove subscribers from an unsub list
# Define location of subscribers text file: subfile="/home/username/listname_newsubs.txt"
# Define location of unsub file: unsubfile="/home/username/listname_delsubs.txt"
# Add all addresses in the sub file /usr/local/cpanel/3rdparty/mailman/bin/add_members -w y -a y -r $subfile list_domain.com
# Now blank the file for future additions echo "" > $subfile
# Unsub all addresses in the unsub file /usr/local/cpanel/3rdparty/mailman/bin/remove_members -n -f $unsubfile list_domain.com
# Now blank the file for future additions echo "" > $unsubfile
Works great!
Scot Hacker wrote:
I have a list owner who runs a list designed to be subscribed to from text-messaging cell phones (list traffic consists only of a daily haiku the phone, the user replies to it, and then... nada. Seems to go into a
- very neat). People can use the web interface to sign up, but the confirmations are a problem. The subscription confirmation arrives at
bit bucket. I've confirmed that the ID cookie does make it across just fine.
A little later, the list owner receives a copy of the message below. At first I thought there was some discrepancy between the Return-Path: and the From: headers, but when I send a regular email from the phone to myself, they both come through fine, and match. I'm stumped.
The list owner *really* wants to turn off confirmations altogether, but I don't see any way to do this. Is there one? If confirmations absolutely can't be disabled, any ideas on why cell phone subscriptions fail, or what can be done about it?
Thanks, Scot
*From: *mailman-bounces@tinywords.com *Date: *May 2, 2005 9:51:59 PM PDT *To: *Haiku-owner@tinywords.com *Subject: Uncaught bounce notification * The attached message was received as a bounce, but either the bounce format was not recognized, or no member addresses could be extracted from it. This mailing list has been configured to send all unrecognized bounce messages to the list administrator(s).
For more information see: http://tinywords.com/mailman/admin/haiku_tinywords.com/bounce
From: *5102925586@vtext.com *Date: *May 2, 2005 9:51:54 PM PDT *To: *Haiku-bounces@tinywords.com *Subject: confirm f9bc6456b89ec9c10360aaa140411e154f0ff60b
Want to subscribe to tinywords? Please reply to this message
Mailman-Users mailing list Mailman-Users@python.org http://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: http://mail.python.org/mailman/options/mailman-users/shacker%40birdhouse.org
Security Policy: http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq01.027.htp
Argh! Just minutes after finishing this up, I got a message from a friend. Turns out you CAN disable subscription confirmations. See:
http://www.mail-archive.com/mailman-users@python.org/msg31306.html
Grrrr...
(although that solution will affect lists system-wide, rather than per-list; anyone know of a way to make tweaks like this per-list?).
Scot
Scot Hacker wrote:
FWIW, I found the command-line tools add_members and remove_members, which allowed me to write a simple shell script to manage subscribes
and unsubscribes by drawing from files. So we'll have PHP forms for subbing and unsubbing, which will write to new_subs.txt and del_subs.txt. Then we just run this shell script once per hour, and get around the confirmations problem with cell phones (though I'm surprised -- aren't there any other mailman lists out there with people subbing their cell phones? you'd think there would be a better way...)#!/bin/sh # Add new subscribers to mailing list from text file # Then remove subscribers from an unsub list
# Define location of subscribers text file: subfile="/home/username/listname_newsubs.txt"
# Define location of unsub file: unsubfile="/home/username/listname_delsubs.txt"
# Add all addresses in the sub file /usr/local/cpanel/3rdparty/mailman/bin/add_members -w y -a y -r $subfile list_domain.com
# Now blank the file for future additions echo "" > $subfile
# Unsub all addresses in the unsub file /usr/local/cpanel/3rdparty/mailman/bin/remove_members -n -f $unsubfile list_domain.com
# Now blank the file for future additions echo "" > $unsubfile
Works great!
Scot Hacker wrote:
I have a list owner who runs a list designed to be subscribed to from text-messaging cell phones (list traffic consists only of a daily haiku the phone, the user replies to it, and then... nada. Seems to go into a
- very neat). People can use the web interface to sign up, but the confirmations are a problem. The subscription confirmation arrives at
bit bucket. I've confirmed that the ID cookie does make it across just fine.
A little later, the list owner receives a copy of the message below. At first I thought there was some discrepancy between the Return-Path: and the From: headers, but when I send a regular email from the phone to myself, they both come through fine, and match. I'm stumped.
The list owner *really* wants to turn off confirmations altogether, but I don't see any way to do this. Is there one? If confirmations absolutely can't be disabled, any ideas on why cell phone subscriptions fail, or what can be done about it?
Thanks, Scot
*From: *mailman-bounces@tinywords.com *Date: *May 2, 2005 9:51:59 PM PDT *To: *Haiku-owner@tinywords.com *Subject: Uncaught bounce notification * The attached message was received as a bounce, but either the bounce format was not recognized, or no member addresses could be extracted
from it. This mailing list has been configured to send all
unrecognized bounce messages to the list administrator(s).
For more information see: http://tinywords.com/mailman/admin/haiku_tinywords.com/bounce
From: *5102925586@vtext.com *Date: *May 2, 2005 9:51:54 PM PDT *To: *Haiku-bounces@tinywords.com *Subject: confirm f9bc6456b89ec9c10360aaa140411e154f0ff60b
Want to subscribe to tinywords? Please reply to this message
Mailman-Users mailing list Mailman-Users@python.org http://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: http://mail.python.org/mailman/options/mailman-users/shacker%40birdhouse.org
Security Policy: http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq01.027.htp
Mailman-Users mailing list Mailman-Users@python.org http://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: http://mail.python.org/mailman/options/mailman-users/shacker%40birdhouse.org
Security Policy: http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq01.027.htp
On May 5, 2005, at 14:55, Scot Hacker wrote:
(although that solution will affect lists system-wide, rather than per-list; anyone know of a way to make tweaks like this per-list?).
You can set a specific list's subscribe_policy to 0 using bin/
withlist. (You won't be able to set it to 0 via the GUI or
config_list without setting ALLOW_OPEN_SUBSCRIBE.)
$ bin/withlist -l mylist m.subscribe_policy = 0 m.Save() {ctrl-D}
P.S. I have been able to use the normal subscribe/confirm scheme
from my phone, so it is probably a function of your carrier or your
phone. One quirk I've noticed in my lists is that if I use
VERP_CONFIRMATIONS=Yes, the reply address gets truncated in the reply
message, and I have to manually add the last few characters of the
domain name. In my case that is presumably a phone bug, since it is
picking up fewer reply address characters than it allows me to enter
and successfully send.
Jim Tittsler http://www.OnJapan.net/ GPG: 0x01159DB6 Python Starship http://Starship.Python.net/crew/jwt/ Mailman IRC irc://irc.freenode.net/#mailman
Jim Tittsler wrote:
You can set a specific list's subscribe_policy to 0 using bin/ withlist. (You won't be able to set it to 0 via the GUI or
config_list without setting ALLOW_OPEN_SUBSCRIBE.)$ bin/withlist -l mylist m.subscribe_policy = 0 m.Save() {ctrl-D}
Jim, this is cool to know, thanks. So is there a corresponding option for unsubscribe confirmations? I tried it with
m.unsubscribe_policy = 0
and withlist didn't complain, but unsub confirmations are still sent out
:( . I also tried setting ALLOW_OPEN_UNSUBSCRIBE in mm_cfg.py, but no
luck. I find it strange that Mailman is so insistent on confirmations
(CommuniGate has a simple checkbox for confirmations on or off), but
if I can find a workaround, my user will be very pleased.
P.S. I have been able to use the normal subscribe/confirm scheme
from my phone, so it is probably a function of your carrier or your
phone. One quirk I've noticed in my lists is that if I use
VERP_CONFIRMATIONS=Yes, the reply address gets truncated in the reply
message, and I have to manually add the last few characters of the
domain name. In my case that is presumably a phone bug, since it is
picking up fewer reply address characters than it allows me to enter
and successfully send.
We're not using VERP, but thanks for that.
Scot
On May 9, 2005, at 04:47, Scot Hacker wrote:
Jim, this is cool to know, thanks. So is there a corresponding
option for unsubscribe confirmations? I tried it withm.unsubscribe_policy = 0
and withlist didn't complain, but unsub confirmations are still
sent out :( . I also tried setting ALLOW_OPEN_UNSUBSCRIBE in mm_cfg.py, but no luck.
There is an unsubscribe_policy variable for each list (set to the
value of DEFAULT_UNSUBSCRIBE_POLICY in your mm_cfg.py/Defaults.py),
but as the comments in Defaults.py explain it only controls whether
or not administrative approval is necessary to allow the
unsubscription to be honored.
There is no setting called ALLOW_OPEN_UNSUBSCRIBE.
If you send your member password when you send the unsubscribe
command, there should be no confirmation step.
unsubscribe mypasswd
(You could modify Mailman/Commands/cmd_unsubscribe.py if you wanted
to allow unverified unsubscriptions from your list by mail.)
-- Jim Tittsler http://www.OnJapan.net/ GPG: 0x01159DB6 Python Starship http://Starship.Python.net/crew/jwt/ Mailman IRC irc://irc.freenode.net/#mailman
At 10:45 PM -0700 2005-05-04, Scot Hacker wrote:
Then we just run this shell script once per hour, and get
around the confirmations problem with cell phones (though I'm surprised -- aren't there any other mailman lists out there with people subbing their cell phones? you'd think there would be a better way...)
If people are subscribing from their cell phones, then this means
they have a way to read and write e-mail messages. That's all the capability you need, in order to have a confirmation work. They just respond to the confirmation request by e-mail, and you're done.
-- Brad Knowles, <brad@stop.mail-abuse.org>
"Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety."
-- Benjamin Franklin (1706-1790), reply of the Pennsylvania
Assembly to the Governor, November 11, 1755
SAGE member since 1995. See <http://www.sage.org/> for more info.
Brad Knowles wrote:
If people are subscribing from their cell phones, then this means
they have a way to read and write e-mail messages. That's all the capability you need, in order to have a confirmation work. They just respond to the confirmation request by e-mail, and you're done.
You'd think so, wouldn't you? I can have an email conversation all day long between my cell phone and my mail client, but try to subscribe a cell phone to a mailing list and it's a different story. The confirmation message is received, you reply to it from the phone, and off it goes into a black hole, the phone never actually subscribed. Eventually the list admin gets the failure report I pasted into the original message of this thread.
Some phones are able to subscribe to mailing lists, but most are not. I
really have no idea what mechanisms cause them to fail. This is not
specific to Mailman either - we had the same problem when we were
running on CommuniGate Pro. I suspect that the problem could be at the
service provider level. Most people get close to zero spam on our
phones, and I think it's because cell providers do some extremely
aggressive blocking. But once subscribed, the phones can receive these
daily haiku digests just fine (he currently has about 3000 cell
subscribers, but most of them had to be manually added or approved).
Scot
--On May 5, 2005 09:32:08 -0700 Scot Hacker <shacker@birdhouse.org> wrote:
Some phones are able to subscribe to mailing lists, but most are not. I really have no idea what mechanisms cause them to fail.
My guess would be that maybe the phone is sending the reply to the wrong email address.
-- Ian Eiloart Servers Team Sussex University ITS
Ian Eiloart wrote:
--On May 5, 2005 09:32:08 -0700 Scot Hacker <shacker@birdhouse.org> wrote:
Some phones are able to subscribe to mailing lists, but most are not. I really have no idea what mechanisms cause them to fail.
My guess would be that maybe the phone is sending the reply to the wrong email address.
Hmm... good point. Perhaps replying to the From: address rather than the Return-Path, or vice versa. Would there be any way to fix this in mailman, or is it just a question of broken email implementations in phones?
Scot
In a flurry of recycled electrons, Scot Hacker wrote:
Hmm... good point. Perhaps replying to the From: address rather than the Return-Path, or vice versa. Would there be any way to fix this in mailman, or is it just a question of broken email implementations in phones?
Most likely, it's a broken MTA at the phone service provider. If any message is actually reaching mailman, it might be possible to do something with it, but I suspect that starts to become major hackery.
I've found that many cell phone and pager companies have -no-idea- what email even is or how it works. They often don't understand much basic concepts as spam filtering, or even de-mimeing a message before sending it to the non-mime-aware pager (240 char of message space, first hundred are a mime header..). Things have gotten better in the last few years, though.
Of course, I've been wrong before.
z!
Is there a configurable open unsubscribe option similar to:
ALLOW_OPEN_SUBSCRIBE = Yes in mm_cfg.py? Doesn't seem to be one, but I'm hoping there's a way to allow this. If not, we can always use a web form that talks to the CLI remove_members tool.
Thanks, Scot
participants (5)
-
Brad Knowles
-
cpz@tuunq.com
-
Ian Eiloart
-
Jim Tittsler
-
Scot Hacker