T M D A |
|
||||||||||||||||||||||||||
|
TMDA FREQUENTLY ASKED QUESTIONS
You can allow senders, or groups of senders directly into your mailbox by adding lines to your incoming filter file. By default this is ~/.tmda/filters/incoming. For example, the following line will allow both anyone@domain.dom, and anyone@sub.domain.dom into your mailbox: If you have many individual addresses and/or expressions you'd like to whitelist, you may keep them in a seperate file and then have TMDA access it using the following line:from *@=domain.dom ok ~/.tmda/lists/whitelist would contain e-mail addresses and/or wildcard expressions, one per line. e.g,from-file ~/.tmda/lists/whitelist ok See the TMDA Filter Specification for details on filter file syntax, options, and more examples.king@grassland.com *@myisp.net *@cs.myuni.edu *@=mycompany.com bobby*@peru.com The collectaddys script in the contrib directory can be used to build an initial whitelist from your existing mail archives. NOTE: For incoming mail, tmda-filter compares the address in the Envelope-Sender, the "From:" header and the "Reply-To:" header when looking for a match. By default, TMDA doesn't automatically whitelist senders who confirm. If you want that behavior, set the CONFIRM_APPEND variable in your ~/.tmdarc to point to a file, something like: and add a rule to your incoming filter:CONFIRM_APPEND = os.path.expanduser("~/.tmda/lists/whitelist_confirmed") Then TMDA will the add the senders from successfully confirmed messages to the 'whitelist_confirmed' file. A possible variation on this includes setting CONFIRM_APPEND to your main whitelist file rather than a secondary file.from-file ~/.tmda/lists/whitelist_confirmed ok You do have a couple of options under Windows on the client side of TMDA. You can configure a qmail relay to re-write the address of outgoing messages with tmda-inject, so that those running Microsoft Windows clients can take full advantage of TMDA. See `README.RELAY' in the contrib/ directory or one user's configuration details for a site-wide TMDA install using smtpd-auth, vpopmail and qmailadmin. TMDA also works perfectly under cygwin with the sSMTP sendmail emulation program (which comes with cygwin). Just add the following line your tmdarc: You'll then be able to tag your outgoing mail with cygwin-compatible MUAs (mutt, XEmacs, etc.) just as under UNIX.SENDMAIL = "/usr/sbin/ssmtp.exe" Yes. QAdmin-TMDA is a TMDA configuration and administration tool which integrates into the QmailAdmin web managment system. For more information on QAdmin-TMDA's capabilities, see its original announcement. See Lou Hevly's TMDA/vpopmail tutorial TMDA comes with a utility called `tmda-pending' which can do all of this and more for you. You can run it by hand, or periodically from cron. Run ``tmda-pending -h'' to get a listing of available options and usage examples. tmda-pending should be run by the user account that owns the pending queue, not by root (unless root is running TMDA). NOTE: Use tmda-pending's delete and release options with discretion. If a sender tries to confirm a message which you have manually removed, they will get back an error message from TMDA about the missing message. Thus, it is best to only manually remove messages which have no (or little) chance of being confirmed by their sender such as mailing list messages. If you want to maintain a copy of pending messages to play with, you can set CONFIRM_CC in your .tmdarc. This variable takes an e-mail address that will receive a copy of any message that triggers a confirmation request. See your sample.tmdarc for an example. You can whitelist the ezmlm list using wildcard characters. Here is an example whitelist entry for the qmail mailing list:
This will allow messages to get through when initially interacting
with ezmlm to get subscribed, as well as after when list messages are
delivered.
You can either set bbdb-always-add-addresses to 'never, or use a BBDB hook to filter the addresses before adding them to the database. You need some special entries in the virtusertable (and a modern version of sendmail - at least 8.10.1 to get full functionality) to preserve the +detail portion of the address. For example, if you currently have the following in your virtusertable: you can change this to:bobby@peru.com bobby See the virtusertable section in your Sendmail's cf/README for the full details.bobby+*@peru.com bobby+%2 If you have the following in /etc/aliases: +detail is preserved for mail to johndoe+detail, but not for webmaster+detail.webmaster: johndoe Unfortunately, there is no way to preserve the +detail portion in /etc/aliases. You must use the virtusertable for this (see previous question). Support for Exchange Server is not planned primarily because I have no access to this software, and no interest in developing in a Microsoft environment. Also, TMDA is an opensource endeavor, not a profit oriented one, and therefore I'm also not interested in turning it into an "enterprise solution". I enjoy my personal freedom too much to consider this. In theory yes, but in practice this is not likely to happen. Most SPAM is unrepliable, so TMDA's confirmation requests are never delivered to them. They use non-valid return addresses as to not incur the cost of the tremendous number of bounces they generate. Using a valid return address to process all the bounces looking for confirmation messages to auto-reply to would defeat their economies of scale. It would also make them easy to block, track down and report, sue, etc. In short, trying to thwart TMDA in this manner would defeat the cost-effectiveness of the bulk-mailing process. Simple economics keep us safe. Additionally, the majority of the population is not yet using TMDA or a TMDA-like system to protect their mailbox, so we are simply not worth the extra effort. There are plenty of weaker, "less fit" targets to prey on, and so they will, passing over us in the process. But should these facts change, TMDA could modify its (currently very simple) challenge response to make it more difficult for a computer to auto-reply to. The level of difficulty could increase as much as is necessary for the sender to prove their humanity and legitimacy. Until someone teaches a computer program to read and understand the English language, I think we will have plenty of options here. In short, no. Python is an excellent choice for an application like TMDA. Python is a simple yet extremely powerful language with a reasonably fast interpreter and comprehensive standard library. It is also very portable and has several features that make it very attractive for Rapid Application Development. I also happen to enjoy programming in Python more than any other language. There are a few reasons that I can think of why you are asking this question:
Yes. See this message for how to do this; these instructions apply to any mailing list protected by TMDA. If you are using TMDA's CONFIRM_APPEND feature to implement an "auto-whitelist", you may have noticed that it is being filled with the 'dated' addresses of other TMDA users, which isn't very helpful. This can be disconcerting to the other party who must confirm each of their messages to you. The solution is to mix social practices with the technology. Don't rely entirely on CONFIRM_APPEND. When you start to interact with a new TMDA user, add a wildcard entry for them to your whitelist (e.g, jason*@mastaler.com ). This will work for
other TMDA-like schemes that use one-time addresses as well.
|