You can subscribe to this list here.
2009 |
Jan
(7) |
Feb
(1) |
Mar
(1) |
Apr
(16) |
May
(6) |
Jun
(4) |
Jul
(2) |
Aug
(9) |
Sep
|
Oct
|
Nov
|
Dec
|
---|---|---|---|---|---|---|---|---|---|---|---|---|
2010 |
Jan
(1) |
Feb
(1) |
Mar
(12) |
Apr
(10) |
May
(3) |
Jun
|
Jul
|
Aug
|
Sep
(2) |
Oct
(1) |
Nov
|
Dec
(1) |
2011 |
Jan
|
Feb
(22) |
Mar
(26) |
Apr
(26) |
May
(5) |
Jun
(4) |
Jul
(1) |
Aug
(1) |
Sep
|
Oct
(2) |
Nov
(9) |
Dec
(1) |
2012 |
Jan
(2) |
Feb
(1) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2013 |
Jan
|
Feb
|
Mar
(3) |
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
(1) |
Sep
|
Oct
|
Nov
|
Dec
|
2014 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(4) |
Sep
|
Oct
(2) |
Nov
|
Dec
|
2015 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2016 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(3) |
Dec
(4) |
2017 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2019 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2022 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
(7) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2023 |
Jan
(1) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Ron S. <ron...@ya...> - 2014-08-10 00:55:14
|
I am testing TMDA offered by a e-mail provider. My interface to it is via the TMDA cgi. I have created a 'keyword' address and tested it, it works. But how to I revoke it? I can't find anything in the TMDA cgi screens which permit me to do so. Ron |
From: Michael A. <mi...@ze...> - 2013-08-01 06:55:54
|
Hi folks, After upgrading my Gentoo/Linux box, tmda-pending started throwing "ValueError: too many values to unpack". I don't know what change brought it on (my Python version is still at 2.7.3). Anyway, here's the patch that fixed it for me: --- /home/mike/tmp/tmda-1.1.12/TMDA/Pending.py 2007-01-17 16:49:44.000000000 -0500 +++ /usr/lib64/python2.7/site-packages/TMDA/Pending.py 2013-08-01 01:44:50.000000000 -0400 @@ -371,7 +375,7 @@ if Defaults.PENDING_RELEASE_APPEND: Util.append_to_file(self.append_address, Defaults.PENDING_RELEASE_APPEND) - timestamp, pid = self.msgid.split('.') + timestamp, pid, ignore = self.msgid.split('.', 2) # Remove Return-Path: to avoid duplicates. del self.msgobj['return-path'] # Remove X-TMDA-Recipient: @@ -469,7 +473,7 @@ if not self.confirm_accept_address: if self.recipient: import Cookie - (timestamp, pid) = self.msgid.split('.') + (timestamp, pid, ignore) = self.msgid.split('.', 2) self.confirm_accept_address = Cookie.make_confirm_address( self.recipient, timestamp, pid, 'accept') -- Michael Allan Toronto, +1 416-699-9528 http://zelea.com/ |
From: Bradley H. <bh...@sh...> - 2013-04-29 02:24:58
|
The company I work for requests to have every email (including inter departmental) from employees be reviewed and approved by their manager. I'm wondering if the confirmation workflow of TMDA can be tailored to this type of confirmation. For instance, if email from Jo...@ex..., then place item in pending queue, and send a confirmation email to Ma...@ex..., and don't place in whitelist. I've looked at the readme with examples, and it seems there's no way to set a rule to request confirmation from a different email address. I was hoping there might be some undocumented syntax that I have not found, or that perhaps you know of an application that may fulfill this function. We are testing Zimbra 8.0.3 on a SLES 11.2 system. We are not locked into Zimbra, and could fall back to basic postfix if this gives up the functionality required. Brad |
From: Kevin G. <kev...@om...> - 2013-03-26 06:06:57
|
On 03/24/2013 11:14 AM, Kevin Goodsell wrote: > > My best guess is that the default Python version for the web hosting > service changed to 2.6, and from the look of it our MoinMoin version has > a 2.6 incompatibility bug. I'll try to get it fixed up, but I'm not sure > how long it's going to take. I put in place a quick patch for the specific problem causing this error and it seems to be working, but there's no way to know what else could be broken. I'm going to try to update MoinMoin, but not until I do some test runs on a non-production system. -Kevin |
From: Kevin G. <kev...@om...> - 2013-03-24 18:30:22
|
On 03/12/2013 09:42 AM, Corey Halpin wrote: > It looks like SF has fiddled with something and broken the wiki. > > http://tmda.sourceforge.net/cgi-bin/moin.cgi gives me the attached > tracebacks. I've tried a number of browsers, so I don't *think* it's anything > to do with my particular configuration. > > It'd be nice to have the wiki back as a reference. I don't know much about > MoinMoin, and I don't have a lot of time to offer, but if there's some way I > can help to resurrect this, I gladly will. > > Thank you. > My best guess is that the default Python version for the web hosting service changed to 2.6, and from the look of it our MoinMoin version has a 2.6 incompatibility bug. I'll try to get it fixed up, but I'm not sure how long it's going to take. Thanks for reporting the problem. -Kevin |
From: Corey H. <ch...@cs...> - 2013-03-12 16:42:28
|
It looks like SF has fiddled with something and broken the wiki. http://tmda.sourceforge.net/cgi-bin/moin.cgi gives me the attached tracebacks. I've tried a number of browsers, so I don't *think* it's anything to do with my particular configuration. It'd be nice to have the wiki back as a reference. I don't know much about MoinMoin, and I don't have a lot of time to offer, but if there's some way I can help to resurrect this, I gladly will. Thank you. ~crh Tracebacks: <!--: spam Content-Type: text/html <body><font style="color: white; font-size: 1px"> --> <html><head> <meta http-equiv="content-type" content="text/html; charset=UTF-8"></head><body><font style="color: white; font-size: 1px"> --> --> </font> <pre>Traceback (most recent call last): File "../moin/lib/python2.4/site-packages/MoinMoin/request.py", line 1207, in run handler(self.page.page_name, self) File "../moin/lib/python2.4/site-packages/MoinMoin/wikiaction.py", line 453, in do_show if not request.user.may.read(pagename): File "../moin/lib/python2.4/site-packages/MoinMoin/security.py", line 52, in <lambda> return lambda pagename, Page=Page, request=request, attr=attr: Page(request, pagename).getACL(request).may(request, self.name, attr) File "../moin/lib/python2.4/site-packages/MoinMoin/wikiacl.py", line 184, in may is_group_member = request.dicts.has_member File "../moin/lib/python2.4/site-packages/MoinMoin/request.py", line 298, in getDicts dicts.scandicts() File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 374, in scandicts self.addgroup(request, pagename) File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 292, in addgroup grp = Group(request, groupname) File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 48, in __init__ self.initRegex() File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 57, in initRegex cls.regex = re.compile(cls.regex, re.MULTILINE | re.UNICODE) File "/usr/lib64/python2.6/re.py", line 190, in compile return _compile(pattern, flags) File "/usr/lib64/python2.6/re.py", line 238, in _compile raise ValueError('Cannot process flags argument with a compiled pattern') ValueError: Cannot process flags argument with a compiled pattern Additionally cgitb raised this exception: Traceback (most recent call last): File "../moin/lib/python2.4/site-packages/MoinMoin/failure.py", line 146, in handle handler.handle() File "../moin/lib/python2.4/site-packages/MoinMoin/support/cgitb.py", line 566, in handle doc = view.format(formatter, self.context) File "../moin/lib/python2.4/site-packages/MoinMoin/support/cgitb.py", line 343, in format return formatter.section(self.formatContent(), {'class': 'cgitb'}) File "../moin/lib/python2.4/site-packages/MoinMoin/failure.py", line 38, in formatContent self.formatDebugInfo(), File "../moin/lib/python2.4/site-packages/MoinMoin/failure.py", line 83, in formatDebugInfo self.formatTraceback(), File "../moin/lib/python2.4/site-packages/MoinMoin/failure.py", line 96, in formatTraceback return self.formatAllTracebacks(self.formatOneTraceback) File "../moin/lib/python2.4/site-packages/MoinMoin/failure.py", line 108, in formatAllTracebacks tracebacks.append(formatFuction((type, value, tb))) File "../moin/lib/python2.4/site-packages/MoinMoin/support/cgitb.py", line 423, in formatOneTraceback self.formatter.orderedList(self.tracebackFrames(info), File "../moin/lib/python2.4/site-packages/MoinMoin/support/cgitb.py", line 434, in tracebackFrames frames.append(frame.format(self.formatter)) File "../moin/lib/python2.4/site-packages/MoinMoin/support/cgitb.py", line 197, in format vars, highlight = self.scan() File "../moin/lib/python2.4/site-packages/MoinMoin/support/cgitb.py", line 286, in scan vars = self.scanVariables(reader) File "../moin/lib/python2.4/site-packages/MoinMoin/support/cgitb.py", line 298, in scanVariables value = getattr(parent, token, __UNDEF__) File "../moin/lib/python2.4/site-packages/MoinMoin/request.py", line 298, in getDicts dicts.scandicts() File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 374, in scandicts self.addgroup(request, pagename) File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 292, in addgroup grp = Group(request, groupname) File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 48, in __init__ self.initRegex() File "../moin/lib/python2.4/site-packages/MoinMoin/wikidicts.py", line 57, in initRegex cls.regex = re.compile(cls.regex, re.MULTILINE | re.UNICODE) File "/usr/lib64/python2.6/re.py", line 190, in compile return _compile(pattern, flags) File "/usr/lib64/python2.6/re.py", line 238, in _compile raise ValueError('Cannot process flags argument with a compiled pattern') ValueError: Cannot process flags argument with a compiled pattern </pre> </body></html> |
From: Sabahattin G. <ma...@sa...> - 2012-02-08 10:24:38
|
Hi all, Recently I noticed that if my emails had multiple recipients they would be injected by tmda-inject into /usr/sbin/sendmail as multiple transactions, one per recipient. Often, this would result in multiple parallel deliveries (by Postfix) to the same host, for multiple recipients at that host. Now, I appreciate that in these times of increased bandwidth this isn't too big of a deal, but I'd really rather use SMTP correctly and deliver just one copy. Most hosts today now do pipelining, so there's really no good reason not to. Is there anything I can do meantime, or is a fix possible? Was there some reason for this particular choice of implementation? Cheers, Sabahattin |
From: Sabahattin G. <ma...@sa...> - 2012-01-25 04:12:55
|
As with 8BITMIME, I'm just whipping up a Sendmail wrapper TMDA can call to turn on DSN support. From this I gather that the feature needed is the ability to pass in arguments to the sendmail program, or the ability to pass parameters into the MAIL and RCPT commands of the downstream submission server, depending on which is selected by the user. Tip: if you're on Sendmail (the real thing), set O RRTImpliesDSN=T and then have TMDA add a "Return-Receipt-To:" header. Sendmail will helpfully turn your request into a DSN success, discard the header value and strip it from the next hop output. Tip: some MTAs (Mercury/32 on Windows, EG) are still honouring Return-Receipt-To even when they themselves do not support DSN. In these enlightened times of greylisting and other spam-vapourising nonsense, I love DSNs. Cheers, Sabahattin |
From: Sabahattin G. <ma...@sa...> - 2012-01-02 15:54:21
|
I just upgraded to Postfix from XMail. With TMDA there is an immediate noticeable difference: my challenges go out with the address in $RECIPIENT in the From: field. $RECIPIENT is my account name after expansion from my alias 'mail', as used in this message. I realise this isn't a "Big deal", and it isn't even worth writing a wrapper shell script to fix it, but is there any way to fix it in TMDA itself? Cheers, Sabahattin |
From: Sabahattin G. <ma...@sa...> - 2011-12-26 19:46:32
|
Hi, I have an (arguably broken) email client which will not do duplicate detection in the absence of Message-Id fields. While I can get Postfix to insert that field along with From/Sender/Date, I'd rather not since it breaks the purity of mail when relayed or for delivery to non-TMDA destinations, and Message-Id can't be enabled without enabling all the other fields. Is there a server-side equivalent to the TMDAINJECT variable, or a way to implement it? Cheers, Sabahattin |
From: Brent S. <bsp...@Xs...> - 2011-11-19 00:35:55
|
I see that the incoming filter (http://tmda.sourceforge.net/cgi-bin/moin.cgi/FilterSpecification#addlfeatures) allows a deliver=&<email_address> for forwarding messages. This works very well. However, the local, original recipient user does not get the email as well. Is there a way to make this happen? Can an incoming filter have multiple actions (e.g. ok AND deliver=&<email_address>)? I've tried to get this to work, but I either get an error or just the first action is performed. Brent |
From: Brent S. <bsp...@Xs...> - 2011-11-18 22:24:52
|
So, I thought I had made some progress on this by adding a rule to my .procmailrc to check for the existence of a "X-TMDA-Action" header in the message as an indicator that the message was validated and should be forwarded. However, this header is not availble when the message passes through .procmailrc as TMDA seems to directly move the validated message to the appropriate /var/spool/mail/<user_name> directory. Is there a way to get TMDA to send the validated message through the .procmailrc routine via sendmail again? Brent |
From: Brent S. <bsp...@Xs...> - 2011-11-18 21:36:43
|
Nevermind. I found ACTION_HEADER_INCOMING. Brent |
From: Brent S. <bsp...@Xs...> - 2011-11-18 20:59:37
|
When an email is accepted by TMDA for delivery, the "Actn:" line in the incoming log file describes why the message was accepted. How would I add this information as a header in the email. For example, if the incoming log message is "Actn: OK (from-file -autocdb ~/.tmda/lists/whitelist ok)", I would want to add a header to that message such as "X-TMDA-Actn: whitelist ok". Is there a way to do this? Brent |
From: Brent S. <bsp...@Xs...> - 2011-11-18 17:11:31
|
I have a working TMDA setup. Currently, as expected, messages that are validated arrive in my Inbox. However, I would like to be able to also forward these messages to another email address. How do I configure this? Brent Spencer XsiteDesigns |
From: Kevin G. <kev...@gm...> - 2011-11-02 02:02:57
|
On Tue, Nov 1, 2011 at 8:52 AM, Anthony Ercolano <ton...@er...> wrote: > > This first suggestion was the best. During the upgrade I lost a portion of the APACHE2_MODULES specification in the /etc/make.conf. When there were added back it I got the login page back. > > However using the "updated" tmda.cgi I would always get an invalid password. when I switched back to the released tmda.cgi the correct password was accepted. > > Any thoughts on what I should pursue? There's some debugging info that gets put into the login page source on a login failure. That might help pinpoint the problem. Also, there's a basic server for testing in the simple-server/ directory, and that might make it easier to see what's going on. You can just run it from a terminal. -Kevin |
From: Anthony E. <ton...@er...> - 2011-11-01 15:52:38
|
On Oct 31, 2011, at 7:35 PM, Kevin Goodsell wrote: > On Mon, Oct 31, 2011 at 3:00 PM, Anthony Ercolano > <ton...@er...> wrote: >> Both the released and "updated" version Kevin "published". >> Basically when attempting to execute the cgi I get the following dump from >> my browser: >> > > Looks like a web server configuration problem to me. Instead of > executing the CGI program, it's sending the program to the browser. This first suggestion was the best. During the upgrade I lost a portion of the APACHE2_MODULES specification in the /etc/make.conf. When there were added back it I got the login page back. However using the "updated" tmda.cgi I would always get an invalid password. when I switched back to the released tmda.cgi the correct password was accepted. Any thoughts on what I should pursue? My configure.ini file for the password invalid password tmda.cgi looks like: [environment-variables] TMDA_CGI_MODE = system-wide TMDA_SESSION_ODDS = 0.01 TMDA_VUSER = vpopmail TMDA_SESSION_EXP = 300 TMDA_SESSION_PREFIX = /tmp/TMDASession. TMDA_AUTH_ARG = imaps://mail.ercolano.net:993 TMDA_VLOOKUP = vpopmail1 /home/vpopmail/bin/vuserinfo ~ TMDA_CGI_DISP_DIR = ../display TMDA_AUTH_TYPE = remote TMDA_BASE_DIR = /usr/local/src/tmda/current TMDARC = None [configure-parameters] auth_type = remote config_file = None session_prefix = /tmp/TMDASession. binary_target = /var/www/localhost/cgi-bin/tmda.cgi python_interpreter = /usr/bin/python2.6 cleanup_odds = 0.01 tmda_path = /usr/local/src/tmda/current auth_uri = imaps://mail.ercolano.net:993 virtual_lookup = vpopmail1 /home/vpopmail/bin/vuserinfo ~ mode = system-wide auth_file = None virtual_user = vpopmail auth_program = /usr/sbin/checkpassword session_expire = 300 display_dir = ../display compiler = gcc The working tmda.cgi configure.ini looks like: (dp0 S'AuthArg' p1 S'imaps://mail.ercolano.net:993' p2 sS'Perm' p3 S'6711' p4 sS'Config' p5 S'None' p6 sS'SessionPrefix' p7 S'/tmp/TMDASession.' p8 sS'InstPath' p9 S'/usr/local/src/tmda-cgi/current' p10 sS'Python' p11 S'/usr/bin/python2.5' p12 sS'Base' p13 S'/usr/local/src/tmda/current' p14 sS'AuthURI' p15 g2 sS'Path' p16 S'/usr/local/src/tmda-cgi/tmda-cgi-0.16.4/' p17 sS'AuthProg' p18 S'/usr/sbin/checkpassword' p19 sS'AuthType' p20 S'remote' p21 sS'Target' p22 S'/var/www/localhost/cgi-bin/tmda.cgi' p23 sS'DispDir' p24 S'../display/' p25 sS'CC' p26 S'gcc' p27 sS'VLookup' p28 S'vpopmail1 /var/vpopmail/bin/vuserinfo ~' p29 sS'Mode' p30 S'system-wide' p31 sS'AuthFile' p32 S'None' p33 sS'SessionOdds' p34 S'0.01' p35 sS'VUser' p36 S'vpopmail' p37 sS'SessionExp' p38 S'300' p39 |
From: Anthony E. <ton...@er...> - 2011-11-01 02:54:28
|
> > Yeah, I would think that would ensure that it would *not* work. CGI > programs need to be in the CGI directory so the web server knows to > execute them, not return them as files. > I should have been more clear. I placed the directory that contained the .py files below the document root in an offshoot directory. The tmda.cgi file was still in its proper cgi-bin directory. > -Kevin Tony |
From: Kevin G. <kev...@gm...> - 2011-11-01 02:35:47
|
On Mon, Oct 31, 2011 at 3:00 PM, Anthony Ercolano <ton...@er...> wrote: > Both the released and "updated" version Kevin "published". > Basically when attempting to execute the cgi I get the following dump from > my browser: > Looks like a web server configuration problem to me. Instead of executing the CGI program, it's sending the program to the browser. > > The thing that I find most interesting is the text Cannot change to > directory: %s….. > This appears to be a failure to do the chdir in the first few executable > lines of code in tmda-cgi.c I don't think that's relevant. You are seeing static text from inside the executable. You would find every error message the program can generate there. It doesn't reflect the error that's actually occurring. > In the above example I actually moved the "install" directory of the cgi > below the document root in case this was a permission problem. This didn't > help. Yeah, I would think that would ensure that it would *not* work. CGI programs need to be in the CGI directory so the web server knows to execute them, not return them as files. -Kevin |
From: Anthony E. <ton...@er...> - 2011-10-31 22:18:18
|
Both the released and "updated" version Kevin "published". Basically when attempting to execute the cgi I get the following dump from my browser: ELFàƒ4T4 (44€4€ TTT€€xxŸŸ Ÿ ŸÐÐhhh QåtdRåtdŸŸôô€e(/lib/ld-linux.so.2GNU KãÀC H)/6<˜†__gmon_start__libc.so.6_IO_stdin_usedexeclprintfchdirputenvopen__libc_start_mainGLIBC_2.0ii ZðŸ U‰åƒìè±è(èóÉÃÿ5øŸÿ%üŸÿ% héàÿÿÿÿ% héÐÿÿÿÿ% héÀÿÿÿÿ% hé°ÿÿÿÿ% h é ÿÿÿÿ% h(éÿÿÿÿ% h0é€ÿÿÿ1í^‰áƒäðPTRhà…hð…QVh¨„èÿÿÿôU‰åSƒìè[Ãä‹“üÿÿÿ…ÒtèZÿÿÿX[ÉÃU‰åSƒì€=$ u5»Ÿ¡( ëŸÁûK9Øs@£( ÿ…Ÿ¡( 9ØrêÆ$ X[]ô&U‰åƒì‹Ÿ…Òt¸…Àt ƒìhŸÿЃÄÉÃL$ƒäðÿqüU‰åQƒìƒìhœ†è ÿÿÿƒÄƒìh¨†èúþÿÿƒÄƒìhÀ†èêþÿÿƒÄƒìhì†èÚþÿÿƒÄƒìh‡èÊþÿÿƒÄƒìh$‡èºþÿÿƒÄƒìhN‡èªþÿÿƒÄƒìhd‡èšþÿÿƒÄƒìhœ‡èŠþÿÿƒÄƒìh‡èzþÿÿƒÄƒìhׇèjþÿÿƒÄƒìjhî‡èøýÿÿƒÄ‰Eôƒìhø‡è5þÿÿƒÄ…Àu jhˆh'ˆh'ˆèøýÿÿƒÄ¸ë¸<ˆƒìhø‡PèëýÿÿƒÄ¸‹MüÉaüÃU‰å]Ãt&¼'U‰åWVSƒìè[Ãöè>ýÿÿƒÿÿÿ“ÿÿÿ)ЉÇÁÿt%1ö´&‹MP‹U‹EQRPÿ”³ÿÿÿFƒÄ9þräeô[^_]ÃU‰åSƒì¡Ÿƒøÿt»ŸfƒëÿЋƒøÿuôX[]ÃU‰åSƒìè[Ãpè ýÿÿY[ÉÃTMDARC=NoneTMDA_AUTH_TYPE=remoteTMDA_AUTH_ARG=imaps://mail.ercolano.net:993TMDA_CGI_MODE=system-wideTMDA_CGI_DISP_DIR=../displayTMDA_BASE_DIR=/usr/local/src/tmda/currentTMDA_VUSER=vpopmailTMDA_VLOOKUP=vpopmail1 /home/vpopmail/bin/vuserinfo ~TMDA_SESSION_PREFIX=/tmp/TMDASession.TMDA_SESSION_EXP=300TMDA_SESSION_ODDS=0.01/dev/null/var/www/localhost/htdocs/tmda-cgitmda-cgi.py/usr/bin/python2.6Content-type: text/html Cannot change to directory: %sÿÿÿÿÿÿÿÿHƒ x†ˆõþÿoÀp‚à dôŸ8ƒƒþÿÿoè‚ÿÿÿoðÿÿoÔ‚ Ÿvƒ†ƒ–ƒ¦ƒ¶ƒÆƒÖƒGCC: (Gentoo 4.3.4 p1.2, pie-10.1.5) 4.3.4GCC: (Gentoo 4.4.5 p1.2, pie-0.4.5) 4.4.5.symtab.strtab.shstrtab.interp.note.ABI-tag.gnu.hash.dynsym.dynstr.gnu.version.gnu.version_r.rel.dyn.rel.plt.init.text.fini.rodata.eh_frame.ctors.dtors.jcr.dynamic.got.got.plt.data.bss.commentTT#hh 5ˆˆ81öÿÿoÀÀ ;ààCp‚pdKÿÿÿoÔ‚ÔXþÿÿoè‚è g ƒp ƒ8yHƒHt`ƒ`€àƒà˜…x†x‹”†”à“tˆtŸ¤Ÿ«Ÿ° Ÿ йðŸð¾ôŸô(Ç Í$ $Ò0$UyÛÜ` <}ThˆÀàp‚Ԃ肃 ƒ Hƒ`ƒàƒ x†”†tˆŸŸŸ ŸðŸôŸ $ ñÿôŸ"Ÿ3ŸF ŸO Zjà… zàƒ ¤”†«x†±Îߘ†î û Ÿð…\ %7HZ$ ñÿf, ñÿk$ ñÿr¨„- wHƒtmda-cgi.c_GLOBAL_OFFSET_TABLE___init_array_end__init_array_start_DYNAMICdata_startopen@@GLIBC_2.0__libc_csu_fini_start__gmon_start___Jv_RegisterClasses_fp_hw_fini__libc_start_main@@GLIBC_2.0execl@@GLIBC_2.0_IO_stdin_used__data_start__dso_handle__DTOR_END____libc_csu_initprintf@@GLIBC_2.0chdir@@GLIBC_2.0putenv@@GLIBC_2.0__bss_start_end_edatamain_init The thing that I find most interesting is the text Cannot change to directory: %s….. This appears to be a failure to do the chdir in the first few executable lines of code in tmda-cgi.c In the above example I actually moved the "install" directory of the cgi below the document root in case this was a permission problem. This didn't help. Any thoughts on how I should proceed? - Thanks in advance Tony |
From: Anthony E. <ton...@er...> - 2011-10-31 22:07:32
|
Both the released and "updated" version Kevin "published". Basically when attempting to execute the cgi I get the following dump from my browser: ELFàƒ4T4 (44€4€ TTT€€xxŸŸ Ÿ ŸÐÐhhh QåtdRåtdŸŸôô€e(/lib/ld-linux.so.2GNU KãÀC H)/6<˜†__gmon_start__libc.so.6_IO_stdin_usedexeclprintfchdirputenvopen__libc_start_mainGLIBC_2.0ii ZðŸ U‰åƒìè±è(èóÉÃÿ5øŸÿ%üŸÿ% héàÿÿÿÿ% héÐÿÿÿÿ% héÀÿÿÿÿ% hé°ÿÿÿÿ% h é ÿÿÿÿ% h(éÿÿÿÿ% h0é€ÿÿÿ1í^‰áƒäðPTRhà…hð…QVh¨„èÿÿÿôU‰åSƒìè[Ãä‹“üÿÿÿ…ÒtèZÿÿÿX[ÉÃU‰åSƒì€=$ u5»Ÿ¡( ëŸÁûK9Øs@£( ÿ…Ÿ¡( 9ØrêÆ$ X[]ô&U‰åƒì‹Ÿ…Òt¸…Àt ƒìhŸÿЃÄÉÃL$ƒäðÿqüU‰åQƒìƒìhœ†è ÿÿÿƒÄƒìh¨†èúþÿÿƒÄƒìhÀ†èêþÿÿƒÄƒìhì†èÚþÿÿƒÄƒìh‡èÊþÿÿƒÄƒìh$‡èºþÿÿƒÄƒìhN‡èªþÿÿƒÄƒìhd‡èšþÿÿƒÄƒìhœ‡èŠþÿÿƒÄƒìh‡èzþÿÿƒÄƒìhׇèjþÿÿƒÄƒìjhî‡èøýÿÿƒÄ‰Eôƒìhø‡è5þÿÿƒÄ…Àu jhˆh'ˆh'ˆèøýÿÿƒÄ¸ë¸<ˆƒìhø‡PèëýÿÿƒÄ¸‹MüÉaüÃU‰å]Ãt&¼'U‰åWVSƒìè[Ãöè>ýÿÿƒÿÿÿ“ÿÿÿ)ЉÇÁÿt%1ö´&‹MP‹U‹EQRPÿ”³ÿÿÿFƒÄ9þräeô[^_]ÃU‰åSƒì¡Ÿƒøÿt»ŸfƒëÿЋƒøÿuôX[]ÃU‰åSƒìè[Ãpè ýÿÿY[ÉÃTMDARC=NoneTMDA_AUTH_TYPE=remoteTMDA_AUTH_ARG=imaps://mail.ercolano.net:993TMDA_CGI_MODE=system-wideTMDA_CGI_DISP_DIR=../displayTMDA_BASE_DIR=/usr/local/src/tmda/currentTMDA_VUSER=vpopmailTMDA_VLOOKUP=vpopmail1 /home/vpopmail/bin/vuserinfo ~TMDA_SESSION_PREFIX=/tmp/TMDASession.TMDA_SESSION_EXP=300TMDA_SESSION_ODDS=0.01/dev/null/var/www/localhost/htdocs/tmda-cgitmda-cgi.py/usr/bin/python2.6Content-type: text/html Cannot change to directory: %sÿÿÿÿÿÿÿÿHƒ x†ˆõþÿoÀp‚à dôŸ8ƒƒþÿÿoè‚ÿÿÿoðÿÿoÔ‚ Ÿvƒ†ƒ–ƒ¦ƒ¶ƒÆƒÖƒGCC: (Gentoo 4.3.4 p1.2, pie-10.1.5) 4.3.4GCC: (Gentoo 4.4.5 p1.2, pie-0.4.5) 4.4.5.symtab.strtab.shstrtab.interp.note.ABI-tag.gnu.hash.dynsym.dynstr.gnu.version.gnu.version_r.rel.dyn.rel.plt.init.text.fini.rodata.eh_frame.ctors.dtors.jcr.dynamic.got.got.plt.data.bss.commentTT#hh 5ˆˆ81öÿÿoÀÀ ;ààCp‚pdKÿÿÿoÔ‚ÔXþÿÿoè‚è g ƒp ƒ8yHƒHt`ƒ`€àƒà˜…x†x‹”†”à“tˆtŸ¤Ÿ«Ÿ° Ÿ йðŸð¾ôŸô(Ç Í$ $Ò0$UyÛÜ` <}ThˆÀàp‚Ԃ肃 ƒ Hƒ`ƒàƒ x†”†tˆŸŸŸ ŸðŸôŸ $ ñÿôŸ"Ÿ3ŸF ŸO Zjà… zàƒ ¤”†«x†±Îߘ†î û Ÿð…\ %7HZ$ ñÿf, ñÿk$ ñÿr¨„- wHƒtmda-cgi.c_GLOBAL_OFFSET_TABLE___init_array_end__init_array_start_DYNAMICdata_startopen@@GLIBC_2.0__libc_csu_fini_start__gmon_start___Jv_RegisterClasses_fp_hw_fini__libc_start_main@@GLIBC_2.0execl@@GLIBC_2.0_IO_stdin_used__data_start__dso_handle__DTOR_END____libc_csu_initprintf@@GLIBC_2.0chdir@@GLIBC_2.0putenv@@GLIBC_2.0__bss_start_end_edatamain_init The thing that I find most interesting is the text Cannot change to directory: %s….. This appears to be a failure to do the chdir in the first few executable lines of code in tmda-cgi.c In the above example I actually moved the "install" directory of the cgi below the document root in case this was a permission problem. This didn't help. Any thoughts on how I should proceed? - Thanks in advance Tony |
From: Dagger <dag...@gm...> - 2011-08-22 04:07:07
|
I got a bit fed up of dropping to a terminal to generate addresses when signing up for things in Firefox, so I wrote this a little while back: https://addons.mozilla.org/en-US/firefox/addon/tmda-address-gen/ Maybe it'll be useful to someone else on this list too. |
From: Brent S. <bsp...@Xs...> - 2011-07-20 00:17:22
|
I have a working sendmail+virtusertable+tmda configuration. However, I would like to be able to send confirmed emails to a group of recipients. For instance, if an email is sent to local user A and the sender positively responds to the challenge, I would like for not only user A but also so...@gm... to receive the email that was initially sent to user A. Any help is greatly appreciated. Brent |
From: Anthony V. E. <ton...@gm...> - 2011-06-03 07:21:36
|
Turns out googling for "connect: Operation now in progress (115)" (which was in the log) gets some good hits. Seems as though stunnel 4.35 has a bug. Downgrading to 4.33 caused everything to start working again. - Tony |
From: Kevin G. <kev...@gm...> - 2011-06-03 02:09:30
|
On Thu, Jun 2, 2011 at 1:20 PM, Anthony V. Ercolano <ton...@gm...> wrote: > > My /etc/tmda/stunnel/tmda-ofmipd-wrapper is: > #!/bin/sh > exec /usr/local/src/tmda/current/bin/tmda-ofmipd -d -f -1 > -R imaps://mail.ercolano.net \ > -S /usr/local/src/tmda/current/contrib/vpopmail-vdir.sh \ > 2>>/var/log/tmdaofmipd > > (I see no messages in /var/log/tmdaofmipd.) It seems quite strange that you would see nothing here. tmda-ofmipd with -d should output something. It may be because you are only capturing stderr. I'd suggest getting stdout also with something like 2>&1 >>/var/log/tmdaofmipd. If you really get no output from tmda-ofmipd, I'd say it probably isn't actually running. You might want to try running it in the foreground from a terminal to see what happens. > 2011.06.01 21:21:10 LOG3[32556:3083143808]: connect: Operation now in > progress (115) > 2011.06.01 21:21:10 LOG5[32556:3083143808]: Connection reset: 0 bytes sent > to SSL, 0 bytes sent to socket > 2011.06.01 21:21:10 LOG7[32556:3083143808]: Service stunnel finished (0 > left) > > > The next to last line seems like where things really break. Is there some > sort of protocol mismatch between stunnel and tmda-ofmipd? You might want to try connecting with openssl to see if there are any clues there: openssl s_client -connect address:port -crlf I'm not sure if you need the -starttls smtp option or not with your configuration. -Kevin |