IntegrityError: duplicate key violates unique constraint "message_pkey"

[bewst]

File "/usr/local/lib/python2.4/site-packages/Trac-0.11dev_r5790-py2.4.egg/trac/web/main.py", line 436, in dispatch_request
  dispatcher.dispatch(req)
File "/usr/local/lib/python2.4/site-packages/Trac-0.11dev_r5790-py2.4.egg/trac/web/main.py", line 221, in dispatch
  resp = chosen_handler.process_request(req)
File "build/bdist.freebsd-6.1-RELEASE-i386/egg/tracdiscussion/core.py", line 74, in process_requestFile "build/bdist.freebsd-6.1-RELEASE-i386/egg/tracdiscussion/api.py", line 51, in process_discussionFile "build/bdist.freebsd-6.1-RELEASE-i386/egg/tracdiscussion/api.py", line 593, in _do_actionFile "build/bdist.freebsd-6.1-RELEASE-i386/egg/tracdiscussion/api.py", line 972, in add_messageFile "/usr/local/lib/python2.4/site-packages/Trac-0.11dev_r5790-py2.4.egg/trac/db/util.py", line 70, in execute
  return c.execute(sql_escape_percent(sql), args)
File "/usr/local/lib/python2.4/site-packages/Trac-0.11dev_r5790-py2.4.egg/trac/db/util.py", line 70, in execute
  return c.execute(sql_escape_percent(sql), args)

[bewst]

This was the first thing I tried with the 0.11 port -- just submitting a quoted reply to an existing message.

[bewst]

Umm, looks like I gave you the wrong backtrace somehow. Please see the attachment.

[bewst]

Backtrace

[Radek Bartoň]

I didn't evoke this error with PosgreSQL. Could you please checkout lastest DiscussionPlugin revision with debug outputs and post me here debug log and info about your system (Trac version, webserver, etc. See BugReporting for more details) Thank you.

[bewst]

BZipped log file

[bewst]

That version is much more broken that the previous one.

DiscussionPlugin: ​http://trac-hacks.org/svn/discussionplugin/0.11 (r2358) Trac: 0.11-dev (r5790) Server: lighttpd + fastcgi proxied through apache 2 Database: postgresql 8.2 OS: FreeBSD

[Radek Bartoň]

OK, I was checking if there is needed ending comas in one-item tuples when passing arguments to SQL queries. Apparently they are on some systems (nor on mine). Never mind, please try it with r2374 aggain and post me relevant part of debug log. It is not needed to post 16.5 MB log when relevant are last 2 kB.

[Radek Bartoň]

Still experiencing this?