![](https://secure.gravatar.com/avatar/930c46db686975260aaad5b255d345ce.jpg?s=120&d=mm&r=g)
I've just upgraded my Mailman installation from a 2.0.x to a 2.1.2 on a Solaris 8 server using Python 2.2.2. And I think I have a problem with the OutgoingRunner but I can't put my finger on it. Here a the symtoms:
- After I start mailman, the mailmanctl never exits. I can still see it in the process table.
- The "mailmanctl stop" command stops everything but NOT the OutgoingRunner. I have to kill that one manually.
- If I send subscription requests (email or web), I see them pending in logs/subscribe. I see a new file get created in qfiles/out. But nothing gets out. I never get the confirmation message.
- I see no errors in logs/smtp-failure or any other files that would help me put my finger on the problem.
Any ideas?
-- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com | but me.
One tequila, two tequila, three tequila, floor. -- Dennis Miller
![](https://secure.gravatar.com/avatar/930c46db686975260aaad5b255d345ce.jpg?s=120&d=mm&r=g)
I have just tested a post to a new list created with the 2.1.2 mailman. Pretty much the same behavior, the .pck and .db files are created in qfiles/out. They look normal when I peek in with dumpdb. But nothing leaves the machine. No apparent errors anywhere.
Can I run the "qrunner --runner=OutgoingRunner:0:1 -s" manually and add some kind of degugging flag?
On Wed, Aug 27, 2003 at 07:56:38AM -0400, Charles Gagnon wrote:
-- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com | but me.
I went to a bookstore and asked the saleswoman where the Self Help section was, she said if she told me it would defeat the purpose. -- Dennis Miller
![](https://secure.gravatar.com/avatar/930c46db686975260aaad5b255d345ce.jpg?s=120&d=mm&r=g)
Another quick update: I had old queue files in qfiles/out. So I cleared out everything in the folder and restarted all the qrunners.
Now I can get the OutgoingRunner to send one post or one message. After it's restarted, it sends one message correctly. The next post or confirmation message after that goes nowhere.
On Wed, Aug 27, 2003 at 08:13:13AM -0400, Charles Gagnon wrote:
-- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com | but me.
There's no such thing as strong coffee, just weak people.
![](https://secure.gravatar.com/avatar/930c46db686975260aaad5b255d345ce.jpg?s=120&d=mm&r=g)
I've just upgraded from Mailman 2.0.x to 2.1.2 using Python 2.2.2 on a Solaris 8 box.
The "qrunner --runner=OutgoingRunner:0:1 -s" process starts okay and stays okay to send one message (either confirmation or post) and than is just hangs.
It does not die but doesn't send anything else. After that one message, even a "mailmanctl stop" won't stop it. I have to manually "kill -9" the process. And outgoing stuff just accumulates in qfiles/out.
The logs all seem clean I have no indication of a problem anywhere?
Any ideas?
-- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com | but me.
640K ought to be enough for anybody. -- Bill Gates, 1981
![](https://secure.gravatar.com/avatar/930c46db686975260aaad5b255d345ce.jpg?s=120&d=mm&r=g)
I have just tested a post to a new list created with the 2.1.2 mailman. Pretty much the same behavior, the .pck and .db files are created in qfiles/out. They look normal when I peek in with dumpdb. But nothing leaves the machine. No apparent errors anywhere.
Can I run the "qrunner --runner=OutgoingRunner:0:1 -s" manually and add some kind of degugging flag?
On Wed, Aug 27, 2003 at 07:56:38AM -0400, Charles Gagnon wrote:
-- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com | but me.
I went to a bookstore and asked the saleswoman where the Self Help section was, she said if she told me it would defeat the purpose. -- Dennis Miller
![](https://secure.gravatar.com/avatar/930c46db686975260aaad5b255d345ce.jpg?s=120&d=mm&r=g)
Another quick update: I had old queue files in qfiles/out. So I cleared out everything in the folder and restarted all the qrunners.
Now I can get the OutgoingRunner to send one post or one message. After it's restarted, it sends one message correctly. The next post or confirmation message after that goes nowhere.
On Wed, Aug 27, 2003 at 08:13:13AM -0400, Charles Gagnon wrote:
-- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com | but me.
There's no such thing as strong coffee, just weak people.
![](https://secure.gravatar.com/avatar/930c46db686975260aaad5b255d345ce.jpg?s=120&d=mm&r=g)
I've just upgraded from Mailman 2.0.x to 2.1.2 using Python 2.2.2 on a Solaris 8 box.
The "qrunner --runner=OutgoingRunner:0:1 -s" process starts okay and stays okay to send one message (either confirmation or post) and than is just hangs.
It does not die but doesn't send anything else. After that one message, even a "mailmanctl stop" won't stop it. I have to manually "kill -9" the process. And outgoing stuff just accumulates in qfiles/out.
The logs all seem clean I have no indication of a problem anywhere?
Any ideas?
-- Charles Gagnon | My views are my views and they http://unixrealm.com | do not represent those of anybody charlesg at unixrealm.com | but me.
640K ought to be enough for anybody. -- Bill Gates, 1981
participants (1)
-
Charles Gagnon