[
http://issues.apache.org/jira/browse/JAMES-426?page=comments#action_12365067 ]
Daniel Perry commented on JAMES-426:
I agree. The code i provided shouldn't affect any user of james, unless they
add the extra attribute, so therefore could be
[
http://issues.apache.org/jira/browse/JAMES-426?page=comments#action_12365064 ]
Daniel Perry commented on JAMES-426:
Whilst i think that this a great idea (not that i'm biased atall!), i dont
think this code provides the best solution.
The cu
> AFAIK if it has it's own virtual machine (so it's in network mode) and if
has
> more than 64MB it can gracefully handle such situations. The problem is
that under
> 64MB it can do to much for such situations - it's like an IDE that doesn't
do well
> without a minimal size :).
see below...
> No
> I am still concerned about the Derby related crash that I saw earlier.
> Anyone have a clue? And is there anything that we need to do to cleanly
> shut Derby down?
>
> --- Noel
Given that the whole crash starts with: java.lang.OutOfMemoryError, then
maybe that caused the problem with derb
[ http://issues.apache.org/jira/browse/JAMES-426?page=all ]
Daniel Perry updated JAMES-426:
---
Attachment: patch.txt
Patch for this functionality against trunk. Note: the code has not been
compiled or tested with trunk, but has been used in a live
Reporter: Daniel Perry
Priority: Trivial
Attached is a patch that makes james use the virtualusertable database for
identifying local hosts. It also allows for continued use of servernames
block. It is backwards compatible, and unless an extra attribute is added in
the configuration, it
Due to problems with our DNS servers, i shortened the times of outgoing
retries to 1 hour, and bumped up the maxRetries to 50.
Anyway, the DNS servers are sorted... so i put the config back. Everything
seemed happy... until monday morning.
60GB of log files later... and a full hard disk... duh!
[
http://issues.apache.org/jira/browse/JAMES-334?page=comments#action_12318291 ]
Daniel Perry commented on JAMES-334:
This is an SHA hash but not in the format you'd expect.
It's a base64 encoded rather than hex (reduces the size).
Its then
[
http://issues.apache.org/jira/browse/JAMES-383?page=comments#action_12314394 ]
Daniel Perry commented on JAMES-383:
Looks like same issue as: http://issues.apache.org/jira/browse/JAMES-311 which
has been fixed, but not in a release yet.
Patch as
> Cool,
>
> That will be another source of information that will be extremely helpful.
>
> One of the design ideas I had was to use the "Chain of Responsiblity"
> pattern which will allow functionality to be added in stages and make it
> easy to configure. Adding new features could them be done qu
I mentioned a while ago that I had started work on a new mailing list
system.
Anyway, it's basically a (messy) refactoring of the CommandListServ, to
allow multiple dynamically-configured lists. It uses a helper class to
manage users, lists, etc, so it can be modified to use any data source
witho
Yup, in our server i decided to just catch Throwables from the bounce
attempt (if a bounce fails, tough!)
Still, it seems a little concerning that the server may get low on memory,
then attempt to deliver a reasonably sized mail (within server limits) and
fail, and consequently keep repeating (and
> * Is it useful/desirable to parse the message-id field (into left and
> right parts)? I get the impression that a relatively high number of
> mail messages have syntactically illegal message-id values--two @ signs
> seems to be a particularly common offense. Since the message-id is
> really int
> The project I have in mind is the List Management portion of the ToDo.
Decent lists would be nice. (i have started some code for a configurable
list manager, ie. one mailet/matcher for multiple dynamically configured
lists, but not finished it)
> Specifically I'd like to add command capability
Did anyone else ever look into the (serious) issues i highlighted with the
SMTP protocol handler and remote delivery mailet?
1. endless remote delivery retries on an exception when remote delivery
bounces a message (initial problem) - can be delt with using the patch i
supplied, to catch the error
>
> Another question: Are people happier with JSP or Java Servlets? Yes,
> in the end, they compile out to the same thing, but sometimes I hear
> quasi-religious wars on which deployment solution is the best ;-)
Both!
Niether are particularly good at producing a sensible system.
JSPs are great f
> On 5/11/05, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
> > > IMHO when the limit is reached, we should either (soft fail)
> > > listen to everything and ignore the data as you describe
> > > above, or (hard fail) kill the connection. Right now what we
> > > does not make a lot of sense, since
(cross posted as code change proposal at bottom)
Having had a quick look at the RFCs it looks like the 552 error should not
be sent until the final . has been received.
Looking at james code (SMTPHandler.java):
James recieves data command. Responds accordingly.
James starts receiving data. Once
Daft question, but is it not more sensible to have 3 branches here? head,
old-head, and branch_2_1_fcs where:
branch_2_1_fcs is current stable use-this-one branch. For now it is what
people should check out to get james source, but also keep it for any
necessary maintenence releases later.
head i
+1
Will help when i can, but dont have much time at the moment.
Daniel.
> -Original Message-
> From: Danny Angus [mailto:[EMAIL PROTECTED]
> Sent: 11 April 2005 09:58
> To: server-dev@james.apache.org
> Subject: [VOTE] POJO pattern
>
>
> I propose that work commence to extract James's "v
>
> This is a very good point!
> It is very easy to make a CDI wrapper around an SDI POJO, but the
> other way
> around is rather difficult.
>
> This is probably the strongest argument for SDI, I gave seen.
>
> --Søren
So, so far we've settled on SDI POJOs (possibly with CDI wrappers :)
And presu
> Okay, so you prefer code looking like:
> p = new SomePOJO ("bla", null, null, null, null, "bli", null,
> null, null);
>
> to code looking like:
> p = new SomePOJO ();
> p.setBla("bla");
> p.setBli("bli");
>
> I must say, I prefer the latter.
I prefer the CDI one.
Take two cases:
In a normal en
[ http://issues.apache.org/jira/browse/JAMES-367?page=history ]
Daniel Perry updated JAMES-367:
---
Attachment: patch.txt
This patch also includes the addfooter patch i submitted previously!
> Added option to CommandListServ to change the to addr
)
Environment: N/A
Reporter: Daniel Perry
Priority: Minor
Attachments: patch.txt
I have added the option: [EMAIL PROTECTED]
to the CommandListServProcessor. When an email is sent to the list, it changes
the to address into this. Is useful in hiding the list address, or in any
other
[ http://issues.apache.org/jira/browse/JAMES-364?page=history ]
Daniel Perry updated JAMES-364:
---
Attachment: addfooterpatch.txt
The patch as a txt file.
> Made adding of footer optional with CommandListServProces
)
Environment: N/A
Reporter: Daniel Perry
Priority: Minor
Attachments: addfooterpatch.txt
The CommandListServ attaches a footer. This patch makes that optional, and
defaults to true. If you add: false then it will not
add the footer, and if you leave out that element (as in all existing
N/A
Reporter: Daniel Perry
James automatically adds a PostmasterAlias mailet to the root processor. This
is ok for a simple setup, but causes problems in a virtual hosting environment,
as different domains may have different postmaster accounts.
The fix is to remove the hard-coded mailet, and
> > [...]
> > The Spring framework's Java-Bean approach is IMO a way better
> > solution to the POJO-ification issue.
> >
> > Now I have started to convert James into POJO's the Spring
> > way (only James.java so far), but as I stated above I hate
> > doing things over, so let us take a vote. Is it
[
http://issues.apache.org/jira/browse/JAMES-311?page=comments#action_61203 ]
Daniel Perry commented on JAMES-311:
Patch to fix this and make james rcf compliant with respect to messages not
ending with \r\n.
--- NNTPHandler_orig.java
Yup, your code seems a little more 'with it' than mine! (first day back at work
after holiday so who can blame me!)
I'll post any other stuff i do on mailing lists up here.
Thanks,
Daniel.
> -Original Message-
> From: Noel J. Bergman [mailto:[EMAIL PROTECTED]
> Sent: 15 March 2005 04:3
I have now added some code to commandlistserv to allow you to limit posting
to certain addresses specified in the config block. Note if the person is
not a member you must turn off membersonly aswell or that will block them!
Example config:
false
...
true
[EMAIL PROTECTED]
There is a bug in
commandlistservfooter where if an html email is sent to a list, and a footer is
added, it causes the content-type to become all screwed up. The fix is to
add message.saveChanges() at appropriate points (see below). Complete
fixed file is attached. Could someone kindly co
Funnily enough, at the end of last week, i was told to sort out some mailing
lists for our server :)
I've had a look at the various list mailets available, and none meet all of
our requirements.
One problem with the mailing lists that ship with james is that they have to
be defined and configured
> My aim is to configure a development environnement upon a stable release
> to study / compile / browse / search JAMES code.
> So, i tried two ways :
> - Unzip james code including phoenix (james-with-phoenix-2.2.0-src.zip)
> and try to configure Eclipse to build the project
> - Install subversion
2-300K users?
That could give some serious number of concurrent pop3 sessions! Of course
this depends on who's using the server, and how often they will be using it!
say 100k users checking email every 20 mins... at 10 seconds per session,
that would give an average of 830 concurrent sessions at
I've finally got round to joining the dev list!
Personally I don't trust databases with things of this nature, and would
prefer to stick with using the file system. Just this week there was another
'I need help urgently' message on the user list - James went screwy because
of database issues. I do
36 matches
Mail list logo
