> Just finished plugging all major leakage. Things look usable now.
Cool, I'll try it out tonight. Will you release 2.1.4 then?
Cheers,
Kevin
Got above error message prompt on the console, however the IMAP seems still
working. Attached is the valgrind output.
Thanks,
Kevin
==22282== Memcheck, a memory error detector for x86-linux.
==22282== Copyright (C) 2002-2004, and GNU GPL'd, by Julian Seward et al.
==22282== Using valgrind-
79]: COMMAND: [. UID
SEARCH UID 1:21529]
Jan 24 14:32:27 centos42-03 dbmail/imap4d[22079]: COMMAND: [. logout]
Thanks,
Kevin
I found out when I restarted the old dbmail-imapd (svn 1918), through
Outlook Express. Spam folder suppose to have 3 emails but now have
200+, and INBOX is empty.
(I use both SquirrelMail and Outlook Express which makes it a little
bit confusing)
Kevin
On 1/23/06, Paul J Stevens <[EM
ll
Thanks,
Kevin
On 1/23/06, Paul J Stevens <[EMAIL PROTECTED]> wrote:
>
>
> Kevin wrote:
> > Got above error message prompt on the console, however the IMAP seems
> > still working.
>
> That assertion is actually harmless. It's what happens when a search
&
It must be SquirrelMail's filter did the move but I have no idea why,
as header doesn't match the filter criteria. Let's leave it for now,
and I'll see if I can find a pattern to reproduce it.
Thanks,
Kevin
On 1/24/06, Kevin <[EMAIL PROTECTED]> wrote:
> I found
If only that were legal. Maximum width for varchar is 255.
Is this a limit of mysql? I don't think pgsql has such a limit, and
oracle might be 2000.
Kevin
t the ldap
integration and Sieve support would be great.
Have you done a hot failover solution ever for DBMail. Just want message
replicated. The idea being that if I have my primary fail I can just swith
DNS to the secondary and go to town.
Thanks for Aaron,
Kevin
> On Sun, 2005-10-30
ction.
Thanks,
Kevin
--- Kevin Baker <[EMAIL PROTECTED]> wrote:
> Yes... here is my exact goal.
>
> - Replace Cyrus with DBMail on a Kolab installation
> - Inteagrate DBMail with Kolab LDAP user schema
> - test Sieve integration with Kolab admin vacation
> message stu
of
where we are on the sieve project..
Thanks,
Kevin
--- Aaron Stone <[EMAIL PROTECTED]> wrote:
> Most definitely!
>
> Since your last email, I've got DBMail 2.1 SVN to compile
> cleanly with
> libSieve 2.1.7. It doesn't actually work, because the
> call
omparison is needed.
Note that in general, you should *not* initially test your SQL queries
with MySQL. Use PostgreSQL instead, since it's much more conformant
with the ANSI SQL standard, and stick to ANSI SQL if you can. I'm
sure that has been said before, but this bug clearly illus
Aaron Stone wrote:
> On Sat, 2005-12-10 at 19:59 -0800, Kevin Brown wrote:
> [snip]
>
> > h = g_list_join((GList *)headers,"'),upper('");
> > g_string_printf(q,"SELECT physmessage_id,headername,headervalue "
> > "F
will be something like:
unsigned int results[4];
The values we're interested in (exists, seen, recent) will then be:
exists = results[0] + results[1] + results[2] + results[3];
seen = results[2] + results[3];
recent = results[1] + results[3]
Now, the above is more efficient, but the real win would be if we
could avoid calling db_getmailbox (or perhaps a modified version of it
that doesn't have to do any of this) altogether after doing a STORE.
I don't know how much more (if any) efficient the above will be on
MySQL or SQLite, but I wouldn't expect it to be less efficient.
Thoughts?
--
Kevin Brown [EMAIL PROTECTED]
Hello Arron,
How are things going with the DBMail Seive support? I saw
that you made some good progress earlier in the month.
Just curious if you've been able to find the time to focus
on it any more.
Thanks,
Kevin
Kevin Baker
Mission Vi Inc.
[EMAIL PROTECTED]
858.454
s='1') AND seen_flag=1 UNION SELECT 'c',COUNT(*) FROM
> dbmail_messages WHERE mailbox_idnr='14' AND (status='0' OR
> status='1') AND recent_flag=1
Umm...should I just submit a patch against 2.0.7 for this? What, if
anything, do you guys want me to do about this?
--
Kevin Brown [EMAIL PROTECTED]
gers to manage a stats table or something, which
would rule out MySQL and SQLite as database engines, and thus rules
out that approach entirely.
> Kevin, I'd like a imap command sequence that triggers this problem
> so I can setup a testcase for 2.1. and start fixing this properly;
> move all
Mail server, the message is perfectly retrievable.
My suspicion (unverified) is that this has to do with end-of-line
processing. But it could be something entirely different. I dunno.
--
Kevin Brown [EMAIL PROTECTED]
ate a patch against 2.0.7 for this if you guys are interested.
--
Kevin Brown [EMAIL PROTECTED]
#x27;t mind implementing that, though Paul seemed a little
underwhelmed with that solution...
I can create a patch of both types, I suppose, and let you and Paul
fight it out as to what to do with them. Might be a fun show. :-)
--
Kevin Brown [EMAIL PROTECTED]
when recent_flag > 0 then 1 else null end) FROM
> > dbmail_messages WHERE status < '2' and mailbox_idnr='1';
>
> and this one works for mysql-4.0 just fine.
It's also as efficient as the query I put up, so I'm cool with using
it.
I'll create a patch using it i
? I checked the release
> notes and I couldn't find mention of it...
If you're looking to find the total size of a table, no. In
PostgreSQL, that's a Fundamentally Hard Problem because of MVCC.
It will make use of indexes in the WHERE clause, of course, to limit
the rows it has to visit. I think that's always been the case.
--
Kevin Brown [EMAIL PROTECTED]
on of the userbase for all I know, so it's certainly
a consideration).
--
Kevin Brown [EMAIL PROTECTED]
even as it is, it's way cooler than any other IMAP
server out there that I've ever seen.
--
Kevin Brown [EMAIL PROTECTED]
format as the backend would be for
such a large mailbox. I would expect it to be slow, too).
Note that the version of DBMail being used is 2.0.7 with the
modification to db_getmailbox() that uses the more efficient query.
It's still not *fast* for huge mailboxes,
Paul J Stevens wrote:
> And I suspect 5.0 will be become a requirement for 2.3+ so we can
> start working with views, triggers, etc...
Hmm...that would imply ditching SQLite, wouldn't it?
Not that I have a problem with that, mind you.
--
epresentation at the table level
completely different between databases if that proved to be
advantageous.
--
Kevin Brown [EMAIL PROTECTED]
, that should just change the line numbers
involved.
--
Kevin Brown [EMAIL PROTECTED]
#! /bin/sh /usr/share/dpatch/dpatch-run
## 05_mailboxspeed.dpatch by <[EMAIL PROTECTED]>
##
## All lines beginning with `## DP:' are a description of the pa
Geo Carncross wrote:
> On Thu, 2005-12-15 at 14:16 -0800, Kevin Brown wrote:
> > Paul J Stevens wrote:
> > > And I suspect 5.0 will be become a requirement for 2.3+ so we can
> > > start working with views, triggers, etc...
> >
> > Hmm...that would imply dit
Paul J Stevens wrote:
>
>
> Kevin Brown wrote:
>
> >>From what I can tell, just doing a STORE of attributes is enough. I
> > can recreate the problem and email you the logs if you're interested.
>
> I was more like thinking about some sort of lazy c
sees a reply to a bug it should add the
contents of the reply to the bug. It's annoying to have to deal with
multiple interfaces just to get the complete history of the bug, since
some of that history comes from discussions here while the rest of it
is in the bug tracking system proper.
In general, I find that email works much better for most bug-related
correspondence.
--
Kevin Brown [EMAIL PROTECTED]
7;t really hear it when I explain just how cool
DBMail is.
- Kevin
--- Aaron Stone <[EMAIL PROTECTED]> wrote:
> I'm going to use the replycache table, with:
>
> to_addr: obviously, the outgoing address.
> from_addr: the envelope recipient address (rather than
> the user
Hey Aaron,
Aaron Stone wrote:
Kevin, thanks very much for your generosity of spirit and of pocketbook.
You gave me the push I needed to get back to the grindstone.
NP It's a great project, psyched to help out in this way since I can
really be on the dev end of things for C
Are there any decent howto's I could use on setting up
DBMail with LDAP?
I saw the page in the wiki, but was hoping for something a
bit more detailed.
Mainly:
- ldap schema example (ldif)
- which conf files to edit
-
Thanks,
Kevin
Kevin Baker
Mission Vi Inc.
[EMAIL PROT
me of the $$ I committed, as it seems you
have made great progress.
Thanks,
Kevin
Kevin Baker
Mission Vi Inc.
[EMAIL PROTECTED]
858.454.5532
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
> series.
> >
> > We don't have our own schema, rather, the configuration
> has been made
> > flexible enough to handle a couple of popular schemas.
> The qmail-ldap
> > schema works, even a basic inetOrgPerson should work.
> MS Exchange /
> > Active
causes the problem
>
> e.g. apt-get install libdbmail-mysql
>
>
>
> On Mon, Mar 13, 2006 at 03:29:10PM -0800, Kevin Baker
> wrote:
> > Great. So I was hoping to install the 2.1.x branch on
> > Debian 3.1 to get the new Sieve features.
> >
> > I update
Oops, Thanks Paul. You are exactly right.
I'll work on tweaking my source.list as you described.
I'll post once I have the solution for this. Can't wait to
play with 2.1.x
- Kevin
--- Paul J Stevens <[EMAIL PROTECTED]> wrote:
> No Kevin, your libs are too old. Are
anks again Paul for the advice on updating apt-get.
- Kevin
--- Kevin Baker <[EMAIL PROTECTED]> wrote:
> Oops, Thanks Paul. You are exactly right.
>
> I'll work on tweaking my source.list as you described.
>
> I'll post once I have the solution for this. Can&#x
d. The comments in -h imply that
> this is the
> > proper behavior, and I think it would be nice to have.
> >
> > Obviously the same idea could be applied to the other
> daemons.
> >
> > Matt
>
> ___
>
version $Revision: 1944 $ (c)
1999-2004 IC&S, The Netherlands
- Kevin
>
> On Tue, Mar 14, 2006, Kevin Baker <[EMAIL PROTECTED]>
> said:
>
> > While on the topic of the letter "V", it would be good
> to
> > have a "--version" flag on dbmail-u
--- Marc Dirix <[EMAIL PROTECTED]> wrote:
> Reboot??
>
> For installing dbmail?
> Did it also update your kernel or what?
Nope... it just had to shut down various services I
guess... and suggested to reboot, to be sure everything
came back up properly.
>
> /Marc
n/off the Sieve Notify
extension (unsupported).
AUTO_NOTIFY=no # Use the auto_notify table to send
email notifications.
AUTO_REPLY=no # Use the auto_reply table to send
away messages.
# end of configuration file
Kevin Baker
Mission Vi Inc.
[EMAIL PROTECTED]
858.454.5532
nect to
> authentication. If you switch
> to authdriver=sql, does it start without error?
>
> Aaron
>
> On Thu, 2006-04-06 at 00:27 -0700, Kevin Baker wrote:
> > So I finally have some time to configure our dev dbmail
> > config to use LDAP.
> >
> >
Ok, I'll go through the perms. I might have missed
something... thanks guys.
Once I get the mysql settings working, I might need a hand
getting the LDap settings right.
--- Paul J Stevens <[EMAIL PROTECTED]> wrote:
>
> Kevin, did you verify the database parameters? Basic
figuration
> has been made
> > flexible enough to handle a couple of popular schemas.
> The qmail-ldap
> > schema works, even a basic inetOrgPerson should work.
> MS Exchange /
> > Active Directory should work, and if it's not working,
> that's a bug.
> >
Reply below...
--- Paul J Stevens <[EMAIL PROTECTED]> wrote:
>
>
> Kevin Baker wrote:
> > So I'm back to the ldap configuration. Still not there.
> >
> > I have DBMail working perfectly on just mysql auth.
> When I
> > switch to the "
--- Kevin Baker <[EMAIL PROTECTED]> wrote:
> Reply below...
>
>
> --- Paul J Stevens <[EMAIL PROTECTED]> wrote:
> >
> >
> > Kevin Baker wrote:
> > > So I'm back to the ldap configuration. Still not
> there.
> > >
TW: I am running OpenLDAP: slapd 2.2.26 and dbmail 2.2
- Kevin
--- Kevin Baker <[EMAIL PROTECTED]> wrote:
>
>
> --- Kevin Baker <[EMAIL PROTECTED]> wrote:
>
> > Reply below...
> >
> >
> > --- Paul J Stevens <[EMAIL PROTECTED]>
I have this setup on my system:
postfix master.cf
dbmail-lmtp unix - - n - - lmtp -v
postfix main.cf
mailbox_transport = dbmail-lmtp:127.0.0.1:24
dbmail dbmail.conf
[LMTP]
PORT=24 # Port to bind to.
Thanks
Kevin Baker
Mission Vi Inc.
[EMAIL PROTECTED]
858.45
lnet localhost 24' to see if you can
> connect to it.
>
> Aaron
>
> On Thu, Apr 20, 2006, Kevin Baker <[EMAIL PROTECTED]>
> said:
>
> > Can some one give me an example of how to have postfix
> pass
> > messages through lmtp to dbmail?
> >
bmail-util -V
DBMail: dbmail-util
Version: 2.1
$Revision: 1897 $
Any thoughts... is this a bug that is fixed in newer
versions?
Thanks.
Kevin
Kevin Baker
Mission Vi Inc.
[EMAIL PROTECTED]
858.454.5532
__
Do You Yahoo!?
Tired of spam? Y
) Checked dbmail.conf for correct settings
5) Problem: I can not connect to server after start.
Any thoughts?
Thanks,
- Kevin
--- Paul J Stevens <[EMAIL PROTECTED]> wrote:
> Kevin, please try the latest unstable packages.
>
> Please note that i've 'normalized'
--- Paul J Stevens <[EMAIL PROTECTED]> wrote:
> Kevin,
>
> what errors do you see?
I was getting the socket error:
[error]
dbmysql.c,db_connect: MySQL host is set to localhost, but
no mysql_socket has been set. Use sqlsocket=... in
dbmail.conf. Connecting will be attempted us
--- Paul J Stevens <[EMAIL PROTECTED]> wrote:
> Kevin Baker wrote:
> > --- Paul J Stevens <[EMAIL PROTECTED]> wrote:
> >
> >
> >>Kevin,
> >>
> >>what errors do you see?
> >
> > I was getting the socket error:
> >
HAT IS THIS]')
Can someone help me fill in the blanks on the [WHAT IS
THAT] stuff... and any query holes?
- Kevin
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Does DBMail support "SHA" passwords?
My current app uses SHA, but I can switch to MD5 if I need to.
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
--- Aaron Stone <[EMAIL PROTECTED]> wrote:
> On Tue, 2006-05-02 at 17:29 -0700, Kevin Baker wrote:
> > Does DBMail support "SHA" passwords?
> >
> > My current app uses SHA, but I can switch to MD5 if I
> need to.
>
> SHA is not supported. I'm
How do I go about specifying MD5 encryption on a dbmail
user with the dbmail-util?
Right now I am using:
dbmail-users -a kbaker -w MYPASSWORD -g 0 -m 250M -s kbaker
at MYDOMAIN [EMAIL PROTECTED]
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has
utils would be depending on an older version of
libdbmail that was installed from a previous install.
To fix this, you just need to add "/usr/lib/dbmail*" to
debian/dbmail.install.
--
Kevin Brown [EMAIL PROTECTED]
I stupidly wrote:
> To fix this, you just need to add "/usr/lib/dbmail*" to
> debian/dbmail.install.
Man, I don't know what my deal is today. Can't type worth a damn.
Anyway, you need to add "/usr/lib/libdbmail*" to
debian/dbmail.install, not &qu
retty much solid. I want to get a group of people together
to test improve it on a separate virtual dbmail project..
figured the list would be a good way to let people know.
Thanks,
Kevin
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the
Problem corrected... thanks.
--- Kevin Baker <[EMAIL PROTECTED]> wrote:
> Hello Dev,
>
> I posted an email about a dbmail/postfix virtualization
> project I'm working on.
>
> It doesn't seem to be posting to the users list.
>
> Could you maybe take a lo
bmail running on two ports on one
machine.
Being able to speficify a non default conf file would be
great!
- Kevin
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
oops... just found the "-f" flag in the man page.
So I'm guessing if I do a
dbmail-imapd -f /etc/dbmail/dbmail1.conf
and
dbmail-imapd -f /etc/dbmail/dbmail2.conf
specifying unique information per server in each conf file
I'll be all set...?
--- Kevin Baker <[
dnr etc. to "id". I remember somewhere I read that you intended to
make dbmail friendly to RoR(Rails). With the old primary key names, it's a
little bit hard to develop webmail in RoR (yes I know IMAP is the choice but
performance gain with direct sql access i
Are there any significant performance gains using Pg over
MySQL on the newer releases?
Just curious... with the discussion of possibly adding
real transaction support on Pg, it might be worth looking
into.
Thanks,
Kevin
not even close to production, could someone
suggest a fitlering solution?
I need:
- message header filtering for amavis
- vacation filters
Thanks,
Kevin
the
system for it. I haven't spent much time with the code,
but couldn't the unique_id be used instead of the existing
ID.
Thanks,
Kevin
ne machine to even numbers and the other to odd?
I'm sure it has lots of holes, but something to throw out
there. It would likely end up skipping some numbers, but
would be pretty much sequential and avoid collisions.
Kevin
> ""Kevin Baker"" <[EMAIL PROTECTED]
s:
1. The message-id could be removed from the db and just
calculated on a mailbox list request. Its just their
sequence.
2. The unique-id could be calculated with a
date/time/sec/millsec base combined with a machine id
added to the dbmail configuration to assure ascending and
unique.
- Kevin
Kevin
2104527002
In this scenario either machine could go down before the
message was replicated to the other. When it came back up
and the synced again it could commit it without problems.
- Kevin
I don't think so.. I'm pretty sure that the email client
will just refresh the mailbox it is displaying.. messages
are then sorted by message-id or other if in the view
settings.
We have had this happen in our current configuration cyrus
as messages can sometimes get caught up in amavis running
th
This could never happen.
The message id is based on the time that it comes in.
If one server goes down, and messages come in while it is
down... when it comes back up the messages are all in the
timeslot while it was down (ids based on this timeslot)
Now any new messages on the server that just
lly, while keeping the ascending
order. However by RFC they only need to be unique to the
mailbox(folder).
- Kevin
> I don't think so.. I'm pretty sure that the email client
> will just refresh the mailbox it is displaying.. messages
> are then sorted by message-id or ot
or so back (although I probably
> wouldn't
> have noticed the collisions if I had used ms resolution
> :).
>
> Jn
>
>
> Original Message
> From: Kevin Baker
> To: "DBMAIL Developers Mailinglist"
>
> Subject: Re: [Dbmail-dev] GUID vs
>> A prev message in thread discussed this solution in
>> project. Apparently it works great.
>
> Are you referring to Simon Cocking's message? I don't
> think he was
> referring to IMAP.
>
Right that was the message I was referring to. He was
talking about message routing through his system. Wh
retty confident a thorough review of the messages
in this thread and the rfc would put us all on the same
page.
Its not that dbmail implementation is wrong.. its just
that the rfc leaves a lot of breathing room from where the
current implementation is. That room can be used to
correct these issues.
> installation. PostgreSQL is no slower than MySQL for
> similar work,
> and it is faster under pressure, since it scales better
> with load than
> MySQL does. (Same pattern, for the same reasons, as when
> comparing
> Oracle to MS-SQL.)
I guess a bit OT for this thread, but has this been
conf
> "Kevin Baker" <[EMAIL PROTECTED]> writes:
> has to do. However, about two years ago, experiments
> conducted by the
> people at Slashdot showed that the two systems were
> comparable even
> for simple tasks, while PostgreSQL totally whipped MySQL
> once t
le to read it. To support shared mailboxes
properly you will need ACLs and to hold the seen flags for a message per
message per user, not per message.
--
Kevin Thorpe
Head of IT
Purchasing Index Ltd
Tel: (+44) 20 7350 4200
Mailto: [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
Paul J Stevens wrote:
Kevin,
What version are you looking at?
Dbmail-2 *does* support shared folders. The only thing missing from
the current ACL implementation is group support. You can't grant
access to mailboxes to groups -- only to specific users and 'anyone'.
Adding gro
I'd be very interested!
>
> Wondering if anyone out there has a need for a class like
> this? I'm
> probably guessing so, as I'm working on this to integrate
> into an existing
> hosting/control panel, and am considering just BSD
> Licensing it and
> letting ya'll have your way with it. :)
>
>
on this.
STATEMENT ABOUT GUID's for Message ID's
http://tinyurl.com/62luv
RFC Overview for Message ID's
http://tinyurl.com/3jjrm
OLD THREAD ABOUT THIS
[Dbmail] MySQL Load Balancing & Failover
http://tinyurl.com/3m6mf
EXAMPLE OF HOWTO GENERATE GUID
http://tinyurl.com/4y696
START OF THREAD
http://tinyurl.com/6mnu9
Thanks,
Kevin Baker
t; > the first place, and use LOAD DATA INFILE (which is standard ANSI, I
> > believe) to stuff it into the database once we've done the delivery stuff.
Ick. Such an operation is definitely not portable, and probably isn't
transactional, either.
--
Kevin Brown [EMAIL PROTECTED]
binary order of magnitude)
to the planner's estimate, so it means the statistics are fairly
good here.
2. Try doing a REINDEX TABLE dbmail_messages.
3. Are you running autovacuum? If not, how often do you vacuum the
tables in this database?
--
Kevin Brown [EMAIL PROTECTED]
> except the mailbox idnr).
>
> That is fast (MyISAM or InnoDB?), but I would expect something below 1s too.
>
> If no one has an idea what could be wrong I'll ask on the PostgreSQL sql
> list tomorrow.
At this point, you might consider posting to pgsql-performance as well
as (or, perhaps, instead of) pgsql-sql.
--
Kevin Brown [EMAIL PROTECTED]
>> Does mysql-4.0 support UNION selects?
Yes MySQL 4.x does support UNION Selects.
http://dev.mysql.com/doc/mysql/en/union.html
I'm using them on a couple projects.
>
> Yes, and I'll test anything you post to the list.
>
>
> Aaron
> ___
> Dbmail-dev
Good catch, 4.0.x is a better description.
"UNION is available from MySQL 4.0.0 on."
[ http://dev.mysql.com/doc/mysql/en/union.html ]
> On Mon, Mar 14, 2005, ""Kevin Baker""
> <[EMAIL PROTECTED]> said:
>
>>
>>>> Does mysql-4.
thoughts.
wiki: Feature Requests, Wish List
http://www.dbmail.org/dokuwiki/doku.php?id=start#6
wiki: Replace unique_id with GUID for Load Balancing/Failover
http://www.dbmail.org/dokuwiki/doku.php?id=unique_id_guid
Thanks,
Kevin Baker
d values.
Each cluster synced through master-master. DNS Round Robin
for load "balance". Heartbeat for failover.
This would give loadbalance and failover on commodity
hardware. With guid unique_id's there would be no id
collision so it should failover and sync well.
Kevin
&
QL installations start the database
using the local time zone, so clients will need to set their time zone
differently if they want a different time zone. I think it would be
very unwise for dbmail to rely on the time zone of the database.
I'm actually somewhat surprised
Geo Carncross wrote:
> On Fri, 2005-05-13 at 01:49 -0700, Kevin Brown wrote:
> > Geo Carncross wrote:
> > > On Thu, 2005-05-12 at 17:08 +, Aaron Stone wrote:
> > > > >> Geo, would you this up as a patch against current dbmail_2_0_branch
> > >
code.
I think this will be a great addition to the system.
Scalable multi-master mailserver with no message id
collisions!!!
Kevin Baker
> Ok.
>
> I've committed some initial code to svn. Next step will be
> to actually
> replace auto_increment columns with the new bigint ki
gt; totally swamped in work :-\
>
Yes thanks again Let me know if I can help in any way,
even if it is just research. It was actually kind fun to
"decode" that giant RFC ;)
Kevin Baker
> Aaron
> ___
> Dbmail-dev mailing list
> Dbmail-dev@dbmail.org
> http://twister.fastxs.net/mailman/listinfo/dbmail-dev
>
lustered databases.
So the problem seems to be generating a sequential 32bit
char based on time and server id.
I'll have to do some reading on 32bit chars... to get my
head around this...
Kevin
was
restarted sync to the cluster before accepting messages..
but this could just be part of the startup process.
Sounds a bit less elagant buy might work fine. Plus just
another idea to put out there. It would likely take care
of the 32bit restrictions with timestamps.
Kevin Baker
> Here&#
o could you exmplain what you mean by clock running fast
or slow?
If all servers in the cluster are synced with a remote NTP
server periodically I can't see how this would happen.
This could just be a requirement of clustered multi-master
systems.
Kevin
>> If all servers in the cluster are synced with a remote
>> NTP
>> server periodically I can't see how this would happen.
>> This could just be a requirement of clustered
>> multi-master
>> systems.
>
> Let's keep it simple. Two machines, one is configured to
> generate odd
> sequence numbers and
years to live from the time of the
> mailbox's creation.
So I've been thinking more about this. Sounds like a great
idea. Would the mailbox_time be system-wide or would it be
a set server setting.
Either way I think 50 years is a good comprimize. Then we
can put in for an update to
> On Fri, 2005-05-27 at 10:51 -0700, Kevin Baker wrote:
>> Thanks for the explination I totally understand now.
>>
>> That being said, what is the actual harm in one message
>> id
>> being even a second off for the UID. They will both be
>> entered in the m
1 - 100 of 117 matches
Mail list logo
