qmail Digest 3 Aug 2000 10:00:00 -0000 Issue 1082
Topics (messages 45990 through 46071):
Re: updated load balancing qmail-qmqpc.c mods
45990 by: JuanE
45992 by: JuanE
45996 by: James Raftery
46002 by: Michael T. Babcock
46010 by: JuanE
46013 by: Austad, Jay
46015 by: Michael T. Babcock
46021 by: Austad, Jay
46022 by: JuanE
46025 by: David Dyer-Bennet
46030 by: JuanE
46036 by: Michael T. Babcock
46039 by: Michael T. Babcock
46067 by: Eric Cox
46071 by: Pavel Kankovsky
rblsmtpd
45991 by: Slider
46000 by: Paul Farber
46029 by: Jon Rust
46057 by: Eric Cox
build-rblsmtpd error
45993 by: Neil D. Roberts
45994 by: Petr Novotny
45995 by: Gerrit Pape
Re: qmail-pw2u error
45997 by: Chris, the Young One
45998 by: Dave Sill
46003 by: Chris, the Young One
46006 by: Dave Sill
46054 by: Chris, the Young One
46059 by: Vincent Danen
Re: Mailing list performance
45999 by: Fernando Almeida
46001 by: Vince Vielhaber
46004 by: Irwan Hadi
46005 by: Dave Sill
46007 by: Dave Sill
46023 by: David Dyer-Bennet
46026 by: Dave Sill
46032 by: Fernando Almeida
46053 by: Irwan Hadi
Re: why qmail is more secure, was: Mailing list performance
46008 by: Ronny Haryanto
46011 by: Dave Sill
46012 by: Ronny Haryanto
46014 by: Michael T. Babcock
46024 by: David Dyer-Bennet
Re: Cannot send mail thru SMTP
46009 by: Dave Sill
Outlook choking on messages
46016 by: Albert Hopkins
46018 by: Petr Novotny
46028 by: Henrik �hman
46046 by: Albert Hopkins
[Announce] oMail 0.90a - Completely redesigned version
46017 by: Olivier M.
46050 by: Olivier M.
help - qmail rejecting mail. "no mailbox here by that name".
46019 by: J
46020 by: Dave Sill
Any special order of install for various qmail "tools" ?
46027 by: Andy Meuse
46040 by: Jack Barnett
Re: Duplicate Message-ID question
46031 by: Dave Sill
how do you apply a patch?
46033 by: Russ Ringer
46035 by: Jack Barnett
46042 by: Erwin Hoffmann
update Re: help - qmail rejecting mail. "no mailbox here by that name"
46034 by: J
change password via web
46037 by: Mark Lo
Re: New to list and to DNS - need just a little help
46038 by: Ben Beuchler
Re: SPAM BLOCKING
46041 by: Robin S. Socha
maildirmake
46043 by: Eddie Greer
46048 by: qmail List
46049 by: Bryan Ischo
EZMLM Moderation Woes
46044 by: JB
qmail - cyrus
46045 by: Wolfgang Wagner
46047 by: Vince Vielhaber
Qmail and redirecting mail
46051 by: Wilson Fletcher
46052 by: Gavin Cameron
ezmlm-idx QMQP support does not work
46055 by: Wayne Chu
46064 by: Oliver Koch
46068 by: P.Y. Adi Prasaja
Re: Microsoft to Move Hotmail to Windows 2000
46056 by: Irwan Hadi
46061 by: Irwan Hadi
46062 by: Sean C Truman
Always got like this ? Fwd: Returned mail: User unknown
46058 by: Irwan Hadi
Refuse long list rcpt!
46060 by: Nguyen Dang Phuoc Dong
source rpm
46063 by: Sumith Ail
svscan/supervise run script
46065 by: John Conover
46066 by: Chris, the Young One
trouble injecting bounce message
46069 by: Joel Gautschi
46070 by: Slider
Administrivia:
To unsubscribe from the digest, e-mail:
[EMAIL PROTECTED]
To subscribe to the digest, e-mail:
[EMAIL PROTECTED]
To bug my human owner, e-mail:
[EMAIL PROTECTED]
To post to the list, e-mail:
[EMAIL PROTECTED]
----------------------------------------------------------------------
Jay,
A small modification to what you have done can get rid of the problem of
banging one server. Instead of just sampling a random starting point for
the search, just get a random number between 1 and N (N being the number of
servers) and if that server is down just sample again until you find a
server that's up. This way, you will redistribute the load randomly to all
servers and not just the next one on the list.
What I propose sounds very expensive computationally, but there might be a
way to bypass using rand. If there is a way to get the time down to
microseconds (I believe gettimeofday() will do this), then you can use the
microseconds in the time as your random variable. Just take the number of
microseconds mod N. This should be much faster than using rand and close
enough to uniform for this purpose (anybody disagree??).
If this is feasible, then it sounds much simpler than trying to deamonize
qmqpc.
Hope this helps.
JES
Austad, Jay writes:
> You are correct. However, since qmail-qmqpc does not continually run, it is
> not possible to do a non-random round robin (as far as I know). If qmqpc
> could be turned into a process that ran continuously, it would be possible
> to do regular round-robin and there would be no need for the random stuff at
> all, this may also increase performance because it wouldn't have to read
> /var/qmail/control/qmqpservers every time it started. I haven't looked at
> it, but does qmail-inject work by just piping it's data into qmail-queue
> (qmail-queue is a link to qmail-qmqpc in a mini-qmail setup)? If so,
> couldn't qmail-queue be turned into a named pipe, and qmqpc could be
> modified slightly to continously run and wait for input on that named pipe?
> Just a random thought. Of course, I'm not a programmer and I may not have
> any idea what the hell I'm talking about. :)
>
> This approach that I'm using obviously isn't going to spread the load
> between servers exactly evenly, but it should be close enough when mailing
> large amounts of mail, unless one qmqp server dies. However, our qmqp
> servers have port 628 for QMQP monitored closely and I'll get paged within a
> couple of minutes if it goes down, so I'm not too worried about it. I guess
> I could redo it so instead of just rotating the array by a random number, it
> would shuffle it it randomly. That would keep it from banging on the server
> immediately after the dead one, and give everything an even chance again.
>
> Jay
>
> -----Original Message-----
> From: JuanE [mailto:[EMAIL PROTECTED]]
> Sent: Tuesday, August 01, 2000 6:57 PM
> To: [EMAIL PROTECTED]
> Subject: Re: updated load balancing qmail-qmqpc.c mods
>
>
>
> Jay,
>
> If I understand this correctly, then I think there is a flaw in this type
> of random round robin. I think I can best describe it with an example.
>
> Assume you have 4 servers A, B, C, D, and that server C is down. When your
> random seed hits server A, B or D (with probability 1/4, respectively) the
> message will go through fine. When your random seeds hits server C (again
> with p=1/4), you will realize that it is down and proceed to D. So in
> effect you will hit servers A and B with probability 1/4 each, and server D
> with probability 1/2. This is not very good load balancing, because
> basically the load of the server that's down is inherited by the next
> server, which is not what you set out to accomplish.
>
> With simple, non-random round-robin, when ever you hit server C, you will
> skip it and go to server D. The next message will go to server A, the next
> to B, etc... Thus, the probability of hitting server D is 1/3 (as well as A
> and B).
>
> All this rests on my interpretation of what you have done, since I am not
> familiar with the code.
>
> JES
>
> Austad, Jay writes:
>
> > Ok, I fixed it. It now just rotates the serverindex.pos[] by a random
> > amount, and then loops through until it finds a good server. rand() is
> still
> > seeded with milliseconds from the system clock. I used memcpy() to move
> the
> > array around instead of loops to make it more efficient.
> >
> > Would it be possible to run qmqpc as some sort of daemon? Then it could
> > have memory of which server it contacted last and make sure it went to the
> > next one, instead of having the chance of picking the same one again.
> >
> > Jay
> >
> >
> >
> > #include <sys/types.h>
> > #include <string.h>
> > #include <memory.h>
> > #include <sys/socket.h>
> > #include <stddef.h>
> > #include <netinet/in.h>
> > #include <arpa/inet.h>
> > #include <stdlib.h>
> > #include <sys/time.h>
> > #include <sys/timeb.h>
> > #include "substdio.h"
> > #include "getln.h"
> > #include "readwrite.h"
> > #include "exit.h"
> > #include "stralloc.h"
> > #include "slurpclose.h"
> > #include "error.h"
> > #include "sig.h"
> > #include "ip.h"
> > #include "timeoutconn.h"
> > #include "timeoutread.h"
> > #include "timeoutwrite.h"
> > #include "auto_qmail.h"
> > #include "control.h"
> > #include "fmt.h"
> >
> > #define PORT_QMQP 628
> >
> > void die_success() { _exit(0); }
> > void die_perm() { _exit(31); }
> > void nomem() { _exit(51); }
> > void die_read() { if (errno == error_nomem) nomem(); _exit(54); }
> > void die_control() { _exit(55); }
> > void die_socket() { _exit(56); }
> > void die_home() { _exit(61); }
> > void die_temp() { _exit(71); }
> > void die_conn() { _exit(74); }
> > void die_format() { _exit(91); }
> >
> > int lasterror = 55;
> > int qmqpfd;
> >
> > int saferead(fd,buf,len) int fd; char *buf; int len;
> > {
> > int r;
> > r = timeoutread(60,qmqpfd,buf,len);
> > if (r <= 0) die_conn();
> > return r;
> > }
> > int safewrite(fd,buf,len) int fd; char *buf; int len;
> > {
> > int r;
> > r = timeoutwrite(60,qmqpfd,buf,len);
> > if (r <= 0) die_conn();
> > return r;
> > }
> >
> > char buf[1024];
> > substdio to = SUBSTDIO_FDBUF(safewrite,-1,buf,sizeof buf);
> > substdio from = SUBSTDIO_FDBUF(saferead,-1,buf,sizeof buf);
> > substdio envelope = SUBSTDIO_FDBUF(read,1,buf,sizeof buf);
> > /* WARNING: can use only one of these at a time! */
> >
> > stralloc beforemessage = {0};
> > stralloc message = {0};
> > stralloc aftermessage = {0};
> >
> > char strnum[FMT_ULONG];
> > stralloc line = {0};
> >
> > struct sindex
> > {
> > int pos[256];
> > int len;
> > };
> >
> > void getmess()
> > {
> > int match;
> >
> > if (slurpclose(0,&message,1024) == -1) die_read();
> >
> > strnum[fmt_ulong(strnum,(unsigned long) message.len)] = 0;
> > if (!stralloc_copys(&beforemessage,strnum)) nomem();
> > if (!stralloc_cats(&beforemessage,":")) nomem();
> > if (!stralloc_copys(&aftermessage,",")) nomem();
> >
> > if (getln(&envelope,&line,&match,'\0') == -1) die_read();
> > if (!match) die_format();
> > if (line.len < 2) die_format();
> > if (line.s[0] != 'F') die_format();
> >
> > strnum[fmt_ulong(strnum,(unsigned long) line.len - 2)] = 0;
> > if (!stralloc_cats(&aftermessage,strnum)) nomem();
> > if (!stralloc_cats(&aftermessage,":")) nomem();
> > if (!stralloc_catb(&aftermessage,line.s + 1,line.len - 2)) nomem();
> > if (!stralloc_cats(&aftermessage,",")) nomem();
> >
> > for (;;) {
> > if (getln(&envelope,&line,&match,'\0') == -1) die_read();
> > if (!match) die_format();
> > if (line.len < 2) break;
> > if (line.s[0] != 'T') die_format();
> >
> > strnum[fmt_ulong(strnum,(unsigned long) line.len - 2)] = 0;
> > if (!stralloc_cats(&aftermessage,strnum)) nomem();
> > if (!stralloc_cats(&aftermessage,":")) nomem();
> > if (!stralloc_catb(&aftermessage,line.s + 1,line.len - 2)) nomem();
> > if (!stralloc_cats(&aftermessage,",")) nomem();
> > }
> > }
> >
> > void doit(server)
> > char *server;
> > {
> > struct ip_address ip;
> > char ch;
> >
> > if (!ip_scan(server,&ip)) return;
> >
> > qmqpfd = socket(AF_INET,SOCK_STREAM,0);
> > if (qmqpfd == -1) die_socket();
> >
> > if (timeoutconn(qmqpfd,&ip,PORT_QMQP,10) != 0) {
> > lasterror = 73;
> > if (errno == error_timeout) lasterror = 72;
> > close(qmqpfd);
> > return;
> > }
> >
> > strnum[fmt_ulong(strnum,(unsigned long) (beforemessage.len + message.len
> +
> > aftermessage.len))] = 0;
> > substdio_puts(&to,strnum);
> > substdio_puts(&to,":");
> > substdio_put(&to,beforemessage.s,beforemessage.len);
> > substdio_put(&to,message.s,message.len);
> > substdio_put(&to,aftermessage.s,aftermessage.len);
> > substdio_puts(&to,",");
> > substdio_flush(&to);
> >
> > for (;;) {
> > substdio_get(&from,&ch,1);
> > if (ch == 'K') die_success();
> > if (ch == 'Z') die_temp();
> > if (ch == 'D') die_perm();
> > }
> > }
> >
> > stralloc servers = {0};
> >
> > main()
> > {
> > int i;
> > int j;
> > int randj;
> > int randarr[256];
> > struct timeb tp;
> > struct sindex serverindex;
> >
> > sig_pipeignore();
> >
> > if (chdir(auto_qmail) == -1) die_home();
> > if (control_init() == -1) die_control();
> > if (control_readfile(&servers,"control/qmqpservers",0) != 1)
> > die_control();
> >
> > getmess();
> > serverindex.len = 1;
> > serverindex.pos[0]=0;
> > for (j = 0; j < servers.len; j++)
> > {
> > if (servers.s[j] == NULL) {
> > serverindex.pos[serverindex.len] = j+1;
> > serverindex.len++;
> > }
> > }
> > serverindex.len--;
> > ftime(&tp);
> > srand(tp.millitm);
> > randj = (serverindex.len*1.0)*rand()/(RAND_MAX+1.0);
> >
> > memcpy(randarr, serverindex.pos + randj, 4 * (serverindex.len - randj));
> > memcpy(randarr + (serverindex.len-randj), serverindex.pos, (randj) * 4);
> >
> > for (j=0; j < serverindex.len; j++)
> > {
> > doit(servers.s + randarr[j]);
> > }
> >
> >
> > _exit(lasterror);
> > }
> >
>
>
I like the idea of statically listing them, but as you mentioned it is not
very scalable. The shuffling can be very expensive. I believe it would grow
with N^2. Instead of shuffling, just plain random sampling would be better
computationally. See me previous posting.
I think this is a "profile, don't speculate" scenario.
JES
Frank D. Cringle writes:
> "Austad, Jay" <[EMAIL PROTECTED]> writes:
> > From: JuanE [mailto:[EMAIL PROTECTED]]
> >> Assume you have 4 servers A, B, C, D, and that server C is down. When your
> >> random seed hits server A, B or D (with probability 1/4, respectively) the
> >> message will go through fine. When your random seeds hits server C (again
> >> with p=1/4), you will realize that it is down and proceed to D. So in
> >> effect you will hit servers A and B with probability 1/4 each, and server D
> >> with probability 1/2. This is not very good load balancing, because
>
> > You are correct. However, since qmail-qmqpc does not continually run, it is
> > not possible to do a non-random round robin (as far as I know).
>
> Choose a permutation of the server list at random. Then step through
> this list until you hit a good server or exhaust the list.
>
> 1. A B C D
> 2. A B D C
> 3. A C B D
> 4. A C D B
> ....
> 24. D C B A
>
> If your random number is 4, you try A, then C, then D and finally B.
>
> If the number of servers is reasonably small, say up to 7, you could
> enumerate all permutations as a static array in the code and then use
> a random index into the array to determine the order. Generating a
> random permutation algorithmically is not too easy. You can either
> approximate by shuffling or make a direct choice using some bignum
> arithmetic described in "Permutation Generation Methods", Robert
> Sedgewick, Computing Surveys, Vol. 9, No. 2, June 1997.
>
> --
> Frank Cringle, [EMAIL PROTECTED]
> voice: (+49 2304) 467101; fax: 943357
>
On Wed, Aug 02, 2000 at 11:13:22AM +0000, JuanE wrote:
> This way, you will redistribute the load randomly to all
> servers and not just the next one on the list.
Not quite. If choose truly randomly the liklehood of you choosing the
same server /every/ time is the same as the liklihood of choosing a
nice even spread across the servers. This isn't what you want.
Regards,
james
--
James Raftery (JBR54) - Programmer Hostmaster - IE TLD Hostmaster
IE Domain Registry - www.domainregistry.ie - (+353 1) 706 2375
"Managing 4000 customer domains with BIND has been a lot like
herding cats." - Mike Batchelor, on [EMAIL PROTECTED]
Again, its unnecessary and does not add anything to the 'system' of
distributing load across servers. Distributing in a true round-robin
fashion is sufficient if the servers are of equal quality (and/or drop
connections sooner if they are not). True round-robin distribution cannot
easily be accomplished (without shared memory or a temporary file) and so
picking a random starting point is the closest that can be achieved. Using
random servers each time though, especially from a near-true random number
source, could cause one server to be hit 3 or 4 (or, technically, an
infinite number of) times in a row. There is no need to go to the expense
of randomly computing each successive server (for example:)
#1 #2 #3 #4 #5
1 3 4 1 2
2 4 1 2 3
3 1 3 4
4 4 1
1 2
2
"1,3,4,1,2" were picked as my random numbers and the number of messages sent
out in a round varying ... each server was hit almost evenly:
1 xxxxxx
2 xxxxx
3 xxxx
4 xxxxx
(Disclaimer: My old stats teacher would kill me for this insufficiently long
example and lack of mathematical proof)
----- Original Message -----
From: "JuanE" <[EMAIL PROTECTED]>
> A small modification to what you have done can get rid of the problem of
> banging one server. Instead of just sampling a random starting point for
> the search, just get a random number between 1 and N (N being the number
of
> servers) and if that server is down just sample again until you find a
> server that's up. This way, you will redistribute the load randomly to all
> servers and not just the next one on the list.
Here's an old post I forgot to cc the list to. Thanks to James for
forwarding it.
----------Forwarded message ----------
From: [EMAIL PROTECTED]
To: 5@ @@ay4it.com
Date: 2 Aug 2000 14:44:14 -0000
Subject: failure notice
>
> Hi. This is the qmail-send program at muncher.math.uic.edu.
> I'm afraid I wasn't able to deliver your message to the following addresses.
> This is a permanent error; I've given up. Sorry it didn't work out.
>
> <[EMAIL PROTECTED]>:
> I do not accept blind carbon copies. Please use To or Cc.
>
> --- Below this line is a copy of the message.
>
> Return-Path: <[EMAIL PROTECTED]>
> Received: (qmail 9238 invoked from network); 2 Aug 2000 14:44:13 -0000
> Received: from maybe.domainregistry.ie (193.1.142.4)
> by muncher.math.uic.edu with SMTP; 2 Aug 2000 14:44:13 -0000
> Received: (qmail 8695 invoked by uid 1000); 2 Aug 2000 14:43:49 -0000
> Resent-Message-ID: <[EMAIL PROTECTED]>
> Delivered-To: [EMAIL PROTECTED]
> Received: (qmail 8382 invoked from network); 2 Aug 2000 14:13:26 -0000
> Received: from drno-int.domainregistry.ie (HELO smtp.domainregistry.ie)
>(?hIO52HHLEyWUw4SKuCTAep5ydEYRa8/[email protected])
> by maybe.domainregistry.ie with SMTP; 2 Aug 2000 14:13:26 -0000
> Received: (qmail 31881 invoked by uid 501); 2 Aug 2000 14:13:25 -0000
> Delivered-To: [EMAIL PROTECTED]
> Received: (qmail 31878 invoked from network); 2 Aug 2000 14:13:25 -0000
> Received: from unknown (HELO fwder.com) (216.42.24.88)
> by smtp0.domainregistry.ie with SMTP; 2 Aug 2000 14:13:25 -0000
> Received: (qmail 9113 invoked by uid 99); 2 Aug 2000 14:05:57 -0000
> Message-ID: <[EMAIL PROTECTED]>
> References: <[EMAIL PROTECTED]>
> <[EMAIL PROTECTED]>
> <[EMAIL PROTECTED]>
> In-Reply-To: <[EMAIL PROTECTED]>
> From: "JuanE" <[EMAIL PROTECTED]>
> To: James Raftery <[EMAIL PROTECTED]>
> Subject: Re: updated load balancing qmail-qmqpc.c mods
> Date: Wed, 02 Aug 2000 14:05:57 GMT
> Mime-Version: 1.0
> Content-Type: text/plain; charset="us-ascii"
> Content-Transfer-Encoding: 7bit
> Resent-From: [EMAIL PROTECTED]
> Resent-Date: Wed, 2 Aug 2000 15:43:49 +0100
> Resent-To: [EMAIL PROTECTED]
>
>
> Not quite. The possibility does exist, but the likelihood os not the same
> as choosing an even spread. If only one server is down, you will pick it
> twice with probability 1/N^2, 3 times with p = 1/N^3, etc...
>
> The probability that you deliver the message in 2 or less attempts is
> (N-1)/N + (N-1)/N^2. So the odds of distributing the load evenly is much
> greater than acuatlly polling the same down server twice.
>
> But, you bring up a valid point. If you want to absolutely avoid the
> possibility of polling the same down server twice, you need to keep track
> of the servers you poll. When you sample a random number, just check if it
> already came up and skip it. This can easily be accomplished with flags. If
> you have less than 32 servers (or is it 16, i don't remember), you can just
> use an integer (or long?) and use every bit as a flag servers you polled.
> If the bit for the server you are going to poll is on, it means you already
> polled it, so you should skip it and get a random sample again.
>
> JES
>
> James Raftery writes:
>
> > On Wed, Aug 02, 2000 at 11:13:22AM +0000, JuanE wrote:
> > > This way, you will redistribute the load randomly to all
> > > servers and not just the next one on the list.
> >
> > Not quite. If choose truly randomly the liklehood of you choosing the
> > same server /every/ time is the same as the liklihood of choosing a
> > nice even spread across the servers. This isn't what you want.
> >
> > Regards,
> >
> > james
> > --
> > James Raftery (JBR54) - Programmer Hostmaster - IE TLD Hostmaster
> > IE Domain Registry - www.domainregistry.ie - (+353 1) 706 2375
> > "Managing 4000 customer domains with BIND has been a lot like
> > herding cats." - Mike Batchelor, on [EMAIL PROTECTED]
> >
>
>
>
Even if one server is getting hit multiple times in a row, I don't think it
will matter. I put in 10 servers in my qmqpservers file and some well
placed printf's in qmqpc and ran it 100,000 times several times through.
Each server was getting picked about 10% of the time +/- 1% or so. If we're
sending millions of messages, this +/- 1% isn't going to make a significant
difference.
I could just use the following code to shuffle the array instead of just
rotating it:
void Shuffle(void *array, int n, size_t size)
{
int f,g;
char *temp,*p;
temp=malloc(size);
p=array;
for(f=0;f<n;f++) {
g=random(f);
memcpy(temp,&p[g*size]);
memcpy(&p[g*size],&p[f*size]);
memcpy(&p[f*size],temp);
}
free(temp);
}
-----Original Message-----
From: Michael T. Babcock [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, August 02, 2000 8:58 AM
To: JuanE; [EMAIL PROTECTED]
Subject: Re: updated load balancing qmail-qmqpc.c mods
Again, its unnecessary and does not add anything to the 'system' of
distributing load across servers. Distributing in a true round-robin
fashion is sufficient if the servers are of equal quality (and/or drop
connections sooner if they are not). True round-robin distribution cannot
easily be accomplished (without shared memory or a temporary file) and so
picking a random starting point is the closest that can be achieved. Using
random servers each time though, especially from a near-true random number
source, could cause one server to be hit 3 or 4 (or, technically, an
infinite number of) times in a row. There is no need to go to the expense
of randomly computing each successive server (for example:)
#1 #2 #3 #4 #5
1 3 4 1 2
2 4 1 2 3
3 1 3 4
4 4 1
1 2
2
"1,3,4,1,2" were picked as my random numbers and the number of messages sent
out in a round varying ... each server was hit almost evenly:
1 xxxxxx
2 xxxxx
3 xxxx
4 xxxxx
(Disclaimer: My old stats teacher would kill me for this insufficiently long
example and lack of mathematical proof)
----- Original Message -----
From: "JuanE" <[EMAIL PROTECTED]>
> A small modification to what you have done can get rid of the problem of
> banging one server. Instead of just sampling a random starting point for
> the search, just get a random number between 1 and N (N being the number
of
> servers) and if that server is down just sample again until you find a
> server that's up. This way, you will redistribute the load randomly to all
> servers and not just the next one on the list.
Re-read my point: its unnecessary. I didn't say it wouldn't work. I said
the CPU use of doing it this way was unnecessary over a simpler round-robin
approach (After picking an initial random server).
Note: I think using an array of pointers to server addresses would allow you
to do your rotations a lot faster. Think "point quicksort".
----- Original Message -----
From: "Austad, Jay" <[EMAIL PROTECTED]>
> Even if one server is getting hit multiple times in a row, I don't think
it
> will matter. I put in 10 servers in my qmqpservers file and some well
> placed printf's in qmqpc and ran it 100,000 times several times through.
> Each server was getting picked about 10% of the time +/- 1% or so. If
we're
> sending millions of messages, this +/- 1% isn't going to make a
significant
> difference.
I agree with you, I forgot to mention that, sorry. I didn't have enough
Mountain Dew yet. :)
-----Original Message-----
From: Michael T. Babcock [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, August 02, 2000 10:44 AM
To: Austad, Jay; [EMAIL PROTECTED]
Subject: Re: updated load balancing qmail-qmqpc.c mods
Re-read my point: its unnecessary. I didn't say it wouldn't work. I said
the CPU use of doing it this way was unnecessary over a simpler round-robin
approach (After picking an initial random server).
Note: I think using an array of pointers to server addresses would allow you
to do your rotations a lot faster. Think "point quicksort".
----- Original Message -----
From: "Austad, Jay" <[EMAIL PROTECTED]>
> Even if one server is getting hit multiple times in a row, I don't think
it
> will matter. I put in 10 servers in my qmqpservers file and some well
> placed printf's in qmqpc and ran it 100,000 times several times through.
> Each server was getting picked about 10% of the time +/- 1% or so. If
we're
> sending millions of messages, this +/- 1% isn't going to make a
significant
> difference.
I agree with you both (Jay and Michael), at least partially. I agree that
altough what Jay proposes will work, it is too much computation and that a
simpler round-robin (after picking initial position) would suffice.
My comment is that in the event of a down server, the simple round robin
will flood the next server in the chain with twice the load of the others.
Jay's solution does not do this (at a high computational cost).
What I proposed earlier is just one of *many* solutions that addresses the
flood problem at a lower computational cost.
Jay, I agree with you that selecting the same server many times in a row is
not an issue. This is guaranteed by the Law of Averages (for you math
wizzes out there, the Law of Large Numbers).
JES
Austad, Jay writes:
> I agree with you, I forgot to mention that, sorry. I didn't have enough
> Mountain Dew yet. :)
>
> -----Original Message-----
> From: Michael T. Babcock [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, August 02, 2000 10:44 AM
> To: Austad, Jay; [EMAIL PROTECTED]
> Subject: Re: updated load balancing qmail-qmqpc.c mods
>
>
> Re-read my point: its unnecessary. I didn't say it wouldn't work. I said
> the CPU use of doing it this way was unnecessary over a simpler round-robin
> approach (After picking an initial random server).
>
> Note: I think using an array of pointers to server addresses would allow you
> to do your rotations a lot faster. Think "point quicksort".
>
> ----- Original Message -----
> From: "Austad, Jay" <[EMAIL PROTECTED]>
>
>
> > Even if one server is getting hit multiple times in a row, I don't think
> it
> > will matter. I put in 10 servers in my qmqpservers file and some well
> > placed printf's in qmqpc and ran it 100,000 times several times through.
> > Each server was getting picked about 10% of the time +/- 1% or so. If
> we're
> > sending millions of messages, this +/- 1% isn't going to make a
> significant
> > difference.
>
JuanE <[EMAIL PROTECTED]> writes on 2 August 2000 at 16:37:36 GMT
>
> I agree with you both (Jay and Michael), at least partially. I agree that
> altough what Jay proposes will work, it is too much computation and that a
> simpler round-robin (after picking initial position) would suffice.
>
> My comment is that in the event of a down server, the simple round robin
> will flood the next server in the chain with twice the load of the others.
> Jay's solution does not do this (at a high computational cost).
>
> What I proposed earlier is just one of *many* solutions that addresses the
> flood problem at a lower computational cost.
>
> Jay, I agree with you that selecting the same server many times in a row is
> not an issue. This is guaranteed by the Law of Averages (for you math
> wizzes out there, the Law of Large Numbers).
Sounds like making repeated random picks is the way to go.
If no server is down, your one random pick will handle the mail (same
cost as picking random starting point for round-robin).
If a server is down *and you hit it*, you pay the cost of a second
random pick. This is slightly expensive, but you only pay it when
you need it. It's cheaper in elapsed time than trying the next server
and having it refuse the connection due to overload, for example. And
it spreads the load more evenly.
On the third hand, if the servers can manage their incoming
connections intelligently (say with tcpserver :-) ), the one after a
down one in a round-robin, while it will get hit a lot, can refuse
some of the connections, which will then go on to the next after it.
So you aren't really constrained to running all your servers at less
than 50% capacity, and the one after the down one won't actually melt
down.
--
Photos: http://dd-b.lighthunters.net/ Minicon: http://www.mnstf.org/minicon
Bookworms: http://ouroboros.demesne.com/ SF: http://www.dd-b.net/dd-b
David Dyer-Bennet / Welcome to the future! / [EMAIL PROTECTED]
David Dyer-Bennet writes:
> JuanE <[EMAIL PROTECTED]> writes on 2 August 2000 at 16:37:36 GMT
> >
> > I agree with you both (Jay and Michael), at least partially. I agree that
> > altough what Jay proposes will work, it is too much computation and that a
> > simpler round-robin (after picking initial position) would suffice.
> >
> > My comment is that in the event of a down server, the simple round robin
> > will flood the next server in the chain with twice the load of the others.
> > Jay's solution does not do this (at a high computational cost).
> >
> > What I proposed earlier is just one of *many* solutions that addresses the
> > flood problem at a lower computational cost.
> >
> > Jay, I agree with you that selecting the same server many times in a row is
> > not an issue. This is guaranteed by the Law of Averages (for you math
> > wizzes out there, the Law of Large Numbers).
>
> Sounds like making repeated random picks is the way to go.
>
> If no server is down, your one random pick will handle the mail (same
> cost as picking random starting point for round-robin).
>
> If a server is down *and you hit it*, you pay the cost of a second
> random pick. This is slightly expensive, but you only pay it when
> you need it. It's cheaper in elapsed time than trying the next server
> and having it refuse the connection due to overload, for example. And
> it spreads the load more evenly.
>
> On the third hand, if the servers can manage their incoming
> connections intelligently (say with tcpserver :-) ), the one after a
> down one in a round-robin, while it will get hit a lot, can refuse
> some of the connections, which will then go on to the next after it.
> So you aren't really constrained to running all your servers at less
> than 50% capacity, and the one after the down one won't actually melt
> down.
> --
> Photos: http://dd-b.lighthunters.net/ Minicon: http://www.mnstf.org/minicon
> Bookworms: http://ouroboros.demesne.com/ SF: http://www.dd-b.net/dd-b
> David Dyer-Bennet / Welcome to the future! / [EMAIL PROTECTED]
>
I did not think of that. Good suggestion.
It seems like it would be a good compropmise if you can take your down
server out of the rotation relatively quickly. If not, then you'll waste
considerable time polling the busy server (and consequently having your
connections rejected by tcpserver) while all other servers are breezing at
50% load.
JES
Since this seems to apply to everyone ...
I don't have a problem with polling the down server. Its relatively simple
to ping or attempt a connection to a server with a short timeout. If it
times out, move on. Otherwise, you don't know when that server comes back
up.
Your random distribution was the only thing I was commenting on. Your logic
of whether to 'keep polling' a down server has _nothing_ to do with whether
you use totally random selection or random select then round-robin.
Nothing!
That said, the best way to accomplish this well would be:
Store servers in an array of structs:
struct server {
ip_addr address;
int down = 0;
int try = 0;
}
struct server * servers; // malloc them ...
1) pick random start # from 0 to (num_servers - 1)
2) if servers[this_server].down > 0 then:
- servers[this_server].try++;
- if (try < down), increment, go to (2)
- else, try = 0 ...
3) connect to server
- if failed:
- servers[this_server].down++;
4) (success) increment start, set server.down to 0.
5) if another message, increment, go to (2)
That's off the top of my head -- but it would allow for linear (not
exponential) decrease of use of down servers. Over time, if a server was
down for a long time, it would stop being polled for a connection as often.
----- Original Message -----
From: "David Dyer-Bennet" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, August 02, 2000 1:03 PM
Subject: Re: updated load balancing qmail-qmqpc.c mods
> JuanE <[EMAIL PROTECTED]> writes on 2 August 2000 at 16:37:36 GMT
> >
> > I agree with you both (Jay and Michael), at least partially. I agree
that
> > altough what Jay proposes will work, it is too much computation and
that a
> > simpler round-robin (after picking initial position) would suffice.
> >
> > My comment is that in the event of a down server, the simple round
robin
> > will flood the next server in the chain with twice the load of the
others.
> > Jay's solution does not do this (at a high computational cost).
> >
> > What I proposed earlier is just one of *many* solutions that addresses
the
> > flood problem at a lower computational cost.
> >
> > Jay, I agree with you that selecting the same server many times in a
row is
> > not an issue. This is guaranteed by the Law of Averages (for you math
> > wizzes out there, the Law of Large Numbers).
>
> Sounds like making repeated random picks is the way to go.
>
> If no server is down, your one random pick will handle the mail (same
> cost as picking random starting point for round-robin).
>
> If a server is down *and you hit it*, you pay the cost of a second
> random pick. This is slightly expensive, but you only pay it when
> you need it. It's cheaper in elapsed time than trying the next server
> and having it refuse the connection due to overload, for example. And
> it spreads the load more evenly.
>
> On the third hand, if the servers can manage their incoming
> connections intelligently (say with tcpserver :-) ), the one after a
> down one in a round-robin, while it will get hit a lot, can refuse
> some of the connections, which will then go on to the next after it.
> So you aren't really constrained to running all your servers at less
> than 50% capacity, and the one after the down one won't actually melt
> down.
> --
> Photos: http://dd-b.lighthunters.net/ Minicon:
http://www.mnstf.org/minicon
> Bookworms: http://ouroboros.demesne.com/ SF: http://www.dd-b.net/dd-b
> David Dyer-Bennet / Welcome to the future! / [EMAIL PROTECTED]
>
>
>
My pseudo code was supposed to infer that one would re-select (randomly, if
you wish) a server at a certain % of the time, based on how many times it
had been polled and turned out to be down. Simply replace my first
(increment, go to #2) with (go to #1).
----- Original Message -----
From: "JuanE" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, August 02, 2000 12:37 PM
Subject: Re: updated load balancing qmail-qmqpc.c mods
>
> I agree with you both (Jay and Michael), at least partially. I agree that
> altough what Jay proposes will work, it is too much computation and that a
> simpler round-robin (after picking initial position) would suffice.
>
> My comment is that in the event of a down server, the simple round robin
> will flood the next server in the chain with twice the load of the others.
> Jay's solution does not do this (at a high computational cost).
Warning: This may be a stupid idea...
Could you take the number of milliseconds in the timeb stucture,
modulo <number or servers>, and use that as the first server to
start from? That could be done with a simple AND of a short,
and I think would provide a nice even spread across the servers.
And it doesn't require state (i.e. which server was used last time).
Just an idea.
I may realize how dumb it is after some sleep.
G'night all.
Eric
On 2 Aug 2000, Frank D. Cringle wrote:
> Generating a random permutation algorithmically is not too easy.
Oh really?
int i, j, x;
int a[N];
for (i = 0; i < N; ++i)
a[i] = i;
for (i = N - 1; i > 0; --i) {
j = random(i);
x = a[i]; a[i] = a[j]; a[j] = x;
}
where random(i) is a "good enough" PRNG generating an integer between
0 and i - 1 with the uniform distribution. The result is in a[].
The proof of correctness (i.e. that the aformentioned piece of code
generates a pseudorandomly chosen permutation of (0, ..., N-1) with
the uniform distribution) is left as an exercise for the reader.
--Pavel Kankovsky aka Peak [ Boycott Microsoft--http://www.vcnet.com/bms ]
"Resistance is futile. Open your source code and prepare for assimilation."
Hi,
Some rather basic questions....
How do I set the $RBLSMTPD environment variable in order for rblsmtpd to
block incoming rbl mails?
Does rblsmtpd need it's own daemon or can it be integrated with the smtpd
daemon if so how?
Thanks
AC
www.qmail.org
download ucspi-tcp-0.88 its included in the package... docs are also on
the site.
Paul Farber
Farber Technology
[EMAIL PROTECTED]
Ph 570-628-5303
Fax 570-628-5545
On Wed, 2 Aug 2000, Slider wrote:
>
> Hi,
>
> Some rather basic questions....
>
> How do I set the $RBLSMTPD environment variable in order for rblsmtpd to
> block incoming rbl mails?
> Does rblsmtpd need it's own daemon or can it be integrated with the smtpd
> daemon if so how?
>
> Thanks
>
> AC
>
>
See 'man rblsmtpd'. Briefly, you don't set the var normally. If the var
is set, but empty, rblsmtpd won't block the mail in any case. If the var
is set to an actual value, it will block the mail. You can set the var
in your tcp.smtp CDB file like so:
63.88.133.:allow,RBLSMTPD="-Yesmail email is not wanted here"
The 'allow' is misleading. It says to allow the TCP connection, but not
necessarily to allow the mail. The $RBLSMTPD var being set
tells rblsmtpd to reject the mail.
HTH,
jon
On Wed, Aug 02, 2000 at 12:31:21PM +0100, Slider wrote:
>
> Hi,
>
> Some rather basic questions....
>
> How do I set the $RBLSMTPD environment variable in order for rblsmtpd to
> block incoming rbl mails?
> Does rblsmtpd need it's own daemon or can it be integrated with the smtpd
> daemon if so how?
>
> Thanks
>
> AC
>
Jon Rust wrote:
>
> See 'man rblsmtpd'.
:-) No man page for rblsmtpd, at least on my 6-month old package.
Docs are actually on the rblsmtpd download page.
[snip.]
> > Some rather basic questions....
> >
> > How do I set the $RBLSMTPD environment variable in order for rblsmtpd to
> > block incoming rbl mails?
> > Does rblsmtpd need it's own daemon or can it be integrated with the smtpd
> > daemon if so how?
Hi !
Running Debian, I done the "build-rblsmtpd" which debian implements in a
DEB rblsmtpd-source package. The only problem is that the
"build-rblsmtpd" part exits with a rather nice error. These are the
lines it does :
Press ENTER to continue...
make
make[1]: Entering directory `/tmp/rblsmtpd/rblsmtpd-0.70�
nroff -man rblsmtpd.8 > rblsmtpd.0
nroff -man antirbl.8 > antirbl.0
./compile rblsmtpd.c
make[1]: execvp: ./compile: Permission Denied
make[1]: *** [rblsmtpd.o] Error 127
make[1]: Leaving directory `/tmp/rblsmtpd/rblsmtpd-0.70�
make: *** [build] Error 2
I don�t know if anyone has come across this problem before, or if I have
missed a step. Any ideas ?
Thanks, Neil
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 2 Aug 00, at 14:08, Neil D. Roberts wrote:
> make
> make[1]: Entering directory `/tmp/rblsmtpd/rblsmtpd-0.70´
[snip]
> ./compile rblsmtpd.c
> make[1]: execvp: ./compile: Permission Denied
[snip]
> I don´t know if anyone has come across this problem before, or if I
> have missed a step. Any ideas ?
Two possibilities:
1. The source package is broken and "compile" is not marked as
executable.
2. Your /tmp directory is mounted as "noexec".
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60
Comment: http://community.wow.net/grt/qdpgp.html
iQA/AwUBOYgEulMwP8g7qbw/EQKEPgCgu16heeXY1O0bU1mRJ8UmU6dfzJIAniDc
ZrxwEuh4JtgHc4qMUtqUmZ1i
=jd5l
-----END PGP SIGNATURE-----
On Wed, Aug 02, 2000 at 02:08:48PM +0200, Neil D. Roberts wrote:
> Hi !
>
> Running Debian, I done the "build-rblsmtpd" which debian implements in a
> DEB rblsmtpd-source package. The only problem is that the
> "build-rblsmtpd" part exits with a rather nice error. These are the
> lines it does :
>
> Press ENTER to continue...
> make
> make[1]: Entering directory `/tmp/rblsmtpd/rblsmtpd-0.70�
> nroff -man rblsmtpd.8 > rblsmtpd.0
> nroff -man antirbl.8 > antirbl.0
> ./compile rblsmtpd.c
> make[1]: execvp: ./compile: Permission Denied
> make[1]: *** [rblsmtpd.o] Error 127
> make[1]: Leaving directory `/tmp/rblsmtpd/rblsmtpd-0.70�
> make: *** [build] Error 2
>
> I don�t know if anyone has come across this problem before, or if I have
> missed a step. Any ideas ?
>
You have wrong permissions on ./compile, execution not allowed. rblsmtpd is
now include in ucspi-tcp-0.88 . You should use this version and build from
source.
Gerrit.
--
[EMAIL PROTECTED]
innominate AG
networking people
tel: +49.30.308806-0 fax: -77 web: http://innominate.de pgp: /pgp/gpa
On Tue, Aug 01, 2000 at 02:32:12PM -0400, Vincent Danen wrote:
! user ids: 400, 401, 402, 0, 403, 404, 405, 406.
Looks good.
! This is the output of ls -dln...
!
! drwxr-sr-x 2 400 400 163 Jul 30 17:50 /var/qmail/alias/
Looks good.
! Could this be it? It's owned by alias.qmail, maybe it should be
! alias.nofiles? I'm guessing because the files in alias/ are owned by
! alias.nofiles.
It doesn't matter. qmail-pw2u doesn't care about the gid of the home
directory. It only cares that the uids match.
Okay, next step. Check to see if you have /var/qmail/users/include
or /var/qmail/users/exclude files. If you have an ``include'', make
sure that ``alias'' is listed. If you have an ``exclude'', make sure
that ``alias'' is not listed.
If you have neither of those files, then I'm going to have to look
deeper...
---Chris K.
--
Chris, the Young One |_ If you can't afford a backup system, you can't
Auckland, New Zealand |_ afford to have important data on your computer.
http://cloud9.hedgee.com/ |_ ---Tracy R. Reed
PGP: 0xCCC6114E/0x706A6AAD |_
"Chris, the Young One" <[EMAIL PROTECTED]> wrote:
>On Tue, Aug 01, 2000 at 02:32:12PM -0400, Vincent Danen wrote:
>! user ids: 400, 401, 402, 0, 403, 404, 405, 406.
>
>Looks good.
>
>! This is the output of ls -dln...
>!
>! drwxr-sr-x 2 400 400 163 Jul 30 17:50 /var/qmail/alias/
>
>Looks good.
Yes and no. That's the right uid but it should display as "alias",
not "400". Sounds like either /etc/password or getpw* is broken.
-Dave
On Wed, Aug 02, 2000 at 09:12:27AM -0400, Dave Sill wrote:
! >! This is the output of ls -dln...
! >!
! >! drwxr-sr-x 2 400 400 163 Jul 30 17:50 /var/qmail/alias/
! >
! >Looks good.
!
! Yes and no. That's the right uid but it should display as "alias",
! not "400". Sounds like either /etc/password or getpw* is broken.
I said to type ``ls -dln''. -d means show directory (and not its
contents), -l means long listing, and -n means show numeric user
and group. If it showed ``alias'' instead of ``400'' I'd say ls
was broken. :-)
---Chris K.
--
Chris, the Young One |_ If you can't afford a backup system, you can't
Auckland, New Zealand |_ afford to have important data on your computer.
http://cloud9.hedgee.com/ |_ ---Tracy R. Reed
PGP: 0xCCC6114E/0x706A6AAD |_
"Chris, the Young One" <[EMAIL PROTECTED]> wrote:
>On Wed, Aug 02, 2000 at 09:12:27AM -0400, Dave Sill wrote:
>! >! This is the output of ls -dln...
>! >!
>! >! drwxr-sr-x 2 400 400 163 Jul 30 17:50 /var/qmail/alias/
>! >
>! >Looks good.
>!
>! Yes and no. That's the right uid but it should display as "alias",
>! not "400". Sounds like either /etc/password or getpw* is broken.
>
>I said to type ``ls -dln''. -d means show directory (and not its
>contents), -l means long listing, and -n means show numeric user
>and group. If it showed ``alias'' instead of ``400'' I'd say ls
>was broken. :-)
D'oh. Missed the "n". I forget: did he post ls output that showed
"alias"?
How about running qmail-pw2u under trace/truss/strace/par?
-Dave
On Wed, Aug 02, 2000 at 10:10:02AM -0400, Dave Sill wrote:
! D'oh. Missed the "n". I forget: did he post ls output that showed
! "alias"?
Not one that I remembered.
! How about running qmail-pw2u under trace/truss/strace/par?
Yeah, that's something I'd suggest if include/exclude files aren't the
obvious culprits.
---Chris K.
--
Chris, the Young One |_ Never brag about how your machines haven't been
Auckland, New Zealand |_ hacked, or your code hasn't been broken. It's
http://cloud9.hedgee.com/ |_ guaranteed to bring the wrong kind of
PGP: 0xCCC6114E/0x706A6AAD |_ attention. ---Neil Schneider
On Thu, Aug 03, 2000 at 03:28:56PM +1200, Chris, the Young One wrote:
> ! How about running qmail-pw2u under trace/truss/strace/par?
>
> Yeah, that's something I'd suggest if include/exclude files aren't the
> obvious culprits.
I think they are... going to send another email just to make sure,
but adding "alias" to include seems like it fixed the problem.
--
[EMAIL PROTECTED], OpenPGP key available on www.keyserver.net
// Danen Consulting Services www.danen.net, www.freezer-burn.org
// MandrakeSoft, Inc. www.linux-mandrake.com
Current Linux uptime: 1 hour 50 minutes.
Irwan Hadi wrote:
> At 08:45 PM 8/1/00 -0300, you wrote:
> >
> > Im setting a mailing list system that will require a VERY good
> >performance. After search for a lot of options, Ive decided to use qmail
> >because I think it is the most quick and configurable MTA.
>
> you wrong.
> postfix (Www.postfix.org) is 3 times faster then qmail.
Ive just read the postfix documentation, and I would like to know about
how secure it is. Its know that qmail is a very secure system....
Thanks,
--
_________________________
Fernando Costa de Almeida
ICQ - 72293951
On Wed, 2 Aug 2000, Fernando Almeida wrote:
> Irwan Hadi wrote:
>
> > At 08:45 PM 8/1/00 -0300, you wrote:
> > >
> > > Im setting a mailing list system that will require a VERY good
> > >performance. After search for a lot of options, Ive decided to use qmail
> > >because I think it is the most quick and configurable MTA.
> >
> > you wrong.
> > postfix (Www.postfix.org) is 3 times faster then qmail.
>
> Ive just read the postfix documentation, and I would like to know about
> how secure it is. Its know that qmail is a very secure system....
Check the bugtraq archives at securityfocus.com
Vince.
--
==========================================================================
Vince Vielhaber -- KA8CSH email: [EMAIL PROTECTED] http://www.pop4.net
128K ISDN from $22.00/mo - 56K Dialup from $16.00/mo at Pop4 Networking
Online Campground Directory http://www.camping-usa.com
Online Giftshop Superstore http://www.cloudninegifts.com
==========================================================================
At 10:24 AM 8/2/00 -0300, Fernando Almeida wrote:
> Ive just read the postfix documentation, and I would like to know about
>how secure it is. Its know that qmail is a very secure system....
qmail will not be a secure system anymore if you delete the rcpthosts file
which makes it open relay.
the same will happens to postfix, if you don't set your main.cf correctly.
So it depends on the administrator to make whether a system is secure or not.
Fernando Almeida <[EMAIL PROTECTED]> wrote:
>Irwan Hadi wrote:
>
>> At 08:45 PM 8/1/00 -0300, you wrote:
>> >
>> > Im setting a mailing list system that will require a VERY good
>> >performance. After search for a lot of options, Ive decided to use qmail
>> >because I think it is the most quick and configurable MTA.
>>
>> you wrong.
>> postfix (Www.postfix.org) is 3 times faster then qmail.
You wrong. See:
http://www.kyoto.wide.ad.jp/mta/eval1/
(Japanese text, but the graphics are labeled in English.)
> Ive just read the postfix documentation, and I would like to know about
>how secure it is.
I don't think it's quite as secure as qmail, but it's way better than
Sendmail.
-Dave
> http://www.kyoto.wide.ad.jp/mta/eval1/
Just noticed that this is now available in English:
http://www.kyoto.wide.ad.jp/mta/eval1/eindex.html
-Dave
Dave Sill <[EMAIL PROTECTED]> writes on 2 August 2000 at 10:14:56 -0400
> > http://www.kyoto.wide.ad.jp/mta/eval1/
>
> Just noticed that this is now available in English:
>
> http://www.kyoto.wide.ad.jp/mta/eval1/eindex.html
His methodology looks reasonably sound, now that I can read the
description of it. And he seems entirely aware of the shortcomings,
which leads me to trust his judgement on other matters as well.
Looks like qmail took 20 seconds and sendmail took 1750 seconds to
deliver his test load. Not surprising! (uncached case)
Also note that in the cached case postfix appears to beat qmail at
delivering all the mail, at least on one graph.
However, did people notice that sendmail actually did *fewer* DNS
queries? I had understood that for total bandwidth use, qmail won
over sendmail partly for doing less DNS traffic, but this doesn't seem
to be the case in this study.
And unfortunately he only counts syn/fin and dns packets, there's
nothing that directly records bandwidth used. I guess we can make
assumptions about data transmitted per SMTP connection to calculate
bandwidth, and not be too far off. His graphs seem to show all
mailers making essentially the same number of SMTP connections, so it
sounds like his address mix doesn't allow for much multi-rcpt.
(postfix took 30 seconds, exim 500, zmailer I can't tell. Am I
reading the graphs wrong? Zmailer shows increasing count of DNS
queries off to the end of the map, but no increase in SMTP syn or
fin. Now I'm confused.)
--
Photos: http://dd-b.lighthunters.net/ Minicon: http://www.mnstf.org/minicon
Bookworms: http://ouroboros.demesne.com/ SF: http://www.dd-b.net/dd-b
David Dyer-Bennet / Welcome to the future! / [EMAIL PROTECTED]
Fernando Almeida <[EMAIL PROTECTED]> wrote:
> Im wondering what can I change to improve the performance of my
>mailing list, I already read the documentation and found a lot of thinks
>like the number of paralell proccess and other things.
> I really need a great performance in this mailing list, so I would
>like to know some tips and the best mailing list manager to use. I would
>like to know also some statistics of performance in mailing list using
>qmail....
Use dnscache (from djbdns).
Can you tell about your lists and the configuration of the hardware
that will be serving them? Does the h/w exist, or do you have a
budget? How big are the lists? Where are the subscribers? How big are
the messages? How quickly do you need to deliver them? Will you be
using ezmlm or some other list manager?
-Dave
Dave Sill wrote:
> Fernando Almeida <[EMAIL PROTECTED]> wrote:
>
> > Im wondering what can I change to improve the performance of my
> >mailing list, I already read the documentation and found a lot of thinks
> >like the number of paralell proccess and other things.
> > I really need a great performance in this mailing list, so I would
> >like to know some tips and the best mailing list manager to use. I would
> >like to know also some statistics of performance in mailing list using
> >qmail....
>
> Use dnscache (from djbdns).
Good.
>
>
> Can you tell about your lists and the configuration of the hardware
> that will be serving them? Does the h/w exist, or do you have a
> budget? How big are the lists? Where are the subscribers? How big are
> the messages? How quickly do you need to deliver them? Will you be
> using ezmlm or some other list manager?
>
> -Dave
I dont know yet the hardware configuration of the machine,
unfortunally.... The lists will be very big, let me explain better: we will
make a newsletter system, with msgs being sent to every our customer,
perhaps several messages per day.
I dont think the messages will be big, just a few lines of text, but
I need that the messages cames to its destination in a few amount of time,
because they will be some kind of "last news".
Im searching the mailing list manager to use, but Im thought in use
ezmlm... Any sugestions?
Thanks!
--
_________________________
Fernando Costa de Almeida
ICQ - 72293951
At 10:14 AM 8/2/00 -0400, Dave Sill wrote:
> > http://www.kyoto.wide.ad.jp/mta/eval1/
>
>Just noticed that this is now available in English:
>
> http://www.kyoto.wide.ad.jp/mta/eval1/eindex.html
>
>-Dave
I saw, at least at evaluation 3, postfix beat qmail ;)
On 02-Aug-2000, Dave Sill wrote:
> I don't think it's quite as secure as qmail
Would you care to shed some light on why you don't think so? Not to
ignite flames but for informational purposes. I use both qmail and
postfix and it is very interesting to understand not just the
strengths, but also the weaknesses of both.
TIA.
Ronny
Ronny Haryanto <[EMAIL PROTECTED]> wrote:
>On 02-Aug-2000, Dave Sill wrote:
>> I don't think it's quite as secure as qmail
>
>Would you care to shed some light on why you don't think so?
Two reasons:
1) Postfix only uses a single uid. qmail uses six.
2) Wietse's code is buggier than Dan's. Check the historical record.
(To be fair, *everyone's* code is buggier than Dan's. :-)
Postfix *can* be configured to run chrooted, which is a plus for
it. But that doesn't outweigh the above.
-Dave
On 02-Aug-2000, Dave Sill wrote:
> 1) Postfix only uses a single uid. qmail uses six.
Why is using more than one uid better? What sort of security problem
would using one uid potentially pose?
> 2) Wietse's code is buggier than Dan's. Check the historical record.
> (To be fair, *everyone's* code is buggier than Dan's. :-)
Well, IMHO, nobody's perfect, period :). Although I'm sure Dan and
Wietse are two of the very best in the field.
> Postfix *can* be configured to run chrooted, which is a plus for
> it. But that doesn't outweigh the above.
Thanks for the insight.
Ronny
The multiple UIDs provide a few failsafes, if nothing else, whereby one
broken / buggy / replaced binary can't do damage to files it doesn't own.
DJB has comments about this in the readmes, if I'm not mistaken.
----- Original Message -----
From: "Ronny Haryanto" <[EMAIL PROTECTED]>
> On 02-Aug-2000, Dave Sill wrote:
> > 1) Postfix only uses a single uid. qmail uses six.
>
> Why is using more than one uid better? What sort of security problem
> would using one uid potentially pose?
Ronny Haryanto <[EMAIL PROTECTED]> writes on 2 August 2000 at 09:35:52 -0500
> On 02-Aug-2000, Dave Sill wrote:
> > I don't think it's quite as secure as qmail
>
> Would you care to shed some light on why you don't think so? Not to
> ignite flames but for informational purposes. I use both qmail and
> postfix and it is very interesting to understand not just the
> strengths, but also the weaknesses of both.
The Postfix code is significantly larger, so if the bug rate per line
is the same, there are more bugs present.
It's less partitioned than qmail -- qmail partitions its functions
among the 7 UIDs so that the extent of damage possible of one program
is insecure is quite limited.
However, as a recently-written piece of software, designed with
security in mind, and written by somebody who is good at security,
postfix should be quite safe, as such things go.
--
Photos: http://dd-b.lighthunters.net/ Minicon: http://www.mnstf.org/minicon
Bookworms: http://ouroboros.demesne.com/ SF: http://www.dd-b.net/dd-b
David Dyer-Bennet / Welcome to the future! / [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
>>What do the qmail-send logs say? Did you look in ~alias/Mailbox for
>>bounces?
>
>There are no messages from the ~alias/Mailbox for bounces.
>About qmail-send logs, I don't know where to find the log file,
/var/log/qmail
>Now I have another problem, I cheked the
>/var/qmail/supervise/qmail-smtpd/run file and tried to edit it here
>is the file:
>
>#!/bin/sh
>QMAILDUID=`id -u qmaild`
>NOFILESGID=`id -g qmaild`
>exec /usr/local/bin/softlimit -m 4000000 \
> /usr/local/bin/tcpserver -v -p -x /etc/tcp.smtp.cdb \
> -u $QMAILDUID -g $NOFILESGID 0 smtp /var/qmail/bin/qmail-smtpd 2>&1
>
>Then I saved it and restart qmail.....
Hmm. What did you change?
>After restarting, a repeated message appears on the screen:
>
>> softlimit: fatal: unable to run : file does not exist
>
>I don't know what it means, so I just ignore it and tried testing
>mail thru smtp.
softlimit is saying it can't find /usr/local/bin/tcpserver.
>Now I can't even connect.
Of course not.
-Dave
Does anyone else experience this? We randomly and unpredictably have an
occurrance whereby Outlook (2000) is unable to POP a message from our
(qmail) pop server. When I look in the users Maildir I see that it's
always dying on the same message. So what I've been doing is moving the
file over to /tmp so that the user can continue to get the other messages.
The contents of the file look fine. And when I copy the message over to
my Maildir and POP it with fetchmail there's no problem.
--
Albert Hopkins
Sr. Systems Specialist
Dynacare Laboratories
[EMAIL PROTECTED]
I always get what I want because I always want what I get.
-Tim Bischoff
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 2 Aug 00, at 10:44, Albert Hopkins wrote:
> Does anyone else experience this? We randomly and unpredictably have
> an occurrance whereby Outlook (2000) is unable to POP a message from
> our (qmail) pop server. When I look in the users Maildir I see that
> it's always dying on the same message.
Isn't this the recent "malformed header" buffer overflow in Outlook?
Please check the Date: field whether it's malformed. If it is,
someone is intentionally sending messages to crash Outlook.
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60
Comment: http://community.wow.net/grt/qdpgp.html
iQA/AwUBOYg2Z1MwP8g7qbw/EQI7TQCgj+UTmSi/SoIXBCA2Fkh65bqD2N4AnRl1
/XN+JG8g3D81Fm6IPcuieBjs
=Lmpi
-----END PGP SIGNATURE-----
--
Petr Novotny, ANTEK CS
[EMAIL PROTECTED]
http://www.antek.cz
PGP key ID: 0x3BA9BC3F
-- Don't you know there ain't no devil there's just God when he's drunk.
[Tom Waits]
Actually, I had this happen when someone sent a mail from an old Eudora client
(Eudora Lite 3.0.1 if I remember correctly) and (I think it was) Outlook
Express
from IE4. I deemed it to be a bug in Outlook Express, removed the mail and
told
the sender not to send any more mail of that type to the recipient.
The mails had a quite large To:-header; I think that was the problem. This
should not be a problem in Outlook 2000 though..
FWIW.
Henrik.
At 05:55 PM 8/2/00 +0200, you wrote:
>On 2 Aug 00, at 10:44, Albert Hopkins wrote:
>
> > Does anyone else experience this? We randomly and unpredictably have
> > an occurrance whereby Outlook (2000) is unable to POP a message from
> > our (qmail) pop server. When I look in the users Maildir I see that
> > it's always dying on the same message.
>
>Isn't this the recent "malformed header" buffer overflow in Outlook?
>Please check the Date: field whether it's malformed. If it is,
>someone is intentionally sending messages to crash Outlook.
>
>-----BEGIN PGP SIGNATURE-----
>Version: PGP 6.0.2 -- QDPGP 2.60
>Comment: http://community.wow.net/grt/qdpgp.html
>
>iQA/AwUBOYg2Z1MwP8g7qbw/EQI7TQCgj+UTmSi/SoIXBCA2Fkh65bqD2N4AnRl1
>/XN+JG8g3D81Fm6IPcuieBjs
>=Lmpi
>-----END PGP SIGNATURE-----
>--
>Petr Novotny, ANTEK CS
>[EMAIL PROTECTED]
>http://www.antek.cz
>PGP key ID: 0x3BA9BC3F
>-- Don't you know there ain't no devil there's just God when he's drunk.
> [Tom Waits]
No the headers are fine.
On Wed, 2 Aug 2000, Petr Novotny wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 2 Aug 00, at 10:44, Albert Hopkins wrote:
>
> > Does anyone else experience this? We randomly and unpredictably have
> > an occurrance whereby Outlook (2000) is unable to POP a message from
> > our (qmail) pop server. When I look in the users Maildir I see that
> > it's always dying on the same message.
>
> Isn't this the recent "malformed header" buffer overflow in Outlook?
> Please check the Date: field whether it's malformed. If it is,
> someone is intentionally sending messages to crash Outlook.
>
> -----BEGIN PGP SIGNATURE-----
> Version: PGP 6.0.2 -- QDPGP 2.60
> Comment: http://community.wow.net/grt/qdpgp.html
>
> iQA/AwUBOYg2Z1MwP8g7qbw/EQI7TQCgj+UTmSi/SoIXBCA2Fkh65bqD2N4AnRl1
> /XN+JG8g3D81Fm6IPcuieBjs
> =Lmpi
> -----END PGP SIGNATURE-----
> --
> Petr Novotny, ANTEK CS
> [EMAIL PROTECTED]
> http://www.antek.cz
> PGP key ID: 0x3BA9BC3F
> -- Don't you know there ain't no devil there's just God when he's drunk.
> [Tom Waits]
>
--
Albert Hopkins
Sr. Systems Specialist
Dynacare Laboratories
[EMAIL PROTECTED]
The man for whom law exists -- the man of forms, the Conservative, is a tame
man.
-Henry David Thoreau
(contributed by Chris Johnston)
oMail 0.90a - 02aug2k
-----------
A PHP4 based qmail+vmailmgrd Maildomain Administration Web interface
New in this version : - complete redesign which now only need PHP4
--------------------- - no more needs for mysql and the suid perl wrapper
- support for mail domain administrator, but also
single users may login to change password, forwarders
and autoresponder.
oMail is a web front end to qmail/vmailmgrd. It can be used by domain
owners to easily administer their mail accounts without bothering the
sysadmin. Working features: manage mailboxes (pop3) and aliases,
change passwords, support for french, englich and german.
URLs:
-----
Download: http://download.sourceforge.net/oMail/omail-admin-0.90a.tar.gz
Homepage: http://omail.omnis.ch
Changelog: http://admin.omnis.ch/admin2/ChangeLog
CVSTree: http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/admin2/?cvsroot=oMail
Please note that it is currently Alpha-state software. I know
there are still bugs :)
Working demo: (domain = test.com / passwd = test)
http://admin.omnis.ch/admin2/
Know bugs: - can't turn off the autoresponder in some conditions
----------
Many thanks to Bruce & Mike for their great support!
Enjoy! Comments are of course welcome!
Olivier
--
_________________________________________________________________
Olivier Mueller - [EMAIL PROTECTED] - PGPkeyID: 0E84D2EA - Switzerland
PGP signature
A new version with some bugfixes and 100% functional autoresponder
support is available. Please tell me if it is working right on your
system!
The file would be: http://download.sourceforge.net/oMail/omail-admin-0.90.2a.tar.gz
Homepage: http://omail.omnis.ch
CVSTree: http://cvs.sourceforge.net/cgi-bin/cvsweb.cgi/admin2/?cvsroot=oMail
Regards,
Olivier
--
_________________________________________________________________
Olivier Mueller - [EMAIL PROTECTED] - PGPkeyID: 0E84D2EA - Switzerland
PGP signature
Hello,
I'm trying to setup a mail server.. I've gotten qmail
completely functional
other than the fact it won't receive mail for users
that have an "Alias"
under /var/qmail/alias/ ..
Here is an example scenario:
Domain name: edronf.com
All the users have accounts on this qmail server, but
the account names are
different than their email address [EMAIL PROTECTED]
.. Job Bob's userid
on the unix machine is "bobj" /home/bobj
Under the alias directory I created a .qmail-joe-bob
(also tried
.qmail-joe.bob) .. the contents of this alias file is
"bobj".
I restart qmail and attempt to send mail to
[EMAIL PROTECTED] and that's
when qmail reports "no mailbox here by that name".
Contents of setup files:
#### locals
Ogabot.edronf.com
Localhost
Edronf.com
#### locals
#### rcpthosts
ogabot.edronf.com
localhost
edronf.com
#### rcpthosts
#### defaultdomain
edronf.com
#### defaultdomain
#### me
ogabot.edronf.com
#### me
#### plusdomain
edronf.com
#### plusdomain
####virtualdomains
nothing at all
####virtualdomains
Does anyone see anything obvious? I appreciate your
feedback!
__________________________________________________
Do You Yahoo!?
Kick off your party with Yahoo! Invites.
http://invites.yahoo.com/
J <[EMAIL PROTECTED]> wrote:
>All the users have accounts on this qmail server, but
>the account names are
>different than their email address [EMAIL PROTECTED]
>.. Job Bob's userid
>on the unix machine is "bobj" /home/bobj
>
>Under the alias directory I created a .qmail-joe-bob
>(also tried
>.qmail-joe.bob) .. the contents of this alias file is
>"bobj".
Should be ~alias/.qmail-joe:bob. See:
http://Web.InfoAve.Net/~dsill/lwq.html#dots-in-extensions
-Dave
Hey All,
I'm in the process of installing my first qmail server and I keep
stumbling across more tools to install, like tcpserver, Qmailadmin,
daemontools, ezmlm etc. Each problem\need I have leads me to a new
"tool" and sometimes those new tools lead me other tools.
Example: I just read the README for qmailadmin which says it
requires ezmlm.
I was wondering if anyone could recommend a basic list of helpful
files\tools that I should look for, and if applicable, the order in
which I should be installing them. The server OS is Redhat Linux 6.2 and
needs to support clients using Outlook 2000 on a 50-100 user network.
Thanks,
Andy Meuse
$WITTYQUOTE
> Hey All,
>
> I'm in the process of installing my first qmail server and I keep
> stumbling across more tools to install, like tcpserver, Qmailadmin,
> daemontools, ezmlm etc. Each problem\need I have leads me to a new
> "tool" and sometimes those new tools lead me other tools.
>
> Example: I just read the README for qmailadmin which says it
> requires ezmlm.
>
> I was wondering if anyone could recommend a basic list of helpful
> files\tools that I should look for, and if applicable, the order in
> which I should be installing them. The server OS is Redhat Linux 6.2 and
> needs to support clients using Outlook 2000 on a 50-100 user network.
>
> Thanks,
> Andy Meuse
>
> $WITTYQUOTE
>
If you get the qmail tar ball, it should come with everything you need (and
probably more) expect for tcpservers. You can run qmail without tcpservers
though inetd, but I would personally recommend switching over to tcpservers
when you put it in a productive envoriment (just for the fact that inetd
gets slammed pretty hard if your running both pop and smtp from it on mail
server). If you got a decent proc inetd should be able to handle a 100
users.
Maybe you should just look first setting up the "minium requirements" of the
mail servers. For example, if users don't need aliases right now, just work
on getting the accounts working and play with the bells and whistles onces
everything else is working properly. If they don't need virtual domains,
just work with getting other "required" funcation of the mail server. Then
once it is setup you can explore other feartures like mailing lists.
Jack
Sten <[EMAIL PROTECTED]> wrote:
>>>>>> Dave Sill writes:
>
>DS> "message id" does the trick.
>
> Not any better than 'message-id'. It finds what it should,
>but also tosses in extraneous messages (like
>http://www.ornl.gov/its/archives/mailing-lists/qmail/2000/03/msg00099.html
>- 'id' doesn't even appear as a substring in that message). I'm
>guessing that 'id' (and 'message-id') are matching the header field,
>which doesn't get displayed.
No, the search engine indexes what you see. I suspect it ignores the
"id" since it's only two characters. :-(
>DS> qmail's message ID's are of the form
>DS> timestamp.pid.qmail@hostname. The only way you could get a
>DS> duplicate would be if the same process ID is reused twice within
>DS> the same second (the resolution of the timestamp) to send
>DS> different messages to the same user.
>
> Are you sure of that?
Yes. See the source code for msgidfmt in newfield.c.
>It looks like the headers may have been
>lost (I should have known better than to leave the 'From ' lines
>unaltered), but the lines which would seem to be relevant are:
>
>Received: (qmail 28582 invoked by uid 99); 1 Aug 2000 02:20:39 -0000
>Message-ID: <[EMAIL PROTECTED]>
>
> and
>
>Received: (qmail 28583 invoked by uid 99); 1 Aug 2000 02:20:39 -0000
>Message-ID: <[EMAIL PROTECTED]>
>
> I can see the timestamp in the message ID, but the process IDs
>doesn't seem to match up.
The process ID in the message ID is that of the process that created
the message ID. Since qmail is modular, many different processes may
touch a message before it's delivered. It's qmail-inject that calls
msgidfmt (indirectly).
>DS> One fix would be to increase the resolution of the timestamp to,
>DS> say, the millisecond.
>
> Would that be something that can be changed in the
>configuration, or would it require a recompile?
It would take a source code change and recompile. Doesn't look too
hard, though.
-Dave
I just got hold of the BADRCPTTO patch, but I don't know how to apply it. Is there a
HOWTO available for this?
thnx,
-->Russ Ringer
> I just got hold of the BADRCPTTO patch, but I don't know how to apply it.
Is there a HOWTO available for this?
>
> thnx,
>
> -->Russ Ringer
>
>
>
cd source_directory
patch -p1 < /tmp/patch.1.0.1
Not sure if this works or not ???
Try `man patch` or the readme file in the patch tarball
Jack
Hi,
alternatively you can you my SPAMCONTROL patch. It comes with an
(un)installation skript and an attempt to provide a good HOWTO.
htttp://www.fehcom.de/qmail_en.html
cheers.
eh.
At 10:34 2.8.2000 -0700, Russ Ringer wrote:
>I just got hold of the BADRCPTTO patch, but I don't know how to apply it.
Is there a HOWTO available for this?
>
>thnx,
>
>-->Russ Ringer
>
>
>
+-----------------------------------------------------------------------+
| fff hh http://www.fehcom.de Dr. Erwin Hoffmann |
| ff hh |
| ff eee hhhh ccc ooo mm mm mm Wiener Weg 8 |
| fff ee ee hh hh cc oo oo mmm mm mm 50858 Koeln |
| ff ee eee hh hh cc oo oo mm mm mm |
| ff eee hh hh cc oo oo mm mm mm Tel 0221 484 4923 |
| ff eeee hh hh ccc ooo mm mm mm Fax 0221 484 4924 |
+-----------------------------------------------------------------------+
Okay,
I changed the alias to:
.qmail-joe:bob
with the contents: bobj
I restarted all qmail daemons.. and unfortunatly I
still get the "no mailbox here by that name".
Any other sugguestions?
Thanks a bunch.
--- J <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I'm trying to setup a mail server.. I've gotten
> qmail
> completely functional
> other than the fact it won't receive mail for users
> that have an "Alias"
> under /var/qmail/alias/ ..
>
> Here is an example scenario:
>
> Domain name: edronf.com
>
> All the users have accounts on this qmail server,
> but
> the account names are
> different than their email address
> [EMAIL PROTECTED]
> .. Job Bob's userid
> on the unix machine is "bobj" /home/bobj
>
> Under the alias directory I created a .qmail-joe-bob
> (also tried
> .qmail-joe.bob) .. the contents of this alias file
> is
> "bobj".
>
> I restart qmail and attempt to send mail to
> [EMAIL PROTECTED] and that's
> when qmail reports "no mailbox here by that name".
>
> Contents of setup files:
>
> #### locals
> Ogabot.edronf.com
> Localhost
> Edronf.com
> #### locals
>
> #### rcpthosts
> ogabot.edronf.com
> localhost
> edronf.com
> #### rcpthosts
>
> #### defaultdomain
> edronf.com
> #### defaultdomain
>
> #### me
> ogabot.edronf.com
> #### me
>
> #### plusdomain
> edronf.com
> #### plusdomain
>
> ####virtualdomains
> nothing at all
> ####virtualdomains
>
>
> Does anyone see anything obvious? I appreciate
> your
> feedback!
>
>
> __________________________________________________
> Do You Yahoo!?
> Kick off your party with Yahoo! Invites.
> http://invites.yahoo.com/
__________________________________________________
Do You Yahoo!?
Kick off your party with Yahoo! Invites.
http://invites.yahoo.com/
|
Hi,
I know there is a web page
exits that teaches people how to use mysql with qmail for chainging password via
the web. Can anyone point me to the correct web site.
Thanks
Mark
|
On Wed, Aug 02, 2000 at 02:23:13PM -0400, Barton Grantham wrote:
> Ok. I thought that maybe it should work this way, but this raises a new
> question - If I have dnscache on 127.0.0.1, how do I point my other boxes to
> it? Or do I have to allocate two IP's for this box? Should I have them the
> other way around: dnscache serving up dns requests and doing local lookups
> to tinydns running on localhost (ie. how John Steniger has it running; see
> next msg on list) ?
The FAQ for dnscache has a section on setting up an "external dnscache"
which is what you want to do. I'm not clear on why you would need a
second IP, unless you are running tinydns on the same box. If you are,
then you do need a second IP.
Ben
--
Ben Beuchler [EMAIL PROTECTED]
MAILER-DAEMON (612) 321-9290 x101
Bitstream Underground www.bitstream.net
* Slider <[EMAIL PROTECTED]> writes:
> Please can someone drop me a hint as to the method of protecting
> myself from spam using the rblsmtpd?
Chris Hardie has written an extremely useful document called "qmail
Anti-Spam HOWTO" in which he explains various methods of thwarting
Spam. http://www.summersault.com/chris/techno/qmail/qmail-antispam.html
--
Robin S. Socha <http://socha.net/>
Question to all the qmail professionals:
when you issue the /var/qmail/bin/maildirmake command to create the Maildir
folder for a perspective user; where does store this information. The
information that I am referring to is the location of the perspective user
newly created Maildir so that when incoming mail is sent, it is delivered
correctly to the user Maildir folder.
Eddie Greer
Network Systems Engineer
University of California San Diego
Ph: (858) 534.0526
Fax: (858) 534.7758
Pager: (619) 406.1055
[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
Not sure that I understand what you're asking, but
/var/qmail/bin/maildirmake ~/Maildir will create ~/Maildir/new,
~/Maildir/tmp and ~/Maildir/cur for incoming mail.
A
----- Original Message -----
From: "Eddie Greer" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Wednesday, August 02, 2000 2:56 PM
Subject: maildirmake
> Question to all the qmail professionals:
>
> when you issue the /var/qmail/bin/maildirmake command to create the
Maildir
> folder for a perspective user; where does store this information. The
> information that I am referring to is the location of the perspective
user
> newly created Maildir so that when incoming mail is sent, it is
delivered
> correctly to the user Maildir folder.
>
> Eddie Greer
>
> Network Systems Engineer
> University of California San Diego
>
> Ph: (858) 534.0526
> Fax: (858) 534.7758
> Pager: (619) 406.1055
> [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
>
>
"qmail List" <[EMAIL PROTECTED]> writes:
> Not sure that I understand what you're asking, but
> /var/qmail/bin/maildirmake ~/Maildir will create ~/Maildir/new,
> ~/Maildir/tmp and ~/Maildir/cur for incoming mail.
>
> A
>
> ----- Original Message -----
> From: "Eddie Greer" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Cc: <[EMAIL PROTECTED]>
> Sent: Wednesday, August 02, 2000 2:56 PM
> Subject: maildirmake
>
>
> > Question to all the qmail professionals:
> >
> > when you issue the /var/qmail/bin/maildirmake command to create the
> Maildir
> > folder for a perspective user; where does store this information. The
> > information that I am referring to is the location of the perspective
> user
> > newly created Maildir so that when incoming mail is sent, it is
> delivered
> > correctly to the user Maildir folder.
> >
> > Eddie Greer
I think what is being asked is, if your default mail delivery
instructions (passed to qmail-start) are something like, "./Mailbox",
and you start creating Maildirs for users, how do you get the mail
to be delivered to the Maildirs instead of the Mailboxes?
And the answer is, there are several ways. The two I am familiar with
are:
* Put:
./Maildir/
into the .qmail file of every user who is going to be using Maildirs,
or, if your whole site is switching over, just change the default
mail delivery to be "./Maildir/", by invoking qmail-start specifying
"./Maildir/" instead of "./Mailbox".
Hope his helps,
Bryan
--
________________________________________________________________________
p l u m b d e s i g n
Bryan Ischo | Software Developer
157 chambers st ny ny 10007
p.212-285-8600 x233 f.212-285-8999
I have a moderated list that is not sending timely moderation notices. In
some cased it takes several (8 or more) hours for a moderator to get a
request. I've never had this problem before and could not find a reference
to it in the archive. Can someone offer some ideas why this is happing in
and how to correct it?
According to "top" the cpu is only using about 15% of its available
resource.
I'm using ezmlm + the idx extensions. I have several un-moderated list that
work fine.
John Burden
Hello,
does anyone know or has working the connection
from qmail to IMAP-daemon Cyrus?
I am experimenting with these two, but qmail
does not deliver mail to cyrus.
I want to use qmail as MTA and cyrus as IMAP-daemon
for all users.
TIA
Wolfgang Wagner
--
[EMAIL PROTECTED]
3.Vorsitzender B�rgernetz Allg�u
Technischer Leiter
On Wed, 2 Aug 2000, Wolfgang Wagner wrote:
> Hello,
>
> does anyone know or has working the connection
> from qmail to IMAP-daemon Cyrus?
>
> I am experimenting with these two, but qmail
> does not deliver mail to cyrus.
>
> I want to use qmail as MTA and cyrus as IMAP-daemon
> for all users.
Go to www.qmail.org. There's at least one pointer to how to set it
up, maybe more.
Vince.
--
==========================================================================
Vince Vielhaber -- KA8CSH email: [EMAIL PROTECTED] http://www.pop4.net
128K ISDN from $22.00/mo - 56K Dialup from $16.00/mo at Pop4 Networking
Online Campground Directory http://www.camping-usa.com
Online Giftshop Superstore http://www.cloudninegifts.com
==========================================================================
I have 2 mail servers. One will be the MX for two domains (as far as the
rest of the world is concerned).
When mail for domain mclachlanlister.com.au arrives I want it to be sent to
the mail server that is located in our other building.
I've currently got it set up according to 7.4 in the FAQ BUT that means I
have to create a tonne of aliases on my mailhost gateway for each user on
the other server.
I would rather that mailhost.mclachlan.com.au receive ALL mail for both
domains mclachlan.com.au and mclachlanlister.com.au and then forward ALL
mclachlanlister.com.au mail to the other server which would then deliver it
locally without the need to setup any aliases.
Is there a way of setting this up ?
TIA,
Wilson
smtproutes would probably do the trick.
man qmail-remote
Gavin
[]-----------------------------------+------------------------------------[]
| Gavin Cameron | ITworks Consulting |
| Ph : +61 3 9642 5477 | Level 8, 488 Bourke Street |
| Fax : +61 3 9642 5499 | Melbourne, Victoria |
| Email : [EMAIL PROTECTED] | Australia, 3000 |
[]-----------------------------------+------------------------------------[]
On Thu, 3 Aug 2000, Wilson Fletcher wrote:
> I have 2 mail servers. One will be the MX for two domains (as far as the
> rest of the world is concerned).
>
> When mail for domain mclachlanlister.com.au arrives I want it to be sent to
> the mail server that is located in our other building.
>
> I've currently got it set up according to 7.4 in the FAQ BUT that means I
> have to create a tonne of aliases on my mailhost gateway for each user on
> the other server.
>
> I would rather that mailhost.mclachlan.com.au receive ALL mail for both
> domains mclachlan.com.au and mclachlanlister.com.au and then forward ALL
> mclachlanlister.com.au mail to the other server which would then deliver it
> locally without the need to setup any aliases.
>
> Is there a way of setting this up ?
>
> TIA,
>
> Wilson
>
>
>
I installed ezmlm-0.53+ezmlm-idx-0.40.
Patched qmail-qmqpc as instructed in qmail-qmqpc.tar.gz
(I checked qmail-qmqpc.c, it's patched.)
Put a "DIR/qmqpservers" file in my list directory.
But ezmlm-send still uses qmail-queue instead of qmail-qmqpc
to deliver posts. The QMQP support in ezmlm-idx-0.40 does not
seem to work at all.
What was wrong?
On Don, 03 Aug 2000, Wayne Chu wrote:
> I installed ezmlm-0.53+ezmlm-idx-0.40.
> Patched qmail-qmqpc as instructed in qmail-qmqpc.tar.gz
> (I checked qmail-qmqpc.c, it's patched.)
> Put a "DIR/qmqpservers" file in my list directory.
You did replace qmail-queue by a link to qmail-qmqpc?
--
Oliver Koch Registered Linux User 163952
"The two most common things in the universe are hydrogen and
stupidity."
On Thu, Aug 03, 2000 at 11:41:19AM +0800, Wayne Chu wrote:
> I installed ezmlm-0.53+ezmlm-idx-0.40.
> Patched qmail-qmqpc as instructed in qmail-qmqpc.tar.gz
> (I checked qmail-qmqpc.c, it's patched.)
> Put a "DIR/qmqpservers" file in my list directory.
>
> But ezmlm-send still uses qmail-queue instead of qmail-qmqpc
> to deliver posts. The QMQP support in ezmlm-idx-0.40 does not
> seem to work at all.
Try to add '-Q' into ezmlm-send invocation, for example
|/path/to/ezmlm-send -Q ...
Indeed this is undocumented (yet).
Salam,
P.Y. Adi Prasaja
Seems that Hotmail will be removed from the list of big companies using
qmail uh ?
>MICROSOFT TO MOVE HOTMAIL TO WINDOWS 2000
>by Dave Murphy, [EMAIL PROTECTED]
>
>Microsoft's free email service, Hotmail, currently boasts
>30 million subscribers. But lately it's been plagued with
>outages and security problems. In a "go for broke" move,
>Microsoft will transition its more than 3,000 email servers
>from the Unix-based FreeBSD operating system (OS) to
>Microsoft's own Windows 2000.
>
>If it works it will be a great PR coup. But I'm thinking that
>you should move your email account to Yahoo!,
>Rocketmail, or another free service for a few months. I'm
>thinking there's a good chance that Hotmail will bite the
>dust a few times before it all gets sorted out.
>
>Reviews from system administrators about the stability of
>Windows 2000 are mixed at best. It's a great system to
>learn, because it's got Microsoft's imprimatur, and that
>means there's consulting work associated with it -- some
>businesses buy Microsoft products the way they used to
>by IBM's -- purely because they don't want to think
>outside the box.
>
>Microsoft has been ribbed for running it's leading free
>email system on a competitor's OS, so I figure it's a
>testosterone move: get the system running on our own
>OS.
>
>What they're missing is that FreeBSD is a great OS. It's
>been around a long time, and there's lots of technical
>folks who know how to keep it running well.
>
>And I'd think Microsoft would learn from it's earlier
>mistake. Back in '97 it tried to migrate Hotmail to
>Windows NT. The project got so royally hung up, they put
>the FreeBSD servers back online and scrapped their own
>Windows NT systems. Now, here we are three years later--
>well, let's just say I'm not willing to sell my front row
>tickets to this show.
>
>I hate to see a great company fall down in public, and I'm
>a big fan of Microsoft's applications -- they're pretty easy
>to use, and they keep business users humming along. But
>in general, most Unix servers are more stable than
>Windows NT/2000 servers. The Windows servers just
>have too much OS overhead.
>
>I've got my seat, popcorn in hand. Somebody dim the
>houselights.
>
>Call for Comments
>What do you think? Leave your comments on the
>message center: http://itrain.org/msg/
>
>References
>Hotmail: http://www.hotmail.com/
>Microsoft: http://www.microsoft.com/
>Message Center: http://itrain.org/msg/
>
>This article is posted to http://itrain.org/itinfo/2000/it000802a.html
>
>Live well, do good,
>
>--Dave Murphy
>X-From_: [EMAIL PROTECTED] Thu Aug 3 00:00:14 2000
>From: "David Dyer-Bennet" <[EMAIL PROTECTED]>
>Date: Wed, 2 Aug 2000 22:59:27 -0500 (CDT)
>To: Irwan Hadi <[EMAIL PROTECTED]>
>Subject: Re: Fwd: Microsoft to Move Hotmail to Windows 2000
>X-Mailer: VM 6.63 under Emacs 20.3.1
>
>Irwan Hadi <[EMAIL PROTECTED]> writes on 2 August 2000 at 21:40:57 -0600
> > Seems that Hotmail will be removed from the list of big companies using
> > qmail uh ?
> >
> > >MICROSOFT TO MOVE HOTMAIL TO WINDOWS 2000
>
>They've tried to move it to NT a number of times before, resulting in
>extensive outages. Maybe this time? Well, Microsoft certainly *can*
>send really large numbers of skilled people at the problem, if they
>want to badly enough. Maybe it can be made to work under 2000. If
>Microsoft can't do it, I guess nobody could!
>
> > >If it works it will be a great PR coup. But I'm thinking that
> > >you should move your email account to Yahoo!,
> > >Rocketmail, or another free service for a few months. I'm
> > >thinking there's a good chance that Hotmail will bite the
> > >dust a few times before it all gets sorted out.
>
>Sounds like *really* good advice to me; although why make it easy for
>them :-) ?
>--
>Photos: http://dd-b.lighthunters.net/ Minicon: http://www.mnstf.org/minicon
>Bookworms: http://ouroboros.demesne.com/ SF: http://www.dd-b.net/dd-b
>David Dyer-Bennet / Welcome to the future! / [EMAIL PROTECTED]
Or you could move to another qmail based Web email system.
www.acmemail.net
Sean
----- Original Message -----
From: Irwan Hadi <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, August 02, 2000 11:40 PM
Subject: Fwd: Microsoft to Move Hotmail to Windows 2000
> Seems that Hotmail will be removed from the list of big companies using
> qmail uh ?
>
> >MICROSOFT TO MOVE HOTMAIL TO WINDOWS 2000
> >by Dave Murphy, [EMAIL PROTECTED]
> >
> >Microsoft's free email service, Hotmail, currently boasts
> >30 million subscribers. But lately it's been plagued with
> >outages and security problems. In a "go for broke" move,
> >Microsoft will transition its more than 3,000 email servers
> >from the Unix-based FreeBSD operating system (OS) to
> >Microsoft's own Windows 2000.
> >
> >If it works it will be a great PR coup. But I'm thinking that
> >you should move your email account to Yahoo!,
> >Rocketmail, or another free service for a few months. I'm
> >thinking there's a good chance that Hotmail will bite the
> >dust a few times before it all gets sorted out.
> >
> >Reviews from system administrators about the stability of
> >Windows 2000 are mixed at best. It's a great system to
> >learn, because it's got Microsoft's imprimatur, and that
> >means there's consulting work associated with it -- some
> >businesses buy Microsoft products the way they used to
> >by IBM's -- purely because they don't want to think
> >outside the box.
> >
> >Microsoft has been ribbed for running it's leading free
> >email system on a competitor's OS, so I figure it's a
> >testosterone move: get the system running on our own
> >OS.
> >
> >What they're missing is that FreeBSD is a great OS. It's
> >been around a long time, and there's lots of technical
> >folks who know how to keep it running well.
> >
> >And I'd think Microsoft would learn from it's earlier
> >mistake. Back in '97 it tried to migrate Hotmail to
> >Windows NT. The project got so royally hung up, they put
> >the FreeBSD servers back online and scrapped their own
> >Windows NT systems. Now, here we are three years later--
> >well, let's just say I'm not willing to sell my front row
> >tickets to this show.
> >
> >I hate to see a great company fall down in public, and I'm
> >a big fan of Microsoft's applications -- they're pretty easy
> >to use, and they keep business users humming along. But
> >in general, most Unix servers are more stable than
> >Windows NT/2000 servers. The Windows servers just
> >have too much OS overhead.
> >
> >I've got my seat, popcorn in hand. Somebody dim the
> >houselights.
> >
> >Call for Comments
> >What do you think? Leave your comments on the
> >message center: http://itrain.org/msg/
> >
> >References
> >Hotmail: http://www.hotmail.com/
> >Microsoft: http://www.microsoft.com/
> >Message Center: http://itrain.org/msg/
> >
> >This article is posted to http://itrain.org/itinfo/2000/it000802a.html
> >
> >Live well, do good,
> >
> >--Dave Murphy
It seems that one member of this mailling list has wrong mail server setup,
because the bounce was not went back to the return path but to the sender.
Can he be unsubscribed then ?
>X-From_: [EMAIL PROTECTED] Wed Aug 2 23:48:47 2000
>X-Internal-ID: 398589B60000D532
>Date: 2 Aug 2000 20:42:31 -0700
>From: [EMAIL PROTECTED]
>Subject: Returned mail: User unknown
>To: [EMAIL PROTECTED]
>
>
>*** This message originated by GCS Client Services ***
>
>----- Delivery could not be made to the following recipients -----
>Invalid Recipient: MichaelG <[EMAIL PROTECTED]> (unrecoverable error)
>Invalid Recipient: qmail <[EMAIL PROTECTED]> (unrecoverable error)
>
>RFC822 Header may follow:
>
>X-Env-Sender: [EMAIL PROTECTED]
>X-Env-Recipient: [EMAIL PROTECTED]
>X-End-of-Envelope:
>X-Internal-ID: 398589B60000D52F
>Received: from amstr.com (162.120.128.9) by dubs0001.amstr.com (NPlex
>2.0.119) for [EMAIL PROTECTED]; Wed, 2 Aug 2000 20:42:23 -0700
>Message-ID: <[EMAIL PROTECTED]>
>Date: Wed, 02 Aug 2000 21:40:57 -0600
>From: (Irwan Hadi) <[EMAIL PROTECTED]>
>To: [EMAIL PROTECTED]
>Subject: Fwd: Microsoft to Move Hotmail to Windows 2000
Hello all,
My question is how to tell qmail not accept long list rcpt (To/Cc/Bcc) both
incoming and outgoing direction. I patched my server with tarpit patch, but
it didn't affect outgoing direction.
Regards,
Dong
Hi All
Can anybody tell me where can I find source rpm of
qmail + patches + init scripts for my RH Linux 6.2 on
i386
The FTP link at www.qmail.org did not work out for
me...
thanks in advance
- Sumith
__________________________________________________
Do You Yahoo!?
Kick off your party with Yahoo! Invites.
http://invites.yahoo.com/
When launching a program under svscan/supervise that has no port
connections, (I just want to keep it running,) what is the correct
line in the "run" script if I want to use syslog?
Would something like:
exec env - PATH="$PATH:/usr/local/bin" my_prog | \
splogger my_prog_id 3
or:
exec env - PATH="$PATH:/usr/local/bin" my_prog 2>&1 | \
splogger my_prog_id 3
work OK?
Both seem to work OK, but are they correct?
Thanks,
John
--
John Conover Tel. 408.370.2688 [EMAIL PROTECTED]
631 Lamont Ct. Cel. 408.772.7733
Campbell, CA 95008 Fax. 408.379.9602 http://www.johncon.com
On Thu, Aug 03, 2000 at 07:08:38AM -0000, John Conover wrote:
! exec env - PATH="$PATH:/usr/local/bin" my_prog 2>&1 | \
! splogger my_prog_id 3
Almost. Use the log directory instead, if you can (replace qmaill with
whatever user you want to run your log process on):
cd /service/foobar
printf '#!/bin/sh\nexec env - PATH="$PATH:/usr/local/bin" my_prog 2>&1' > run
chmod +x run
mkdir log
printf '#!/bin/sh\nexec setuidgid qmaill splogger my_prog_id 3' > log/run
chmod +x log/run
chmod +t .
If the main process dies for some reason, the logging program is kept
running ``so that no data is lost'' (according to the svscan page).
---Chris K.
--
Chris, the Young One |_ If you can't afford a backup system, you can't
Auckland, New Zealand |_ afford to have important data on your computer.
http://cloud9.hedgee.com/ |_ ---Tracy R. Reed
PGP: 0xCCC6114E/0x706A6AAD |_
hi,
I have a lot of these entries in my /var/log/syslog file. what do these
entries mean? do I have to care about them or are these messages just
'information'?
Aug 3 06:26:22 joshua qmail: 965276782.869518 warning: trouble injecting
bounce message, will try later
thanks for an answer
Joel
Hi,
I have just recently had the same problem although in my case I have to
remove alot of spam that was caught in my queue and the ever growing queue
was choking the system... I very brutally removed alot of the spammers mail
from /var/qmail/queue/mess and the others since then the problem started.
Cure came around from rebuilding the queue following the instructions on the
qmail.com page
http://qmail.plig.org/top.html#large
Or if you like:
Patches for high-volume servers (On your mirror)
Good Luck
Slider
hi,
I have a lot of these entries in my /var/log/syslog file. what do these
entries mean? do I have to care about them or are these messages just
'information'?
Aug 3 06:26:22 joshua qmail: 965276782.869518 warning: trouble injecting
bounce message, will try later
thanks for an answer
Joel
