postfix version 2.10-20120902 relay issue

[Naval saini]

I want to configure Haproxy for load balanceing of outgoing mails . for this
configuration of haproxy and postfix i have googled various artical since
last month finally i got some information from Haproxy mailling list.
To balance postfix i am using ..
HA-Proxy version 1.5-dev11 2012/06/04
postfix mail_version = 2.10-20120902

In /etc/main.cf
postscreen_upstream_proxy_protocol = haproxy
smtpd_upstream_proxy_protocol = haproxy

In /etc/haproxy.cfg

global
maxconn 4096 # Total Max Connections. This is dependent on
ulimit
nbproc 4
log 127.0.0.1 local4
defaults
mode tcp
# timeout connect 20s
# timeout client 20s
# timeout server 22s
frontend smtp_proxy 0.0.0.0:10021
mode tcp
log global
option tcplog
default_backend bk_postfix
backend bk_postfix
mode tcp
log global
option tcplog
balance roundrobin
#source 0.0.0.0 usesrc clientip
server postfix 127.0.0.1:25 send-proxy
server r23.mtp.org 74.117.56.x:25 send-proxy
server r17.mtp.org 199.71.212.x:25 send-proxy
both service haproxy and postfix running
but when i relay mails from my php application and from load balancer to
another smtp (which also have postfix 2.10) this shows error
" status=deferred (host r23.mtp.org[74.117.x.x] refused to talk to me: 421
4.3.2 No system resources)

and i am testing this all on vps server with centos 5.8
Is this configuration error or system resource related error.?
please help me thank you in advance




--
View this message in context: http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315.html
Sent from the Postfix Users mailing list archive at Nabble.com.

[Wietse Venema]

Naval saini:

> another smtp (which also have postfix 2.10) this shows error
> " status=deferred (host r23.mtp.org[74.117.x.x] refused to talk to me: 421
> 4.3.2 No system resources)

Look for warnings in the mail logfile of r23.mtp.org[74.117.x.x].

Wietse

[Naval saini]

this are the log of r23.mtp.org
/var/log/maillog

Sep 5 13:35:32 s1 postfix/smtpd[13312]: connect from unknown[199.83.x.x]
Sep 5 13:35:32 s1 postfix/smtpd[13312]: lost connection after CONNECT from
unknown[199.83.x.x]
Sep 5 13:35:32 s1 postfix/smtpd[13312]: disconnect from unknown[199.83.x.x]
Sep 5 13:35:32 s1 postfix/smtpd[12286]: warning: 199.83.x.x: hostname
unassigned.psychz.net verification failed: Name or service not known
Sep 5 13:35:32 s1 postfix/smtpd[12286]: connect from unknown[199.83.x.x]
Sep 5 13:35:32 s1 postfix/smtpd[12286]: lost connection after CONNECT from
unknown[199.83.x.x]

> ------------------------------
> If you reply to this email, your message will be added to the discussion
> below:
>
> http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50316.html
> To unsubscribe from postfix version 2.10-20120902 relay issue, click here<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=50315&code=c2FpbmluYXZhbDU4QGdtYWlsLmNvbXw1MDMxNXwyODE4NTg0MTY=>
> .
> NAML<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>




--
View this message in context: http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50322.html

[Wietse Venema]

Naval saini:

> " status=deferred (host r23.mtp.org[74.117.x.x] refused to talk to me:
> 421 4.3.2 No system resources)

Note: the above is a POSTSCREEN error reply. There is no other
program in Postfix 2.10 that responds with that text.

Naval saini:

> Sep 5 13:35:32 s1 postfix/smtpd[13312]: connect from unknown[199.83.x.x]
> Sep 5 13:35:32 s1 postfix/smtpd[13312]: lost connection after CONNECT from
> unknown[199.83.x.x]
> Sep 5 13:35:32 s1 postfix/smtpd[13312]: disconnect from unknown[199.83.x.x]
> Sep 5 13:35:32 s1 postfix/smtpd[12286]: warning: 199.83.x.x: hostname
> unassigned.psychz.net verification failed: Name or service not known
> Sep 5 13:35:32 s1 postfix/smtpd[12286]: connect from unknown[199.83.x.x]
> Sep 5 13:35:32 s1 postfix/smtpd[12286]: lost connection after CONNECT from
> unknown[199.83.x.x]

That is not POSTSCREEN logging.

I you want to be helped, provide the logfile record including the
TIME STAMP and the POSTFIX PROGRAM NAME for both sides of the same
connection.

Wietse

[Naval saini]

my load balancer /var/log/maillog file

Sep 6 16:36:04 r22 postfix/smtp[3985]: 062058B074B: to=<naval...@snet.in>,
relay=r23.mtp.org[74.117.x.x]:25, delay=5.3, delays=0.2/0.01/5.1/0,
dsn=4.3.2, status=deferred (host r23.mtp.org[74.117.x.x] refused to talk to

me: 421 4.3.2 No system resources)

Sep 6 16:37:48 r22 postfix/postscreen[3992]: warning: haproxy read: time
limit exceeded
Sep 6 16:45:38 r22 postfix/qmgr[3978]: 062058B074B: from=<ro...@r22.mtp.org>,
size=297, nrcpt=1 (queue active)
Sep 6 16:45:38 r22 postfix/smtp[5228]: connect to r23.mtp.org[74.117.x.x]:25:
Connection refused
Sep 6 16:45:39 r22 postfix/smtp[5228]: 062058B074B: to=<naval...@snet.in>,
relay=none, delay=580, delays=580/0.01/0.04/0, dsn=4.4.1, status=deferred
(connect to r23.mtp.org[74.117.x.x]:25: Connection refused)
Sep 6 16:46:51 r22 postfix/postscreen[5233]: warning: haproxy read: time
limit exceeded
Sep 6 16:48:18 r22 postfix/postfix-script[5250]: stopping the Postfix mail
system
Sep 6 16:48:18 r22 postfix/master[3549]: terminating on signal 15
Sep 6 16:48:19 r22 postfix/postfix-script[5325]: starting the Postfix mail
system
Sep 6 16:48:19 r22 postfix/master[5327]: daemon started -- version
2.10-20120902, configuration /etc/postfix
Sep 6 16:49:20 r22 postfix/pickup[5329]: 67C068B075C: uid=0 from=<root>
Sep 6 16:49:20 r22 postfix/cleanup[5398]: 67C068B075C: message-id=<
2012090612492...@r22.mtp.org>
Sep 6 16:49:20 r22 postfix/qmgr[5328]: 67C068B075C: from=<ro...@r22.mtp.org>,
size=297, nrcpt=1 (queue active)
Sep 6 16:49:20 r22 postfix/smtp[5400]: connect to r23.mtp.org[74.117.x.x]:25:
Connection refused
Sep 6 16:49:20 r22 postfix/smtp[5400]: 67C068B075C: to=<naval...@snet.in>,
relay=none, delay=0.23, delays=0.17/0.01/0.05/0, dsn=4.4.1, status=deferred
(connect to r23.mtp.org[74.117.x.x]:25: Connection refused)
Sep 6 16:50:49 r22 postfix/postscreen[5448]: warning: haproxy read: time
limit exceeded
Sep 6 16:52:47 r22 postfix/pickup[5329]: 46C908B0761: uid=0 from=<root>
Sep 6 16:52:48 r22 postfix/qmgr[5328]: 46C908B0761: removed
Sep 6 16:57:48 r22 postfix/postfix-script[5591]: stopping the Postfix mail
system
Sep 6 16:57:48 r22 postfix/master[5327]: terminating on signal 15
Sep 6 16:57:49 r22 postfix/postfix-script[5665]: starting the Postfix mail
system
Sep 6 16:57:49 r22 postfix/master[5667]: daemon started -- version
2.10-20120902, configuration /etc/postfix
Sep 6 16:57:49 r22 postfix/qmgr[5668]: 062058B074B: removed
Sep 6 16:57:49 r22 postfix/qmgr[5668]: 67C068B075C: removed
Sep 6 16:58:19 r22 postfix/pickup[5669]: 8D1CE8B0737: uid=0 from=<root>
Sep 6 16:58:19 r22 postfix/cleanup[5677]: 8D1CE8B0737: message-id=<
2012090612581...@r22.mtp.org>
Sep 6 16:58:19 r22 postfix/qmgr[5668]: 8D1CE8B0737: removed
Sep 6 17:02:18 r22 postfix/postscreen[5791]: warning: haproxy read: time
limit exceeded
Sep 6 19:05:53 r22 postfix/postscreen[11662]: warning: haproxy read: lost
connection



and r23.mtp.org /var/log/maillog

Sep 6 21:03:04 r23 spamd[1667]: logger: removing stderr method
Sep 6 21:03:09 r23 spamd[1670]: spamd: server started on port 783/tcp
(running version 3.3.1)
Sep 6 21:03:09 r23 spamd[1670]: spamd: server pid: 1670
Sep 6 21:03:09 r23 spamd[1670]: spamd: server successfully spawned child
process, pid 1671
Sep 6 21:03:09 r23 spamd[1670]: spamd: server successfully spawned child
process, pid 1672
Sep 6 21:03:09 r23 spamd[1670]: prefork: child states: IS
Sep 6 21:03:09 r23 spamd[1670]: prefork: child states: II
Sep 6 21:03:18 r23 postfix/postfix-script: starting the Postfix mail system
Sep 6 21:03:18 r23 postfix/master[1756]: daemon started -- version 2.3.3,
configuration /etc/postfix
Sep 6 21:38:16 r23 postfix/postfix-script[13489]: stopping the Postfix
mail system
Sep 6 21:38:16 r23 postfix/master[1756]: terminating on signal 15
Sep 6 21:38:16 r23 postfix/postfix-script[13563]: starting the Postfix
mail system
Sep 6 21:38:16 r23 postfix/master[13565]: daemon started -- version
2.10-20120902, configuration /etc/postfix
Sep 6 21:48:42 r23 postfix/postfix-script[13824]: stopping the Postfix
mail system
Sep 6 21:48:42 r23 postfix/master[13565]: terminating on signal 15
Sep 6 21:48:43 r23 postfix/postfix-script[13898]: starting the Postfix
mail system
Sep 6 21:48:43 r23 postfix/master[13900]: daemon started -- version
2.10-20120902, configuration /etc/postfix
Sep 6 21:58:28 r23 postfix/postfix-script[14160]: refreshing the Postfix
mail system
Sep 6 21:58:28 r23 postfix/master[13900]: reload -- version 2.10-20120902,
configuration /etc/postfix

if u need any thing else please tell me...i'll try to provide ..

> ------------------------------
> If you reply to this email, your message will be added to the discussion
> below:
>

> http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50326.html

> To unsubscribe from postfix version 2.10-20120902 relay issue, click here<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=50315&code=c2FpbmluYXZhbDU4QGdtYWlsLmNvbXw1MDMxNXwyODE4NTg0MTY=>
> .
> NAML<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>




--

View this message in context: http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50329.html

[Wietse Venema]

Naval saini:

> Sep 6 16:37:48 r22 postfix/postscreen[3992]: warning: haproxy read: time
> limit exceeded

> Sep 6 16:46:51 r22 postfix/postscreen[5233]: warning: haproxy read: time
> limit exceeded

> Sep 6 16:50:49 r22 postfix/postscreen[5448]: warning: haproxy read: time
> limit exceeded

> Sep 6 17:02:18 r22 postfix/postscreen[5791]: warning: haproxy read: time
> limit exceeded

Your load balancer does not send connection information within the
time limit (the default, "postscreen_upstream_proxy_timeout = 5s"
should be sufficient).

Perhaps the load balancer uses a different protocol than HAPROXY.

You can use "tcpdump -s 0 -w /file/name" to record a failed session
and see what protocol the load balancer actually uses.

Wietse

[Willy Tarreau]

Hi Wietse,

That's something totally possible. Naval seems to be using latest haproxy
development snapshot. I think it would help if problem reports on the
postfix ML were done with "more reliable" haproxy code than latest devel
snapshot, and vice-versa.

My non-reg tests have shown that the PROXY protocol in this version seems
to work for me but this does not mean there is no bug. A tcpdump trace is
clearly needed.

Naval, if your trace shows there is no PROXY protocol in the connection,
please try again with 1.5-dev11 and bring the issue back on the haproxy ML.

Regards,
Willy

[Naval saini]

I have did
tcpdump -s0 -w file.cap this shows

tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to
cooked socket
tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture
size 65535 bytes
191 packets captured
192 packets received by filter
0 packets dropped by kernel

when i capture for port 25 this shows

tcpdump -ni venet0 -s0 -w file.cap port 25
tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to
cooked socket
tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture
size 65535 bytes
0 packets captured
1 packets received by filter
0 packets dropped by kernel

and regarding to smtp trafic when i connect my load balancer from my php
application this shows
"* 421 4.3.2 No system resources"*
this problem start when i upgrade postfix from 2.3 to 2.10.
in older version of postfix this load balancer server was sending mails
successfully .
due to this relay error how can i put smtp load on load balancer.


On Fri, Sep 7, 2012 at 12:43 PM, Willy Tarreau <w...@1wt.eu> wrote:

> Hi,
>
> On Fri, Sep 07, 2012 at 12:40:38PM +0530, Naval saini wrote:
> > tcpdump -i venet0 -e host 199.83.x.x
> > tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back
> to
> > cooked socket
> > tcpdump: verbose output suppressed, use -v or -vv for full protocol
> decode
> > listening on venet0, link-type LINUX_SLL (Linux cooked), capture size 96
> > bytes
> > 11:07:38.787889 Out ethertype IPv4 (0x0800), length 252: r22.mtp.org.ssh
> >
> > 117.203.6.14.58397: P 836101525:836101721(196) ack 2820479482 win 17424
> (...)
>
> You need to put this into a file (-s0 -w file.cap) otherwise this is not
> exploitable at all. Also there is no smtp traffic here at all. I suggest
> you limit the capture to port 25 :
>
> tcpdump -ni veth0 -s0 -w file.cap port 25
>
> Regards,
> Willy
>
>




--
View this message in context: http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50347.html

[Wietse Venema]

Naval saini:

> and regarding to smtp trafic when i connect my load balancer from my php
> application this shows
> "* 421 4.3.2 No system resources"*

You have haproxy turned on in postscreen, but your load
balancer is not sending the haproxy protocol.

Where is the tcpdump output file?

Wietse

[Naval saini]

When mails are submited from application these parameter commented in
main.cf file
#postscreen_upstream_proxy_protocol = haproxy
#postscreen_upstream_proxy_timeout = 1s
#smtpd_upstream_proxy_protocol = haproxy

server relaying mails and and tcpdump shows

tcpdump -s0 -w file.cap

tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to
cooked socket
tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture
size 65535 bytes

294 packets captured
296 packets received by filter

0 packets dropped by kernel

tcpdump -ni venet0 -s0 -w file.cap port 25
tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to
cooked socket
tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture
size 65535 bytes

323 packets captured
324 packets received by filter

0 packets dropped by kernel

When mails are submited from application these parameter are not commented
in main.cf file
postscreen_upstream_proxy_protocol = haproxy
postscreen_upstream_proxy_timeout = 1s
smtpd_upstream_proxy_protocol = haproxy

server stops relaying of mails and and tcpdump shows
tcpdump -s0 -w file.cap

tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to
cooked socket
tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture
size 65535 bytes

41 packets captured
43 packets received by filter

0 packets dropped by kernel

tcpdump -ni venet0 -s0 -w file.cap port 25
tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to
cooked socket
tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture
size 65535 bytes

40 packets captured
41 packets received by filter

0 packets dropped by kernel

now tell me where the problem and logs of load balancer are
/var/log/mailllog
Sep 7 17:12:14 r22 postfix/smtp[7362]: 47A8B1F54156: to=<
bourn...@gmail.com>, relay=gmail-smtp-in.l.google.com[173.194.64.26]:25,
delay=1.9, delays=0.4/0/0.21/1.3, dsn=2.0.0, status=sent (250 2.0.0 OK
1347023531 b10si6583182oee.33)
Sep 7 17:12:14 r22 postfix/qmgr[6035]: 47A8B1F54156: removed
Sep 7 17:12:14 r22 postfix/cleanup[7175]: 154991F54154: message-id=<
a68a1d0a9b810e18...@test.25smtp.com>
Sep 7 17:12:14 r22 postfix/postfix-script[7386]: stopping the Postfix mail
system
Sep 7 17:12:14 r22 postfix/master[6034]: terminating on signal 15
Sep 7 17:12:30 r22 postfix/postfix-script[7461]: starting the Postfix mail
system
Sep 7 17:12:30 r22 postfix/master[7463]: daemon started -- version
2.10-20120902, configuration /etc/postfix
Sep 7 17:12:30 r22 postfix/qmgr[7464]: 1F2FF1F54153: from=<
bou...@test.25smtp.com>, size=7874, nrcpt=1 (queue active)
Sep 7 17:12:31 r22 postfix/smtp[7468]: 1F2FF1F54153: to=<monu...@gmail.com>,
relay=gmail-smtp-in.l.google.com[173.194.64.26]:25, delay=25,
delays=24/0.01/0.21/0.58, dsn=2.0.0, status=sent (250 2.0.0 OK 1347023548
b10si6586615oee.28)
Sep 7 17:12:31 r22 postfix/qmgr[7464]: 1F2FF1F54153: removed
Sep 7 17:12:31 r22 postfix/postscreen[7469]: warning: haproxy read: time
limit exceeded
Sep 7 17:13:02 r22 last message repeated 15 times
~

log of node server
/var/log/maillog

Sep 7 16:31:40 r23 postfix/master[1783]: terminating on signal 15
Sep 7 16:31:40 r23 postfix/postfix-script[2019]: starting the Postfix mail
system
Sep 7 16:31:41 r23 postfix/master[2021]: daemon started -- version
2.10-20120902, configuration /etc/postfix
Sep 7 16:55:14 r23 postfix/postfix-script[5465]: stopping the Postfix mail
system
Sep 7 16:55:14 r23 postfix/master[2021]: terminating on signal 15
Sep 7 16:55:15 r23 postfix/postfix-script[5539]: starting the Postfix mail
system
Sep 7 16:55:15 r23 postfix/master[5541]: daemon started -- version
2.10-20120902, configuration /etc/postfix

> ------------------------------
> If you reply to this email, your message will be added to the discussion
> below:
>

> http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50354.html

> To unsubscribe from postfix version 2.10-20120902 relay issue, click here<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=50315&code=c2FpbmluYXZhbDU4QGdtYWlsLmNvbXw1MDMxNXwyODE4NTg0MTY=>
> .
> NAML<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>

--
View this message in context: http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50361.html

[Wietse Venema]

Naval saini:

> When mails are submited from application these parameter are not commented
> in main.cf file
> postscreen_upstream_proxy_protocol = haproxy
> postscreen_upstream_proxy_timeout = 1s
> smtpd_upstream_proxy_protocol = haproxy
>
> server stops relaying of mails

Your first mistake is to use postscreen to submit mail from
an application.

postscreen is a ZOMBIE DETECTOR.

Wietse

[Naval saini]

Than please tell me how can i balance load of outgoing mails using haproxy and what postfix program i should use to receive mails from php application and balancing load using haproxy backhand servers.

Does relayhost parameter play any role in haproxy like it plays in DNS load balancing.

please help me.

[Wietse Venema]

Wietse:

> Your first mistake is to use postscreen to submit mail from
> an application.

Naval saini:

> Than please tell me how can i balance load of outgoing mails using haproxy
> and what postfix program i should use to receive mails from php application
> and balancing load using haproxy backhand servers.

Turn off postscreen.
http://www.postfix.org/POSTSCREEN_README.html#turnoff

Turn on smtpd_upstream_proxy_protocol.
http://www.postfix.org/postconf.5.html#smtpd_upstream_proxy_protocol

> Does relayhost parameter play any role in haproxy like it plays in DNS load
> balancing.

Your problem is RECEIVING mail. relayhost is for SENDING mail.
http://www.postfix.org/BASIC_CONFIGURATION_README.html

Wietse

[Naval saini]

As you told now i have turned off postscreen and using
smtpd_upstream_proxy_protocol = haproxy
smtpd_upstream_proxy_timeout = 5s
 in my /etc/postfix/main.cf file



and in /etc/postfix/master.cf
smtp    inet    n       -       n       -       -       smtpd -o smtpd_sasl_auth_enable=yes

and my /etc/haproxy.cfg file looks like

global
        maxconn     4096 # Total Max Connections. This is dependent on ulimit

        daemon
        nbproc      4 # Number of processing cores. Dual Dual-core Opteron is 4 cores for example.

        log 127.0.0.1 local4
defaults
        mode        tcp

        clitimeout  4s
        srvtimeout  4s

frontend smtp_proxy 0.0.0.0:10021
        mode tcp
        log global
        option tcplog
        default_backend bk_postfix
backend bk_postfix
        mode tcp
        log global
        option tcplog

        contimeout 3000
        srvtimeout 3000
        balance roundrobin  # here i am testing things only one node
        server r23.mtp.org 74.117.x.x:25 send-proxy


And Now i am submiting mails from another mail server on load balancer using sasl authentication but now it not receiving mails
and When i submit mails to load balancer /var/log/maillog showing

Sep  8 13:09:47 r22 postfix/smtpd[3170]: connect from unknown[unknown]
Sep  8 13:09:47 r22 postfix/smtpd[3170]: disconnect from unknown[unknown]
Sep  8 13:09:53 r22 postfix/smtpd[3170]: warning: haproxy read: timeout error
Sep  8 13:09:53 r22 postfix/smtpd[3170]: connect from unknown[unknown]
Sep  8 13:09:53 r22 postfix/smtpd[3170]: disconnect from unknown[unknown]
Sep  8 13:09:58 r22 postfix/smtpd[3170]: warning: haproxy read: timeout error
Sep  8 13:09:58 r22 postfix/smtpd[3170]: connect from unknown[unknown]
Sep  8 13:09:58 r22 postfix/smtpd[3170]: disconnect from unknown[unknown]
Sep  8 13:10:04 r22 postfix/smtpd[3170]: warning: haproxy read: timeout error

now please tell me where i am doing mistake ..?

[Wietse Venema]

Naval saini:

> Sep 8 13:09:47 r22 postfix/smtpd[3170]: connect from unknown[unknown]
> Sep 8 13:09:47 r22 postfix/smtpd[3170]: disconnect from unknown[unknown]
> Sep 8 13:09:53 r22 postfix/smtpd[3170]: warning: haproxy read: timeout error
> Sep 8 13:09:53 r22 postfix/smtpd[3170]: connect from unknown[unknown]
> Sep 8 13:09:53 r22 postfix/smtpd[3170]: disconnect from unknown[unknown]
> Sep 8 13:09:58 r22 postfix/smtpd[3170]: warning: haproxy read: timeout error
> Sep 8 13:09:58 r22 postfix/smtpd[3170]: connect from unknown[unknown]
> Sep 8 13:09:58 r22 postfix/smtpd[3170]: disconnect from unknown[unknown]
> Sep 8 13:10:04 r22 postfix/smtpd[3170]: warning: haproxy read: timeout error
>
> now please tell me where i am doing mistake ..?

Capture a session with:

tcpdump -s 0 -w OUTPUTFILE port 25

Make THAT OUTPUTFILE available so that Willy and I can have a look.

Then we can make progress.

Wietse

[Naval saini]

this the output of dump file tcpdump -s 0 -w file.cap port 25

tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to cooked socket
tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture size 65535 bytes

260 packets captured
261 packets received by filter

0 packets dropped by kernel

what i am thinking regarding this problem
If you check haproxy listening on port 10021
frontend smtp_proxy 0.0.0.0:10021
 according to me this means haproxy will accept client request on port 10021 and will respond by backend bk_postfix servers which are listed i have checked the log of bakend servers looks like

warning: hostname unassigned.psychz.net does not resolve to address 199.83.95.70: Name or service not known
Sep  8 17:00:59 r23 postfix/smtpd[30403]: connect from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: lost connection after CONNECT from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: disconnect from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: warning: hostname unassigned.psychz.net does not resolve to address 199.83.95.70: Name or service not known
Sep  8 17:00:59 r23 postfix/smtpd[30403]: connect from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: lost connection after CONNECT from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: disconnect from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: warning: hostname unassigned.psychz.net does not resolve to address 199.83.95.70: Name or service not known
Sep  8 17:00:59 r23 postfix/smtpd[30403]: connect from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: lost connection after CONNECT from unknown[199.83.95.70]
Sep  8 17:00:59 r23 postfix/smtpd[30403]: disconnect from unknown[199.83.95.70]

means haproxy sending proxy on bakend server.
But it not accepting mails.
And the reason for this according to me (not sure but) we are sending mails to postfix on port 25 while haproxy listening on port 10021 how it will accept mails.
is this correct. if this is the issue than please tell me on which port i have to bind frontend and on which i have to bind postfix. ?

[Wietse Venema]

Naval saini:
> *this the output of dump file tcpdump -s 0 -w file.cap port 25*

> tcpdump: WARNING: arptype 65535 not supported by libpcap - falling back to
> cooked socket
> tcpdump: listening on venet0, link-type LINUX_SLL (Linux cooked), capture
> size 65535 bytes

I asked for the OUTPUTFILE CONTENT of this command:

tcpdump -s 0 -w OUTPUTFILE port 25

In your case, the file.cap CONTENT file.cap, not the error messages.

Wietse

[Naval saini]

wietse
This is the tcpdump output file i am sending u please find the attachment.

> ------------------------------
> If you reply to this email, your message will be added to the discussion
> below:
>

> http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50405.html

> To unsubscribe from postfix version 2.10-20120902 relay issue, click here<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=unsubscribe_by_code&node=50315&code=c2FpbmluYXZhbDU4QGdtYWlsLmNvbXw1MDMxNXwyODE4NTg0MTY=>
> .
> NAML<http://postfix.1071664.n5.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>




--

View this message in context: http://postfix.1071664.n5.nabble.com/postfix-version-2-10-20120902-relay-issue-tp50315p50409.html

[Reindl Harald]

Am 08.09.2012 19:43, schrieb Naval saini:
> wietse
> This is the tcpdump output file i am sending u please find the attachment

there is NO attachment

upload large files to whatever service and post links
and please get rid of your top-posting