Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Sending email time(s)out on postfix

Status
Not open for further replies.

ISManager007

IS-IT--Management
Aug 12, 2004
5
US
We currently have postfix version 1.1.7-2. Over the last 3 or so years, since installing this postfix server every couple of months we expierence a problem trying to send email where it sometimes takes up to 2 minutes to send the message(from client to postfix mail server) or times out completely.

I'm completely stumped. I've done tcpdump's when we expierence the problem, changed email clients(goldmine/pegasus/outlook) and get the same results. Restart the postfix process, talked with Cisco about problems with the PIX(I have version 6.3(3)135), they say that the pix issue was resolved in version 6.2. I've disabled the smtp fixup protocol anyways... When I try to do a telnet to the box via smtp it just hangs, then eventually connects after a considerable amount of time. When I do a nsookup -sil clientname.AD.domain.com it finds the right client instantley. If I do the same nslookup for redhat for example the I get the response back instantley. One strange point is that on a different PIX segment, if I try to do a telnet to the STATIC address of the mail server that I'm static'ing in from one of my outside segemnts, I get a 220 response right away. If I then retry to telnet via smtp on my inside segment to the mail server(inside segment) it continues to hang.

I don't know postfix very well, One of my sys admins from a while back set this installation up so I'm a bit limited when it comes to trouble shooting this.

Wierdly enough, the problem usually goes away in a couple of days?????

Any nudge in the right direction would be appreciated

Config:

alias_database = hash:/etc/postfix/aliases
alias_maps = hash:/etc/postfix/aliases
command_directory = /usr/sbin
config_directory = /etc/postfix
daemon_directory = /usr/libexec/postfix
mailbox_size_limit = 512000000
mailq_path = /usr/bin/mailq.postfix
manpage_directory = /usr/share/man
maps_rbl_domains = relays.ordb.org, dnsbl.njabl.org, spam.dnsrbl.net, sbl-xbl.spamhaus.org, cbl.abuseat.org
message_size_limit = 40960000
mydestination = $myhostname
mynetworks = %%%.%.%%.%/24, 192.168.3.0/24, 192.168.1.0/24, 127.0.0.0/8
myorigin = $myhostname
newaliases_path = /usr/bin/newaliases.postfix
qmgr_message_recipient_limit = 1000
readme_directory = /etc/postfix/README_FILES
sample_directory = /usr/share/doc/postfix-1.1.7/samples
sendmail_path = /usr/sbin/sendmail.postfix
setgid_group = postdrop
smtpd_helo_required = yes
smtpd_sender_restrictions = permit_mynetworks, regexp:/etc/postfix/access, reject_unknown_sender_domain, reject_maps_rbl, permit
virtual_maps = hash:/etc/postfix/virtual

Tcpdump failure:
[root@%NAME% root]# tcpdump host 192.168.3.244
tcpdump: listening on eth0

16:34:58.324629 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: S 4100659813:4100659813(0) win 64512 <mss 1460,nop,nop,sackOK> (DF)
16:34:58.459698 192.168.3.244 > 225.0.0.1: igmp v2 report 225.0.0.1 [ttl 1]
16:35:01.548563 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: S 4100659813:4100659813(0) win 64512 <mss 1460,nop,nop,sackOK> (DF)
16:35:01.548605 %NAME%.%domain%.com.smtp > 192.168.3.244.2159: S 3244244911:3244244911(0) ack 4100659814 win 5840 <mss 1460,nop,nop,sackOK> (DF)
16:35:01.548725 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: . ack 1 win 64512 (DF)
16:35:05.315241 %NAME%.%domain%.com.smtp > 192.168.3.244.2159: S 3244244911:3244244911(0) ack 4100659814 win 5840 <mss 1460,nop,nop,sackOK> (DF)
16:35:05.315437 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: . ack 1 win 64512 (DF)
16:35:11.315242 %NAME%.%domain%.com.smtp > 192.168.3.244.2159: S 3244244911:3244244911(0) ack 4100659814 win 5840 <mss 1460,nop,nop,sackOK> (DF)
16:35:11.315558 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: . ack 1 win 64512 (DF)
16:35:23.315243 %NAME%.%domain%.com.smtp > 192.168.3.244.2159: S 3244244911:3244244911(0) ack 4100659814 win 5840 <mss 1460,nop,nop,sackOK> (DF)
16:35:23.315562 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: . ack 1 win 64512 (DF)
16:35:24.304787 %NAME%.%domain%.com.smtp > 192.168.3.244.2158: P 3150255653:3150255693(40) ack 4083749657 win 5840 (DF)
16:35:24.305068 %NAME%.%domain%.com.smtp > 192.168.3.244.2158: F 40:40(0) ack 1 win 5840 (DF)
16:35:24.305102 192.168.3.244.2158 > %NAME%.%domain%.com.smtp: R 4083749657:4083749657(0) win 0 (DF)
16:35:24.305323 192.168.3.244.2158 > %NAME%.%domain%.com.smtp: R 4083749657:4083749657(0) win 0
16:35:28.315245 arp who-has 192.168.3.244 tell %NAME%.%domain%.com
16:35:28.315458 arp reply 192.168.3.244 is-at 0:d:56:cc:fc:ed
16:35:47.515242 %NAME%.%domain%.com.smtp > 192.168.3.244.2159: S 3244244911:3244244911(0) ack 4100659814 win 5840 <mss 1460,nop,nop,sackOK> (DF)
16:35:47.515462 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: . ack 1 win 64512 (DF)
16:36:19.563243 192.168.3.244.2159 > %NAME%.%domain%.com.smtp: F 1:1(0) ack 1 win 64512 (DF)


Tcpdump success:
[root@%NAME% root]# tcpdump host 192.168.3.244
tcpdump: listening on eth0

16:55:05.984158 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: . ack 1 win 64512 (DF)
16:55:05.986604 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: P 1:41(40) ack 1 win 5840 (DF)
16:55:05.989099 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: P 1:18(17) ack 41 win 64472 (DF)
16:55:05.989143 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: . ack 18 win 5840 (DF)
16:55:05.989231 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: P 41:67(26) ack 18 win 5840 (DF)
16:55:05.990499 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: P 18:58(40) ack 67 win 64446 (DF)
16:55:05.991288 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: P 67:75(8) ack 58 win 5840 (DF)
16:55:05.991636 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: P 58:96(38) ack 75 win 64438 (DF)
16:55:05.992639 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: P 75:83(8) ack 96 win 5840 (DF)
16:55:05.993044 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: P 96:102(6) ack 83 win 64430 (DF)
16:55:05.993259 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: P 83:120(37) ack 102 win 5840 (DF)
16:55:05.993884 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: P 102:1439(1337) ack 120 win 64393 (DF)
16:55:06.025241 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: . ack 1439 win 8022 (DF)
16:55:06.025512 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: P 1439:1442(3) ack 120 win 64393 (DF)
16:55:06.025561 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: . ack 1442 win 8022 (DF)
16:55:06.029579 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: P 120:150(30) ack 1442 win 8022 (DF)
16:55:06.186758 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: . ack 150 win 64363 (DF)
16:55:07.061907 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: F 1442:1442(0) ack 150 win 64363 (DF)
16:55:07.062191 %NAME%.%DOMAIN%.com.smtp > 192.168.3.244.2319: F 150:150(0) ack 1443 win 8022 (DF)
16:55:07.062447 192.168.3.244.2319 > %NAME%.%DOMAIN%.com.smtp: . ack 151 win 64363 (DF)

 

Dude, that version is ancient. I'd suggest upgrading, then if the problem persists you'll probably get a lot more help.
 
Alright. I believe the latest version is 2.1 with some bug fixes. We are going to do an upgrade and see if the problem persists. Thanx!
 

Be sure to read the 'UPGRADING' files in the docs directory for each major version you are upgrading 'through'. Follow the directions for each upgrade and you should be fine.

 
Did the upgrade to Postfix version postfix-2.1.4-3.rh73 and now connecting to the mail server from any of our segments is instantaneous. I am going to continue to do packet inspection on smtp and tcpdumps to make sure this has resolved the problem. Any idea if this bug is common place?
 

I'm not sure, I've been using postfix for about 4 years and I don't think I ever ran 1.1.x :)

I'm glad things are looking up for you. I think you'll be very please with the new features in 2.1 as you get into them.

 
Well, looks like its time for the shaman to revive the horse so that we can kick it to death again. Was working fine this morning for about the 1 hour or so while I was testing it. Now we're back to the same old thing. Works fine for a little bit, then I get timeout errors trying to connect to the mail server (client --> server [same segment]). Go out to dnsreport.com and do a check on my domain, unable to connect to the mail server.

I thought maybe a bandwidth issue, about the only thing I hadn't checked into much. So I setup a few tasks to eat up all the available bandwidth on pipe to the ISP and then started sending messages. Worked just fine, it took a few seconds, but that was to be expected.

God... you have love computing. Any ideas?
 
I see you have some RBL checks in your config,
Did you check the responses from these RBL Dns Servers ?
Or other DNS related issues ?

--

Maurits

~ There is no place like 127.0.0.1 ~
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top