Hi,

we have the same problem at our company after upgrading Brightmail to version 8.0.0-24. I've opened a Symantec case for that. Let's see what happens.

Regards

Mario

Hi Mark,

So can no mail be sent to these domains?

Would be interesting to capture debugged Mail Transfer Agent logs when trying to send messages to these domains, but I'm guessing that's probably what Support is asking to see...

Kevin

Hi kk76,

in our case nearly all mails can be sent (70.000 per day) but some messages still in the delivery queue. Mails send out to the same email address or domain can be send...

I didn't think that Symantec will find something in our MTA logs because the test mails are send well.

Mario

Hi Mario,

It's not abnormal for there to always be a handful of messages in the delivery queue, what normally gets stuck in there is some missed spam messages where it tries to bounce the message back to an invalid sender or domain.  In that case it'll keep retrying and eventually expire.

What you shouldn't see in the delivery queue is messages destined for internal mail servers, that could definitely indicate a problem, or messages being sent outbound to domains you would be frequently working with. In this situation a debugged mail transfer agent log will definitely help as you capture exactly what's happening when trying to send the message to the remote mail server.

Kevin

Just simple text messages go just fine it seems that larger messages get stuck. It may be a TCP throttle at the other end but eventually I would think the messages should make it - it looks like there are connections establised that just eventually just time out...

tcp        0 150592 ma-relay4.ametek.:57869 ib1.charter.net:smtp    ESTABLISHED
 

2009 Mar 10 10:00:55 EDT (info) ecelerity: [7542] CONN-01040 #sms#00000018 ib1.charter.net 216.33.127.20 [56] 2

2009 Mar 10 10:00:55 EDT (info) ecelerity: [7542] ML-DELIVERY_ATTEMPT: Message ID: F8/41-18974-BA841B94, Audit ID ac100169-b7b96ae000004a1e-95-49b148a9edd7, Delivery attempted, sender: prvs=XXXXXXXX=XXXXXX@ametek.com

Looks right now like this was caused by the 8300 not playing nice during autoneg with the HP switch (or visa versa). I think I had these hard coded to 100 / full before the upgrade. I hope engineering will confirm if it flips it back or not during the upgrade from 7.7 and add to the rel notes if it does.

I'll have to test that and get back to you. I would assume that would be incorrect behavior and not expected. I'll let you know as soon as I get the results.

Thanks!

Hmmm, I'm not so sure about that, if we had changed those settings I probably would have expected that you wouldn't have been able to send anywhere and not to just specific domains or just large messages...

I updated our SMS appliance (8260) last week (march 5), and our problems started.  SMTP on the appliance stops responding after couple of hours and we need to restart it 3-4 times a day.  Many users are complaining because they're receiving many emails with the subject "Delayed Mail (still being retried)", messages stops coming in from the internet, or they're delayed for hours.  The earlier version, 7.7, worked without any issue!  I tried to check the inbound and delivery queue from the http page of the appliance, and it just stays loading and never finishes.  Also, I tried to see hoy much space the queues are using on Status->Host->Messages queues, and it shows 0kb on all queues.  Is there any way to roll back the 8.0.0-24 patch?

Hi Hector,

Just curious, how are you DNS settings configured on the Scanners?  Have you configured the Brightmail Gateway to use one of your internal DNS servers, or is it set to 'Use internal DNS server'?

Kevin

It was using our internal DNS servers.  To make it work, I had to check every 15 minutes an do an external test to the smtp, to see if it responds.  Sometimes, after a few minutes or hours, it stopped responding and I had to restart it.   I had another appliance with a damaged raid controller.  I swapped the hard drives (the damaged one had Brightmail 7.7), booted it up, and presto!  Its working!  Only had to change the mx records on the DNS.  Dont know what happened with the update, because that appliance was working for more that three years (it was installed on 2005, think it was version 4.1, dont remember).

I tried to reproduce the behavior of the duplex settings and was unable to. After setting our test box to 100/full and updating, the settings persisted. So looks like this isn't a reproducable issue.

Thanks for bringing it up though. And sorry it took a while to finish.

we have upgraded to v 8 when it was released.Until now it was working with no problem.But now all mails are stuck.We could not send any mail to any domain.It was using our internal DNS servers.Some mails have "421 4.4.0 [internal] no MXs for this domain could be reached at this time" and "451 4.4.1 [internal] No valid hosts (unable to make any connections)"
What will do?

Are you sure you aren't having network or DNS problems?  What have you done to verify...

We have no problem with dns.All mails are coming with no problem.But outgoing was  full.

At smtp configuration /advanced settings there is "SMTP Delivery Bindings" part it was writing our internal ip adresses.Now i made all "auto" and everting is fine.All mail that stuck at outgoing was send.I don't understand the problem but it is solved.
thanks

I actually upgraded all of my 8260's to 8.0 from 7.7 this Monday.  I also added a VM scanner.  The VM scanner had almost 4,000 emails stuck trying to go out.  I tried all of the suggestions above and I've got it down to about 600.  The annoying problem is that after I made the changes it would deliver a hundred or so and then seemed to stop so I ended up forcing a flush several times to get it down to where it is now.  Once my queue is empty I'm going to remove the VM scanner and see how the two 8260's do.  I'd do it sooner if I knew how to move pending messages from one scanner to another.  I think I'll go research that now...

I send this problem to support so i will see.

Diagnostic-Code: smtp; 451 4.4.1 [internal] No valid hosts (too many connection failures) from Symantech Bright mail gateway
  

also all outbound mails are stuck in antispam with following error  .Please audit log error.

Message Data
	ID:	0a8c0247-b7b67ae00000023a-9e-4b810b625949
	Message-ID:	<003601cab2e1$72f2a210$12030e0a@kfupm34e40dae7>
	Tracker:	AAAAAwAAAZES015ZEu9qaA==
	Accepted From:	10.14.3.18
	Scanners:	KFUPM Mail Antispam Server1
	Time accepted:	Sunday, Feb 21, 2010 01:30:58 PM AST
	Direction:	Outbound
	Sender:	ptashraf@kfupm.edu.sa
	Original recipients:	ashraf.rhce@gmail.com
	Original Subject:	testing rhce
	Full attachment list:	None
	Suspect attachments:	None
Recipient Data
	Intended recipient:	ashraf.rhce@gmail.com
	Verdict:	Verdict Filter Policy Group Details Content Compliance violation: green-desclaimer  green-desclaimer  green-test  None
	Actions taken:	Add annotation
	Delivery:	Delivered To Delivery Time None
	Untested verdicts:	Message was sent from a suspect spammer, Symantec Global Good Senders, Symantec Global Bad Sender, Locally identified suspected virus, Suspected virus, Content Compliance violation: Block Objectionable Content, Content Compliance violation: spammer-lists-quarantine, Content Compliance violation: block-credential-spoofing, Content Compliance violation: KFUPM-From-Spoofed-quarantine, Content Compliance violation: Exception-for-KFUPM-spoofed-email, Content Compliance violation: kfupm-group-block-inbound, System denied IP, System allowed IP, System allowed email address or domain, System denied email address or domain, System allowed third party domain, System denied third party domain, User allow, User reject, Virus attack, Directory Harvest Attack, Unknown recipient, Connection Class, Default Connection Class, Connection Class 1, Connection Class 2, Connection Class 3, Connection Class 4, Connection Class 5, Connection Class 6, Connection Class 7, Connection Class 8, Connection Class 9, Sender authentication failed, Bounce attack signature present, Bounce Attack, Blocked language, Known language
	Other recipients:

Hi Ashraf,

Since ALL the outbound mail seems to be stuck in the delivery queue, it seems like it's a DNS or firewall or some other configuration issue.  I had requested details on the configuration and some screenshots in ther other forum thread that you opened for outbound issue.  Can you please respond to my requests in that thread?

Regards,

Adnan

Hi,

I have the same issue and most of the mails are stucked in the delivery queue stating the same error "421 4.4.0 [internal] no MXs for this domain could be reached at this time"

Can anyone have the solution for this issue.......... we are using SBG 8.0.3-11 on Vmware Esx 4

Thanks in advance for the response........

Hi Prashant,

Is the "most" of te mails or "all" the mails?  Are you able to delivery any mail outside?  If it's most of the mail, then the issue is probably with the DNS server.  Is this a new setup?

Regards,

Adnan

Hi,
I am having the exact same issue using 8.03_11. Only a 10 of my emails are stuck in the Delivery queue with the
"21 4.4.0 [internal] no MXs for this domain could be reached at this time" error on the error column. However when I use the Nslookup utility to check for the MX record it succesfully finds the  MX record of the domain. Could it be a temporary issue with the other end?

If SBG is able to resolve the MX record for the domain as your nslookup test on SBG indicates, then it's most likely the other end.  You can try sending test mail using telnet to the recipient domain from SBG to confirm that.

Regards,

Adnan

I am having the same issue. seems to be just some emails stuck, not all. and not alwasy the same ones. its very very strange. i'm starting to get more calls to the helpdesk about it. does an upgrade to 9 fix this?

Thanks, from a frustrated email admin.