Amazon SES: Hangs at the end of each group?

[Gizmocracy]

I use Amazon SES through SMTP. It worked perfectly in testing, which is why I went ahead and bought MBM. However, now that I use it in production, I'm getting a weird problem: It will send fine for however many emails I have set in "Group emails", then it starts to count down for the "Interval" delay as expected. However, when the "Next Delivery in" counter reaches 00:00:01, it hangs there forever.

The delivery window doesn't show any obvious errors:

[01] 1/4 09:33:05     ~ Stage 3 dispatch done
[01] 1/4 09:33:05     ~ Delay of 10s [00:00:10]
[01] 1/4 09:33:05 > QUIT
[01] 1/4 09:33:05 221 Bye
[01] 1/4 09:33:05     ~ Stage 4 dispatch done
[01] 1/4 09:33:05     ~ Next Delivery in 00:00:10

It hangs there for so long that eventually Amazon gives up and closes the port:

[01] 1/4 09:35:14 *** SMTP TIME-OUT 1 - 120 s
[01] 1/4 09:35:14 --> Reconnecting attempt 1 of 5

And the reconnection attempt also just hangs forever.

All of my settings match http://www.maxprog.com/site/support/us/ ... on_ses.php, except that connections is set to 1.

What's going wrong here and how do we fix it?

 

[stanbusk]

Not sure, all MaxBulk Mailer does is to reconnect. The expected behavior is the server responding. Have you tried a longer delay?

 

[Gizmocracy]

With a delay longer than the 120-second timeout, we get a different error. Here's what a successful login looks like at the start of the group:

[01] 1/4 23:12:52     ~ Opening connection for delivery...
[01] 1/4 23:12:52     ~ Connecting to email-smtp.us-east-1.amazonaws.com Port:465
[01] 1/4 23:12:52 220 email-smtp.amazonaws.com ESMTP SimpleEmailService-296911475
[01] 1/4 23:12:52     ~ Connected
[01] 1/4 23:12:52 > EHLO [10.0.1.59]
[01] 1/4 23:12:53 250-email-smtp.amazonaws.com
250-8BITMIME
250-SIZE 10485760
250-AUTH PLAIN LOGIN
250 Ok
[01] 1/4 23:12:53 > AUTH LOGIN
[01] 1/4 23:12:53 334 VXNlcm5hbWU6
[01] 1/4 23:12:53 > [our SMTP user ID]
[01] 1/4 23:12:53 334 UGFzc3dvcmQ6
[01] 1/4 23:12:53 > ********************************************
[01] 1/4 23:12:53 235 Authentication successful.

But here's what the relogin attempt at the end of a 6-minute window looks like:

[01] 1/4 23:12:55 > QUIT
[01] 1/4 23:12:55 221 Bye
[01] 1/4 23:12:55     ~ Stage 4 dispatch done
[01] 1/4 23:12:55     ~ Next Delivery in 00:06:15
[01] 1/4 23:19:11 220 email-smtp.amazonaws.com ESMTP SimpleEmailService-296911475
[01] 1/4 23:19:11 > EHLO [10.0.1.59]
[01] 1/4 23:19:11     ~ Connected
[01] 1/4 23:19:11 > EHLO [10.0.1.59]
[01] 1/4 23:19:11 250-email-smtp.amazonaws.com
250-8BITMIME
250-SIZE 10485760
250-AUTH PLAIN LOGIN
250 Ok
[01] 1/4 23:19:11 > AUTH LOGIN
[01] 1/4 23:19:11 250-email-smtp.amazonaws.com
250-8BITMIME
250-SIZE 10485760
250-AUTH PLAIN LOGIN
250 Ok
[01] 1/4 23:19:11 > AUTH LOGIN
[01] 1/4 23:19:11 334 VXNlcm5hbWU6
[01] 1/4 23:19:11 > [our SMTP user ID]
[01] 1/4 23:19:11 334 VXNlcm5hbWU6
[01] 1/4 23:19:11 > [our SMTP user ID again]
[01] 1/4 23:19:11 334 UGFzc3dvcmQ6
[01] 1/4 23:19:11 > ********************************************
[01] 1/4 23:19:11 535 Authentication Credentials Invalid
[01] 1/4 23:19:11     ~ Fatal error, giving up!

Notice how you send EHLO twice, then send AUTH LOGIN twice, which appears to be causing the login to fail. This terminates the run, but then the app tries to send the delivery status report email, and this same sequence of doubled messages persists in that send, too, which also fails.

 

[stanbusk]

What version of MaxBulk Mailer are you using?

 

[Gizmocracy]

"8.4.1 - US" Pro.

 

[stanbusk]

Try 8.4.2 and let me know.

 

[yuewupostbox]

Hello stanbusk,

i have the same issue which the first blast is success but stuck at 2nd turn group interval "Next Delivery 00:00:01".

i am using zimbra 8 with maxbulk 8.4.2 & both of the logs (zimbra server log & maxbulk log) didn't reflect a smtp auth in the 2nd turn of group interval, while i click stop, it triggered smtp auth to send report and it's success, please help, thanks!

 

[stanbusk]

Is 'Zimbra' a SMTP server?

 

[yuewupostbox]

Hello stanbusk,

Yes, thanks!

 

[yuewupostbox]

Hello stanbusk,

V 8.4.3 fixed the issue, thanks!

 

[hq_sysadmin]

I just upgraded to 8.4.3 and I still have the problem with Amazon SES. Any other ideas?

 

[lockside]

Hi,

I have the same problem with Amazon SES

[01] 5/21 14:50:27 535 Authentication Credentials Invalid
[01] 5/21 14:50:27 ~ Fatal error, giving up!
[00] 5/21 14:50:27 ~ [-]
[00] 5/21 14:50:27 ~ Closing all streams
[00] 5/21 14:50:27 ~ Creating report...
[00] 5/21 14:50:27 ~ Sending report...
[01] 5/21 14:50:27 ~ Opening connection for delivery...
[01] 5/21 14:50:27 ~ Connecting to email-smtp.us-east-1.amazonaws.com Port:25
[01] 5/21 14:50:27 ~ Connected
[01] 5/21 14:50:27 220 email-smtp.amazonaws.com ESMTP SimpleEmailService-376766033

Do you have any work arounds or fixes for this?

Help would be appreciated.

 

[craigimass]

I can confirm that this happens to me in SES.

It happened in my old 6.7 version and I just upgraded to the newest Mac version and it still happens.

I have tried numerous settings and it happens every time. It simply cannot reconnect.

Yet, if I manually stop the sending and then start it again, it goes right ahead and finds the server.

IMHO, it is a bug in the sw of some kind. A reconnect after the pause should work just like the stopping and starting, yet it does not.

 

[craigimass]

Here is the log - it sends the first group, pauses and then does not connect back up.

[03] 5/22 23:15:32 250 Ok 0000013ecf626946-77c16d02-7ee7-42f8-91ac-74bd459f84ca-000000
[03] 5/22 23:15:32 ~ Stage 3 dispatch done
[03] 5/22 23:15:32 ~ Delay of 0s [00:00:00]
[03] 5/22 23:15:32 > QUIT
[03] 5/22 23:15:32 221 Bye
[03] 5/22 23:15:32 ~ Stage 4 dispatch done
[01] 5/22 23:17:29 *** SMTP TIME-OUT 1 - 120 s
[01] 5/22 23:17:29 --> Reconnecting attempt 1 of 5
[02] 5/22 23:17:30 *** SMTP TIME-OUT 1 - 120 s
[02] 5/22 23:17:30 --> Reconnecting attempt 1 of 5
[03] 5/22 23:17:32 *** SMTP TIME-OUT 1 - 120 s
[03] 5/22 23:17:32 --> Reconnecting attempt 1 of 5

 

[anonify]

I have the same problem, I posted about it here before I saw this thread was already up. Like somebody else stated, stopping and starting manually appears to work just fine. Given that, I doubt it's SES that is causing the problem.

 

[stanbusk]

I believe it has to do with SSL, are you suing SSL?

 

[anonify]

You mean on my domain server? I am in fact using SSL. What would that have to do with this?

 

[stanbusk]

SSL reconnection perhaps, not sure. I will check MaxBulk Mailer connection code however as far has I know MaxBulk Mailer uses new sockets each time it connects. It never reuses them.

 

[stanbusk]

Try this new version: Mac OS | Windows

Does it work better? reconnect correctly?

 

[cindy.yu]

Hi

I met this big issue too. Could anyone advise how to solve it?

The connection is droped at the end of each group. No matter how many emails I choose in each group.

But when I stop the sending right away, then click "start". The connection works again. Amazon can only work in MaxBulk maile by mannually clicking in this way.

Thanks
Cindy