Maximum concurrent threads for mail

[Steven Weiner]

I had previously been on a acf 9 enterprise server that was taken offline so now I am running on a new server with acf11 and lucee 4.5.

I'm tasked with sending out 2 newsletters a month, that hit 100k people each.  Each newsletter is customized and sent via cfmail.  I basically run through queries of 5000 records at a time and reload the page with a javascript refresh until all 100k records are done.

On CF9 ent cfadmin, I was able to check the keep connection alive, and work it up to 40 connections at a time with the mail settings.  I would be able to send 100k emails in about 45 minutes to an hour.

When I initially tried acf11 standard to find that its mail is single threaded and would take days to send the 100k records and cause a bottle neck with the rest of the outgoing emails on the server..

I then installed Lucee.  I'm getting better performance then with  acf11 standard, but still kind of slow.I've got it sending out about 300 messages a minute.

I went into Lucee server and web admin, turned it from 20 up to 40 maximum concurrent threads.... restarted, and then up to 60, restarted, and then up to 100 and restarted....  I tried throwing more ram at it... and the performance I get with 20, seems to be the same performance I get with 100... so I'm basically wondering if the concurrent tasks listed in server/web admin even have an affect on mail sending?

The remote-client/open folder, reached a maximum of 95k messages and has been working for the last 6+ hours to bring it down... but I was wondering if there is any other setting that I should look at for increasing the speed it sends out.

I have always been connecting to sendgrid to send my messages on both cf9ent and lucee.

Any tips would be appreciated.

Thanks

Steve

[Michael Offner]

we are aware of this and it is on our to do list to improve this.

Micha

--
You received this message because you are subscribed to the Google Groups "Lucee" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lucee+un...@googlegroups.com.
To post to this group, send email to lu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/a1e50b5d-055f-48fe-8e9f-5f09de854bab%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Michael Offner]

i assume the following is the problem, Lucee makes sure to reuse connection to the server and does open/close the connection for every mail, what is a good thing. But maybe Lucee is using only one connnection for all this threads?

I will check that ...

Micha

[Steven Weiner]

I am going to assume that yes it only uses one thread for all connections since it only appears to be sending 3-5 messages a second.  Also, in trying to trace it, I ran a netstat expecting to see 100 open connections to the mail server and it appeared to only have 1 while running.

In regards to opening and closing the connection, I find that to be a good thing too, although it does slow mail transferring down.  On ACF I once had an issue where I was sending mail for multiple clients through sendgrid, and since they all used mail.sendgrid.com as the server, but with different usernames and password, ACF was just looking at the address and pushing all email through one connection so mail created for different accounts if they were in queue at the time were being pushed through the wrong connection.

steve

[Michael Offner]

my assumption was not correct but close, problem should be solved now for the next patch release

https://bitbucket.org/lucee/lucee/issue/319/mail-sending-is-synchronized

Micha

To view this discussion on the web visit https://groups.google.com/d/msgid/lucee/7734c3f0-a222-4601-96a2-d0a5136d55ad%40googlegroups.com.

[Terry Whitney]

At a previous company We had similar issues with mass emails.

My solution was to create a relay server to handle the actual delivery of mail. 

A little postfix server with 1GB of memory was more than enough to handle 25K mails an hour sustained.