And just another crash of Mercury during transmitting of a 17 MB attachment by MercuryC which cause to a 24 MB .MC file (single client session file). But this time I had additionally activated the session.log which is showing for this e-mail:

T 20141210 120120 230 Begin processing job MO0001A4 from user@domain.com
E 20141210 120407 230 TCP/IP error.

There are no any special settings in our MercuryC configuration:

- Connection port: 587 - SSL encryption via STARTTLS command

- TCP/IP timeout 180 seconds

- Poll the queue every 60 seconds

- Use extended SMTP features where available > checked

Even if there are any TCP/IP problems with our internet connection - Mercury should not immediately crash in such case, isn't it?

I will test Brians idea to change-over to MercuryE during the next time although I would prefer sending all outgoing mails via our ISP mail server.

Cheers

joerg

Hi Guys,

Since I have installed the beta version 4.8, Mercury is incidentally crashing round about once per month. But it seems not to be a SSL problem like in v4.7, although the crash happens always during a SMTP connection of Mercury C (SMTP Client) when sending an e-mail with attachment to the internet.

I have checked the SMTP session log: When transmitting an attachment, it stops from one second to the other and Mercury is crashing. Here are only the last lines from the SMTP session log for the affected e-mail (all other header lines are ok from my point of view):

16:11:24.484: << NSzKcxQ5s3UfzSJ90wurNlKmVY6a4oePhe00m816MNbRuPhnvzxAoKCX5hOkUulEVQNpiMz+<cr><lf>
16:11:24.484: << kI5XpSxxdsv8jW1RmWXeXul4VQWsstPxqnd6nasmfmtZDO5LQT9yU/yQzfzxSP99woWUok7u<cr><lf>
16:11:24.484: << LoY3m788DahpH1AZGkNsSbhOgba9JZ++CP784Zujjd17LidMhuV94wE68DM74/xwkC+2TMlH<cr><lf>
16:11:24.484: << KN7BDFiuTP7tYnizsb/M5QA4NDpVVfMXQR9s2ijf/LHRTCIq6zfpzRYEcw8bHb8KQLA5hckw<cr><lf>
16:11:24.484: << fBrk+bizLKI4gu9t1JqafE4xxRv39M5HKSbKfU6hZKdNNHIGOkfcFGwOgDfjDsHmWm5U/5sr<cr><lf>
16:11:24.500: << wcLQ0YX66M2VamHo6EsWjwbgk13ACpo4xUfTdxWk3U3ogjQDtKcL5r4MN9+1+a7dGAbDpPS2<cr><lf>
16:11:24.500: << LcfdO/eYaQ+Kcld3PkA2mfSen2wzJxUWXAoWQ+7WSWgMOYH/JoZcQzgk+e5PGhjOF240Fly9<cr><lf>
16:11:24.500: << /+NiMeT8UrKp4KMuvo81cmYrtirsog7WyFnGuq2T64UK6jo2XVjpuNTTVvlpNSbCWpP9U43c<cr><lf>
16:11:24.500: << qv8/bWd2HUvOK2tX2gTOgx1yo8v/1ysVAkCASam1/7Ov3r6FLCZHkEmGwLm9WADfVJzA9wde<cr><lf>
16:11:24.500: << irZEv+su/lOg2TKAntTW6ce2711Q2dIWPzRrsC2KIleXhpv7SkT+/7vOmGEZP/iVVtRyaZmC<cr><lf>
16:11:24.500: << nGdXXMCqbxONjpxYVLp/FF4Dgdcq3T/6Ce8zlRq1Ozog6BRaS2Z7v3pq+h6gE/B+UBUNVHG2<cr><lf>
16:11:24.500: << A08h5fCgTYR0e6oVxZxqYluikklhtGSTS9ZCDds1ou8Mb3r1/204o1klpJw26+A2lt4vV/jU<cr><lf>
16:11:24.500: << xgCLh4B7Z1SlbC/NNTQGFfed/pwEJRDqfPsuPWQKApZ4eNO0x347HOnCX9mVgBxNQU66vIOh<cr><lf>
16:11:24.500: << b/OvDC4dYjW25eIx5AB6e+qpoNPgQi8FvyjS4tJ9JTJ6fMF0BZ3aWrR1tqFAJWe3lUzOtuQS<cr><lf>
16:11:24.500: << H+wufgO4gi67gq6Iqlm2RgD6YEhEf4XxlFlBl6OCrsjcJl+2ALs9VY4scXuq3M6O21Ojgq7I<cr><lf>
16:11:24.500: << /zVK1Qr47al0XfhnFbHj/99uT5VI1rjerXqEOQO5600es9tTZUsYt6cK2O2pQLbZ7anyP0G4<cr><lf>
16:11:24.500: << PVX2tu26KkHcniong4lerrensmguY79MoHmEOQVx51nOC/FYXQ/4EwWdNvG4tI7a2nVQcMtV<cr><lf>

I'm not able to reproduce this failure. Mercury works properly for many days and weeks and then it crashes again. Has anybody an idea?

Greetings

Joerg

It seems nobody else has experienced such rare but regular crashes of Mercury. Is it recommended to regularly restart Mercury e.g. on a daily basis to improve its stability?

Joerg

Hi Guys,

Mercury is just crashed again. It seems this happens when big attachments are being transmitted. Normally Mercury terminates a SMTP session with:

10:51:16.656: << --Message-Boundary-13134--<cr><lf>
10:51:16.656: << .<cr><lf>
10:51:17.109: >> 250 Requested mail action okay, completed: id=0McAb5-1YEcdv1HbL-00Jc93<cr><lf>
10:51:17.109: << QUIT<cr><lf>
10:51:17.312: >> 221 kundenserver.de Service closing transmission channel<cr><lf>
10:51:17.312: [ * ] OpenSSL secure session normally terminated.
10:51:17.312: --- Connection closed normally at Mon Dec 08 10:51:17 2014. ---
10:51:17.312:

but with attachment (presently about 16 MB), as already stated in my first posting:

11:05:28.062: << sh6S3LGs/Ow1fyC1rU0LrRsMcgpDlXrkeacP/ALas2GyqyYKSkAZqABVoAZ0afRmKzyxp0vo<cr><lf>
11:05:28.062: << 51Umu1AECkAeSN+wmKdzBmpAFagAZSBbevDjEzajlhSgCtSALtQJ1DXxeKAAtHqUpq2xnjSc<cr><lf>
11:05:28.062: << GxpnjpYn8r6uqlteaqmfI+NCHkifN+1r0YUakEQgzJFt1k+XgRKQ2LK++c666zyQ7r0+zv+t<cr><lf>
11:05:28.062: << mtzXdlKTLVpf++Q1wdWe59vO42yn3KCMQH2OJ/1aBurWyO3KCrS+CXq5czGR/0PPG+XsLN7o<cr><lf>
11:05:28.062: << UF8x7xrh0X949LNSigN5ILnRiVJsKMVei6QUW9Sl2FCKfb4opdiyLkW1/p910elozDJtl0aX<cr><lf>
11:05:28.062: << A+KYP96EljOd15pT+3G+V43xw+MLkWqGzWDnF90Z9deUrE/S6rmtoOuPUztb78LImYUXkGbn<cr><lf>

it immediately aborts the transmission when sending the attachment.

I need a plan.

Joerg

I am running v4.74 so am not sure I can be of help but here are my thoughts...

1.  Does your relay server impose a fixed size limit?

2.  Have you tried MercuryE?

As for daily restarts, I have the "Exit and restart each day after performing daily maintenance" option enabled (Advanced tab in Core). 

[quote user="bfluet"]

1.  Does your relay server impose a fixed size limit?

2.  Have you tried MercuryE?

As for daily restarts, I have the "Exit and restart each day after performing daily maintenance" option enabled (Advanced tab in Core). 

[/quote]

Hi Brian,

1. Our internet service provider imposes a size limitation but it is quite bigger than 16 MB. We often send much bigger attachments.

2. No Mercury E never tested. As long as we are using Mercury, always Mercury C (relay) is running. Do you use Mercury E to forward your user's emails to the receiving SMTP server of your ISP? I will read the manual for the differences between E and C and what they are for.

Regards

Joerg

I have run MercuryE successfully but have ended up switching back to MercuryC twice because of blacklisting issues (the result of new ISP's assigning us a static IP address that is blacklisted from day one).  A nice thing is that switching between the two is as easy enabling the one not enabled.  You can check the MercuryE log periodically for delivery issues to help determine whether it is the right answer. 

Obviously you would want to send that 16MB message as a test to determine whether the crashing problem remains. 

I know the problem with the blacklisting. Sometimes different e-mails are being returned to us because the mailserver of our ISP is erroneously blacklisted. But usually they already know the problem and take care that they will be removed from that list as fast as possible. Often this is no problem for such big ISPs.

But in case I would use MercuryE my own server is acting as mail server (and not longer as relay only) and could be set on a blacklist, isn't it? And then we have to take care by ourselves to remove us from those lists.

Does Mercury saving the MercuryC settings when switching to MercuryE and backwards?

You are correct in that you would have to handle removal from a blacklist yourself.  My last go around with this I was blacklisted by two lists.  Removal from one was quick and easy but the other (AT&T) took multiple requests over several weeks before they finally acted forcing me to switch back to MercuryC.  I have only experienced blacklisting when being given a previously blacklisted IP address.  I would be curious to here from others about whether blacklisting is a recurring problem.

As for MercuryC & E remembering previous settings, I am pretty certain that they do.  From what I remember, the last time I switched back to MercuryC it contained all of the settings from my previous ISPs relay server.  I don't know enough about Mercury to know what file these setting are maintained in or whether they would be easy to restore but neither module requires much in the way of configuration so a screenshot of the config screen is all you need.

[quote user="Joerg"]

It seems nobody else has experienced such rare but regular crashes of Mercury. Is it recommended to regularly restart Mercury e.g. on a daily basis to improve its stability?

Joerg

[/quote]

There is a built-in facility to do this - Configuration / Core module - Advanced tab.  In fact I've never used this and although I run the beta version with MercuryC, I'm not experiencing this problem - even with very large attachments - 20-30MB.

Hi Paul,

I know the built-in opportunity to regularly restart Mercury e.g. during daily maintenance. But I do not know whether this could help improving the stability. But it seems, it is not a stability issue. Rather I think there is something wrong with the attachment, or it is a reply from another sender outside our user group which is corrupted or something like this. But in any case Mercury should not completely crash.

Cheers

Joerg