The entire site.  enter mail.myserversite.com just hangs.   My test page on var/www/html does not appear.  Receiving no emails.  So that means no access to roundcube, iredadmin.  I uninstalled SOGO.   giving me too me error messages.

I am not sure which log you want.  Tell me specifcally where and which log.   But I'm not sure that is the answer. 

Again,  it works then stops.   Starts again after an hour.   Flaky operation.

Thank you.  That is a valid suggestion.   I am using a "small instance" which is 2 GB.    I loaded up on 15 emails.  That was the whole point.  ALL my emails. 

I reduced that to about 6.  Then I reconfigured the virtual hosting files which are not precisely correct (need to call the private IP address not public addresses).   There was some change after that.  If was running out of memory, how long would that take to recover?    I am trying to construct a mirror site on the east coast.

Thank you for your response. Appreciate the interest.  So often problems are solved in terms of "what I have or do."

It does appear that memory is not the issue. Deep within the program it appears that Spam and Virus checking defaults to non-checking. In other words, that is not source of memory leak.

This was a significant problem. The site would just die and I would get no emails for a few hours and it would start again.

It is taken me months to learn that virtual hosting using Amazon Web services requires very specific coding. If does not follow exact conventions it does not work perfectly. 1st, you must use the private not public IP address. 

This how a typical AWS virtual host must be written.  This is specific to AWS.  Others may not experience the same problem although I know AWS is very common now.  The directory is not essential but it must be correctly pointed to the exact directory.  I chose to use a much more secure directory. 

So it appears for now the 000-default.conf and default-ssl.conf were the cause of the problem.  Not perfectly coded.  It works now.  And I wrote this in one conf file not two.

<VirtualHost 10.0.x.xxx:443>
    ServerAdmin  me@mydomaine.com
    ServerName mail.xxx.com
    ServerAlias xxx.com
   
    DocumentRoot /var/www/vhosts/xxx.com
    <Directory />
        Options FollowSymLinks
        AllowOverride All
    </Directory>
    <Directory /var/www/vhosts/xxx.com>
        Options Indexes FollowSymLinks MultiViews
        AllowOverride All
        Order allow,deny
        allow from all
    </Directory>   
    SSLEngine on

    SSLCertificateFile /etc/letsencrypt/live/mail.xxx.com/fullchain.pem
        SSLCertificateKeyFile /etc/letsencrypt/live/mail.xxx.com/privkey.pem

    Alias /iredadmin/static "/opt/www/iredadmin/static/"
    WSGIScriptAlias /iredadmin "/opt/www/iredadmin/iredadmin.py/"
    Alias /mail "/opt/www/roundcubemail/"
    ProxyPass /Microsoft-Server-ActiveSync http://127.0.0.1:20000/SOGo/Microsoft-Server-ActiveSync retry=60 connectiontimeout=5 timeout=3540
    ProxyPass /SOGo http://127.0.0.1:20000/SOGo retry=0
    Alias /awstats/icon "/usr/share/awstats/icon/"
    Alias /awstatsicon "/usr/share/awstats/icon/"
    ScriptAlias /awstats "/usr/lib/cgi-bin/"

</VirtualHost>

BTW, this is the third time I have reinstalled iRedMail.  Not interested in doing that everytime there is a problem.

No iptables is clean.  That was not the issue.  It reallys seems the issue is probably coding the virtual hosting.  For AWS it must conform to certain parameters.  That is why every service seemed to be fine.  There was no UFW block.  I just could not see any evidence of html activity.   It appears to be working flwalesssly again.

These are all learning lessons for everyone including myself.   And my understanding is that the UFW is not enabled by default.  Correct or not?

Only these seem a bit suspect, otherwise all ports are open and perfectly functional.   

Chain FORWARD (policy DROP)
target     prot opt source               destination

Chain OUTPUT (policy ACCEPT)
target     prot opt source               destination