Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
46
E-Mail / Re: spamassassin.service: Failed with result 'start-limit-hit'
« on: October 25, 2021, 11:39:53 PM »
First of all, sorry for the delay of about 2 months in replying. I read your answer, and I really appreciate it.
However, during this time I have been looking for information by conducting tests ... etc ... and although I could have answered before, I have chosen to answer after having carried out tests and to say forcefully, that the results are positive after a month having implemented the last change / test.
Effectively your answer / solution solves the initial problem that I expose in this thread, even so, it has always surprised me a lot that, spamassassin raises the cpu consumption to 0.90 - 1.00 on average, regardless of the fact that, no matter how much we configure spamassassin such and as the cwp wiki says, also as it supports cwp (I requested support for it) never send any mail to the spam folder ... this was very strange to me.
The answer to this increase in cpu is that perhaps, I would have many emails on my server ...
The tests I have done on 4 servers, I have 3 for testing 1 in production.
Gradually I was eliminating mails from the main test server, and the consumption was always the same ... there came a day that, I eliminated all the accounts of that server, no account, no mail, nothing, the consumption of spamassassin was still the same, I raised the server to 0.90 - 1.00 on average. Something very very high.
Too bad it didn't save the url where I got the information, I did write it down in kate (linux notepad) and I have it saved in my documents. If you have saved the url, you could say the clear terms of the explanation.
Well, the change happens by modifying / etc / sysconfig / spamassassin file:
In wiki they indicate:
The reason why, despite this, spamassassin does not filter, does not take the identified emails to the spam folder and does not rename them with *** SPAM *** as indicated in the wiki, is because it should not be configured in / usr / share /spamassassin/local.cf
Rather, it must be configured in another path.
After that, you can test by sending an email with the subject:
"XJS * C4JDBQADN1.NSBN3 * 2IDNEN * GTUBE-STANDARD-ANTI-UBE-TEST-EMAIL * C.34X" (without quotes) and it looks like the mail is put in the spam folder and renamed the subject to *** SPAM ***.
Greetings.
However, during this time I have been looking for information by conducting tests ... etc ... and although I could have answered before, I have chosen to answer after having carried out tests and to say forcefully, that the results are positive after a month having implemented the last change / test.
Effectively your answer / solution solves the initial problem that I expose in this thread, even so, it has always surprised me a lot that, spamassassin raises the cpu consumption to 0.90 - 1.00 on average, regardless of the fact that, no matter how much we configure spamassassin such and as the cwp wiki says, also as it supports cwp (I requested support for it) never send any mail to the spam folder ... this was very strange to me.
The answer to this increase in cpu is that perhaps, I would have many emails on my server ...
The tests I have done on 4 servers, I have 3 for testing 1 in production.
Gradually I was eliminating mails from the main test server, and the consumption was always the same ... there came a day that, I eliminated all the accounts of that server, no account, no mail, nothing, the consumption of spamassassin was still the same, I raised the server to 0.90 - 1.00 on average. Something very very high.
Too bad it didn't save the url where I got the information, I did write it down in kate (linux notepad) and I have it saved in my documents. If you have saved the url, you could say the clear terms of the explanation.
Well, the change happens by modifying / etc / sysconfig / spamassassin file:
In wiki they indicate:
Code: [Select]
SPAMDOPTIONS = "- d -c -m5 -u nobody -s $ {SAHOME} spamd.log" But we actually remove the "-d" from the beginning, looking like this:Code: [Select]
SPAMDOPTIONS = "-c -m5 -u nobody -s $ {SAHOME} spamd.log" If my memory does not fail me (and it is easy for it to fail) this causes spamassassin to be executed as a service (I do not remember exactly if it was a service or another ...) and now it does, spamassassin, without making the changes that you mention in your post and that I had also seen in other internet threads, it is always executed, without any crash and without load on the server, the server load is normal, if you are using apache, then whatever Apache loads or whatever, and during non-use hours, the server load is 0.00 and spamassassin is active.The reason why, despite this, spamassassin does not filter, does not take the identified emails to the spam folder and does not rename them with *** SPAM *** as indicated in the wiki, is because it should not be configured in / usr / share /spamassassin/local.cf
Rather, it must be configured in another path.
After that, you can test by sending an email with the subject:
"XJS * C4JDBQADN1.NSBN3 * 2IDNEN * GTUBE-STANDARD-ANTI-UBE-TEST-EMAIL * C.34X" (without quotes) and it looks like the mail is put in the spam folder and renamed the subject to *** SPAM ***.
Greetings.
47
E-Mail / Re: CWP Email Compromised & SPAM Filtering
« on: August 17, 2021, 10:21:26 AM »Hi DNA, put the solution here or send it as a public me on other 100 forums, anyway the people from cwp do not give you any solution how to configure correct postfix, spamassassin, any mail that enters and leaves to be scanned in time real antivirus, I have the same problem with clients, I receive a lot of unwanted emails, not to mention the mail quota, which even now has not been resolved.I know that / those problems you indicate ...
48
E-Mail / Re: sync 3 mail servers
« on: August 17, 2021, 10:20:30 AM »
In my case, I don't need the 3 servers syncro in real time.
I have a script created, the cron call this script 2 times at day, and syncro all mails.
But only one server is the "master server mail", the rsync is from this master server mail to the others 2 servers.
When access to others 2 servers are the same emails of master server mail.
I have a script created, the cron call this script 2 times at day, and syncro all mails.
But only one server is the "master server mail", the rsync is from this master server mail to the others 2 servers.
When access to others 2 servers are the same emails of master server mail.
49
E-Mail / Re: spamassassin.service: Failed with result 'start-limit-hit'
« on: August 16, 2021, 06:15:24 PM »
Any info about this?
I search a lot for google but...lot info, any concret about fixed this issue....
Regards.
I search a lot for google but...lot info, any concret about fixed this issue....
Regards.
50
E-Mail / spamassassin.service: Failed with result 'start-limit-hit'
« on: August 01, 2021, 08:46:35 PM »
Hello,
I put this post here, mail section, because although it is from spamassassin, it is the most suitable section that I see.
I have this problem, for a long time, spamassassin only lasts a few hours, the most I have managed to work has been 3 or 4 days in a row.
At the moment that, for whatever reason, it stops working (according to the line of the subject and log that I will put below) I can no longer solve it in any way, if I do a restart, it takes just a few minutes.
I have done many tests, even another vps that I have without use, I have transferred the entirety of this server, so they have exactly the same accounts, same mails ... everything is the same, even both vps are the same (same cores, same ram) except that one is intel, the other epyc (it is the one that fails).
I have even transferred this account to another vps that has 2 accounts and added the server epyc (which are 2 accounts) in total 4, and it does not fail either.
To understand us:
Server1: Epyc (two accounts, is the one that gives me the problems).
Server2: Intel: Not used, I have transferred the whole of server1 and it does not fail.
Server3: Intel: With use of 2 accounts, I have added by transfer the 2 accounts of server1, and this does not fail.
So that they always have the same mails, through rsync I transfer the mails from server1 to server2 and server3.
I have followed the spamassassin installation wiki, also in rebuild from cwp, with Amavis / Clamd and without it.
Always, sooner or later, it ends up failing.
The other difference between server1 and server2 and server3 is that server1 is with php-fpm, the other two are not.
As I comment, same cores, same ram, same s.o. all equal.
I doubt that the cause is because it is an Epyc instead of an Intel ...
This is the log (I put 3 logs).
If when this happens, I run this statement:
Any ideas?
Thanks in advance.
I put this post here, mail section, because although it is from spamassassin, it is the most suitable section that I see.
I have this problem, for a long time, spamassassin only lasts a few hours, the most I have managed to work has been 3 or 4 days in a row.
At the moment that, for whatever reason, it stops working (according to the line of the subject and log that I will put below) I can no longer solve it in any way, if I do a restart, it takes just a few minutes.
I have done many tests, even another vps that I have without use, I have transferred the entirety of this server, so they have exactly the same accounts, same mails ... everything is the same, even both vps are the same (same cores, same ram) except that one is intel, the other epyc (it is the one that fails).
I have even transferred this account to another vps that has 2 accounts and added the server epyc (which are 2 accounts) in total 4, and it does not fail either.
To understand us:
Server1: Epyc (two accounts, is the one that gives me the problems).
Server2: Intel: Not used, I have transferred the whole of server1 and it does not fail.
Server3: Intel: With use of 2 accounts, I have added by transfer the 2 accounts of server1, and this does not fail.
So that they always have the same mails, through rsync I transfer the mails from server1 to server2 and server3.
I have followed the spamassassin installation wiki, also in rebuild from cwp, with Amavis / Clamd and without it.
Always, sooner or later, it ends up failing.
The other difference between server1 and server2 and server3 is that server1 is with php-fpm, the other two are not.
As I comment, same cores, same ram, same s.o. all equal.
I doubt that the cause is because it is an Epyc instead of an Intel ...
This is the log (I put 3 logs).
Code: [Select]
● spamassassin.service - Spamassassin daemon
Loaded: loaded (/usr/lib/systemd/system/spamassassin.service; enabled; vendor preset: disabled)
Active: failed (Result: start-limit-hit) since Thu 2021-07-15 14:00:49 CEST; 6h ago
Process: 390132 ExecStart=/usr/bin/spamd $SPAMDOPTIONS (code=exited, status=0/SUCCESS)
Main PID: 390132 (code=exited, status=0/SUCCESS)
Jul 15 14:00:49 server.mydomain.com systemd[1]: spamassassin.service: Succeeded.
Jul 15 14:00:49 server.mydomain.com systemd[1]: spamassassin.service: Service RestartSec=100ms expired, scheduling restart.
Jul 15 14:00:49 server.mydomain.com systemd[1]: spamassassin.service: Scheduled restart job, restart counter is at 9444.
Jul 15 14:00:49 server.mydomain.com systemd[1]: Stopped Spamassassin daemon.
Jul 15 14:00:49 server.mydomain.com systemd[1]: spamassassin.service: Start request repeated too quickly.
Jul 15 14:00:49 server.mydomain.com systemd[1]: spamassassin.service: Failed with result 'start-limit-hit'.
Jul 15 14:00:49 server.mydomain.com systemd[1]: Failed to start Spamassassin daemon.
● spamassassin.service - Spamassassin daemon
Loaded: loaded (/usr/lib/systemd/system/spamassassin.service; enabled; vendor preset: disabled)
Active: failed (Result: start-limit-hit) since Sun 2021-08-01 09:02:50 CEST; 1min 33s ago
Process: 3493 ExecStart=/usr/bin/spamd $SPAMDOPTIONS (code=exited, status=0/SUCCESS)
Main PID: 3493 (code=exited, status=0/SUCCESS)
Aug 01 09:02:50 server.mydomain.com systemd[1]: spamassassin.service: Service RestartSec=100ms expired, scheduling restart.
Aug 01 09:02:50 server.mydomain.com systemd[1]: spamassassin.service: Scheduled restart job, restart counter is at 115.
Aug 01 09:02:50 server.mydomain.com systemd[1]: Stopped Spamassassin daemon.
Aug 01 09:02:50 server.mydomain.com systemd[1]: spamassassin.service: Start request repeated too quickly.
Aug 01 09:02:50 server.mydomain.com systemd[1]: spamassassin.service: Failed with result 'start-limit-hit'.
Aug 01 09:02:50 server.mydomain.com systemd[1]: Failed to start Spamassassin daemon.
● spamassassin.service - Spamassassin daemon
Loaded: loaded (/usr/lib/systemd/system/spamassassin.service; enabled; vendor preset: disabled)
Active: failed (Result: start-limit-hit) since Sun 2021-08-01 20:26:21 CEST; 2h 6min ago
Process: 57453 ExecStart=/usr/bin/spamd $SPAMDOPTIONS (code=exited, status=0/SUCCESS)
Main PID: 57453 (code=exited, status=0/SUCCESS)
Aug 01 20:26:21 server.mydomain.com systemd[1]: spamassassin.service: Service RestartSec=100ms expired, scheduling restart.
Aug 01 20:26:21 server.mydomain.com systemd[1]: spamassassin.service: Scheduled restart job, restart counter is at 5916.
Aug 01 20:26:21 server.mydomain.com systemd[1]: Stopped Spamassassin daemon.
Aug 01 20:26:21 server.mydomain.com systemd[1]: spamassassin.service: Start request repeated too quickly.
Aug 01 20:26:21 server.mydomain.com systemd[1]: spamassassin.service: Failed with result 'start-limit-hit'.
Aug 01 20:26:21 server.mydomain.com systemd[1]: Failed to start Spamassassin daemon.If when this happens, I run this statement:
Code: [Select]
systemctl reset-failed spamassassinWorks again for a few hours, then crashes again.Any ideas?
Thanks in advance.
51
Postfix / Re: Dovecot failed: Permission denied
« on: July 23, 2021, 12:52:07 AM »
I know this is a old topic, sorrys.
I have exact the same issue, but in my dovecot.conf I don't have this lines:
I think is for last update, I have this problem since 2 or 3 days ago and in 2 servers.
In this moment I have some mails in Postfix Mail Queue and if view details have the same error that the initial post.
I search info but I don't view how solved this...
Regards.
I have exact the same issue, but in my dovecot.conf I don't have this lines:
Code: [Select]
service stats {
unix_listener stats-reader {
user = dovecot
group = wheel
mode = 0660
}
unix_listener stats-writer {
user = dovecot
group = wheel
mode = 0660
}
}I think is for last update, I have this problem since 2 or 3 days ago and in 2 servers.
In this moment I have some mails in Postfix Mail Queue and if view details have the same error that the initial post.
I search info but I don't view how solved this...
Regards.
52
Installation / Re: Restore cPanel account / transfer mails
« on: July 11, 2021, 10:16:57 AM »Sure, you can transfer the emails with rsync but mail accounts must be transferred by CWP.Ok, perfect!
Thanks for your reply!!!
53
Nginx / Re: Just Nginx run (wordpress permalink 404)
« on: July 06, 2021, 10:54:22 PM »
nginx does not use .htaccess file.
Follow these Steps
1) Log into the CLI via SSH
2) Cd /usr/local/cwpsrv/htdocs/resources/conf/web_servers/vhosts/nginx/php-fpm
3) " cp default.stpl wordpress.stpl " and " cp default.tpl wordpress.tpl "
4) edit wordpress.stpl and wordpress.tpl (nano/vi)
5) Find location / {
6) Immediately under that line add " try_files $uri $uri/ /index.php?$args; " and save it
7) Log into CWP Admin panelWebservers -> Webservrs Domain Config
9) Select user of domain running wordpress
10) Click Create Configuration
11) Select NGINX-PHPFPM
12) Nginx default vhost template type: Set to PHP-FPM
13) Nginx default vhost template: Select Wordpress
14) Select your desired PHP-FPM Version
15) Check box to rebuild Webserver Config for domain.
16) Click Save
Wordpress should now work
Repear 9-16 for other wordpress installations.
I "re upload" this thread just to thank you for this post, excellent post, indication and solution, really great !!
Maybe it goes off the topic of the thread a bit but ... I'm looking for a way to add the line to the .tpl so that it also listens to ipv6, add it manually, I add it and it works, but when there is a reload or similar, not being in template disappears.
Any advice?
Thank you again for this excellent post!
54
Installation / Re: Restore cPanel account / transfer mails
« on: July 04, 2021, 08:14:35 PM »
Thanks for reply.
Yes, ok, I can deleted all mails from backup manually, but then, I need transfer from cPanel to CWP, but after days, I made some tests and now, the option of transfer cPanel to CWP transfer all (include mails) without problems).
In the past (some months ago) this option not work fine, but with this new server I can transfer all good.
But continue my doubt, its possible transfer mails with rsync after the accounts are created or "semi restored"?
Sorry if I'm not explained very well.
Regards.
Yes, ok, I can deleted all mails from backup manually, but then, I need transfer from cPanel to CWP, but after days, I made some tests and now, the option of transfer cPanel to CWP transfer all (include mails) without problems).
In the past (some months ago) this option not work fine, but with this new server I can transfer all good.
But continue my doubt, its possible transfer mails with rsync after the accounts are created or "semi restored"?
Sorry if I'm not explained very well.
Regards.
55
E-Mail / Re: What best way for enabled and configure spamassasin.
« on: June 29, 2021, 12:28:20 PM »
If I understand well, the best option, ticket in CWP?
If I want by my self (I like learn) you indicate one url, the url of the wiki of cwp https://wiki.centos-webpanel.com/how-to-configure-spamassassin
is not good?
I view similar points in this other link:
https://forum.centos-webpanel.com/index.php?topic=1746.15
Regards.
If I want by my self (I like learn) you indicate one url, the url of the wiki of cwp https://wiki.centos-webpanel.com/how-to-configure-spamassassin
is not good?
I view similar points in this other link:
https://forum.centos-webpanel.com/index.php?topic=1746.15
Regards.
56
E-Mail / Re: What best way for enabled and configure spamassasin.
« on: June 29, 2021, 11:34:01 AM »
In other words, open ticket for CWP reconfigure SA and spamassasin?
Its correct?
Its correct?
57
E-Mail / What best way for enabled and configure spamassasin.
« on: June 29, 2021, 11:19:04 AM »
Hi,
I have enabled and auto start SpamAssasin like servide, but view that, I received a lot spam.
I search for the forum, I view this link: https://forum.centos-webpanel.com/index.php?topic=1746.0
But I think is some old.
I view this guide / article in the wiki cwp:
https://wiki.centos-webpanel.com/how-to-configure-spamassassin
Is from 2019, is good this guide or better made some modification about this?
Thanks in advance.
I have enabled and auto start SpamAssasin like servide, but view that, I received a lot spam.
I search for the forum, I view this link: https://forum.centos-webpanel.com/index.php?topic=1746.0
But I think is some old.
I view this guide / article in the wiki cwp:
https://wiki.centos-webpanel.com/how-to-configure-spamassassin
Is from 2019, is good this guide or better made some modification about this?
Thanks in advance.
58
E-Mail / Re: Transfer server from cPanel, problems with mail-SSL
« on: June 26, 2021, 11:23:13 AM »
Yes, I am aware that Cloudflare maintains a cache, so I do not start the tests until I see that it has cached the new IP (with a simple ping you can see it).
About it ... well it is true that I did not explain myself well, I am not native English and my English is not very fluent, as I said, I used a translator and the result is not very good.
On the contrary, SAN must be used, it is not that there was a bug and I put a patch, the explanation is difficult because I cannot find the English word that describes what I wanted to express.
It is not about going back to cPanel or hiring an advanced systems administrator, this is another panel, a sys admin can be very good and not know CWP then ...
In my job I am a sys admin, neither the best nor the worst, I only do my job the best I can.
Here, in any case, it would have been hiring a CWP administrator, but as I mentioned before, I did everything exactly the same and, the first time it did not work, the second it worked perfectly.
Difference? None, the whole process was the same.
What I did the first time that it didn't work, what do I say I didn't like? (I mean I didn't like the kind of solution I did), create a subdomain mail.mydomain.com
With it, then fine.
Now, as I did rollback, I made the transfer and I didn't need it, from the first moment, everything was fine.
Not trying or trying to go back to cPanel or sys admin advanced ...
This problem is like the AutoSSL of cPanel, the difference is the type of support, and I understand it, the price difference is also huge!
Having said all that, I like CWP and I hope to continue for a long time and migrate more servers that I have spread over the wide width of the internet ...
About it ... well it is true that I did not explain myself well, I am not native English and my English is not very fluent, as I said, I used a translator and the result is not very good.
On the contrary, SAN must be used, it is not that there was a bug and I put a patch, the explanation is difficult because I cannot find the English word that describes what I wanted to express.
It is not about going back to cPanel or hiring an advanced systems administrator, this is another panel, a sys admin can be very good and not know CWP then ...
In my job I am a sys admin, neither the best nor the worst, I only do my job the best I can.
Here, in any case, it would have been hiring a CWP administrator, but as I mentioned before, I did everything exactly the same and, the first time it did not work, the second it worked perfectly.
Difference? None, the whole process was the same.
What I did the first time that it didn't work, what do I say I didn't like? (I mean I didn't like the kind of solution I did), create a subdomain mail.mydomain.com
With it, then fine.
Now, as I did rollback, I made the transfer and I didn't need it, from the first moment, everything was fine.
Not trying or trying to go back to cPanel or sys admin advanced ...
This problem is like the AutoSSL of cPanel, the difference is the type of support, and I understand it, the price difference is also huge!
Having said all that, I like CWP and I hope to continue for a long time and migrate more servers that I have spread over the wide width of the internet ...
59
E-Mail / Re: Transfer server from cPanel, problems with mail-SSL
« on: June 25, 2021, 06:58:30 AM »
That's right, it works, and I'm happy about it too.
As I mentioned, I made the rollback and left the VPS in the starting state just to start transfer from cPanel (everything else was already configured) I made the transfer and the cPanel certificate had 3 days left.
As I mentioned, the only difference is that, this time, I did not give it to renew, but I did it automatically, when I realized it I added the SAN's .... once everything was configured, the first thing I did was change the MX record and the A "mail" record .... surprise to me that it worked ...
I did not make any other changes, well yes, the first time some change in the CWP dns records (I changed the cname of the mail record for an A record) this time no, but it lacks value when the dns from domain point, in this case , to cloudflare, then the dns are handled from there, and there all I did was change the ip of the old vps to the ip of the new one.
If it is something that I have never touched or do not know or similar, it is not difficult for me to indicate that it access CWP but it was so strange that ... either there was a bug or I had to fix it (I like to see the problem and solve it). Be careful, in a "patch" mode I had fixed it, but it was not the best option nor did I like it, having the possibility of SAN ...
Thanks for your reply.
Regards.
As I mentioned, I made the rollback and left the VPS in the starting state just to start transfer from cPanel (everything else was already configured) I made the transfer and the cPanel certificate had 3 days left.
As I mentioned, the only difference is that, this time, I did not give it to renew, but I did it automatically, when I realized it I added the SAN's .... once everything was configured, the first thing I did was change the MX record and the A "mail" record .... surprise to me that it worked ...
I did not make any other changes, well yes, the first time some change in the CWP dns records (I changed the cname of the mail record for an A record) this time no, but it lacks value when the dns from domain point, in this case , to cloudflare, then the dns are handled from there, and there all I did was change the ip of the old vps to the ip of the new one.
If it is something that I have never touched or do not know or similar, it is not difficult for me to indicate that it access CWP but it was so strange that ... either there was a bug or I had to fix it (I like to see the problem and solve it). Be careful, in a "patch" mode I had fixed it, but it was not the best option nor did I like it, having the possibility of SAN ...
Thanks for your reply.
Regards.
60
E-Mail / Re: Transfer server from cPanel, problems with mail-SSL
« on: June 24, 2021, 02:22:29 PM »
In a way I don't agree that opening new threads doesn't help. I have configured many things thanks to threads that exist.
Another thing is that the policy of CWP is that, in the face of a problem of this type, a ticket is opened and not a thread, then, there I have nothing to say because it is the policy of CWP-Studio4Host.
Surprisingly I managed to solve it, as I said at the beginning, I solved it by making a configuration that I did not like at all, I could not say the exact word in English that describes it, I am going to use a translator, so I do not know how it will be: In my language it will be I'd say "botched" but it worked.
Well, I was doing the migration tests, so just before starting the migration I had a snapshot of the VPS, they were tests because the dns in cloudflare still pointed to the old ip (except in the moments of tests that it pointed to the new one) and there I saw that problem ...
Last night, I rolled back the snapshot to make the final transfer / migration ...
Before continuing, I had the snapshot with everything configured but without any migrated accounts. SW. installed from 0, CWP installed from 0 and the necessary settings for php-fpm and php.ini.
Well, my surprise comes when, doing exactly the same as the previous time ... now yes, mail.mydomain.com now works without any problem!
I start to look for the differences, the only one, when I did tests, I renewed the SSL certificate by pressing the renew button now, this time not, as the cPanel certificate had 2 days left, shortly after finishing the migration the system automatically renewed for 90 days, I just had to add SAN for mail and webmail and .... it works! Now it works well.
Yes, between my testing and last night, there was a CWP update but ... I highly doubt it.
Maybe it was that clicking on renew now affects something? I don't think so, but it's the only thing different ...
The rest of the process, identical, first I changed dns in cloudflare pointing to the new server in the mx record and mail.mydomain.com, I saw that everything was going well and .... I already changed the rest of the dns's.
For my part there is no explanation, I do not know if when there is a transfer it is different than if we create an account of 0 in CWP, for example, if I go to Domains-List Domains none appears (and I have two active accounts-domains ), but to think something ...
For the rest, what I mentioned (and this post is long) I did the process exactly the same as the other time, except that I let the certificate renew itself and ... it worked ...
Greetings!
Another thing is that the policy of CWP is that, in the face of a problem of this type, a ticket is opened and not a thread, then, there I have nothing to say because it is the policy of CWP-Studio4Host.
Surprisingly I managed to solve it, as I said at the beginning, I solved it by making a configuration that I did not like at all, I could not say the exact word in English that describes it, I am going to use a translator, so I do not know how it will be: In my language it will be I'd say "botched" but it worked.
Well, I was doing the migration tests, so just before starting the migration I had a snapshot of the VPS, they were tests because the dns in cloudflare still pointed to the old ip (except in the moments of tests that it pointed to the new one) and there I saw that problem ...
Last night, I rolled back the snapshot to make the final transfer / migration ...
Before continuing, I had the snapshot with everything configured but without any migrated accounts. SW. installed from 0, CWP installed from 0 and the necessary settings for php-fpm and php.ini.
Well, my surprise comes when, doing exactly the same as the previous time ... now yes, mail.mydomain.com now works without any problem!
I start to look for the differences, the only one, when I did tests, I renewed the SSL certificate by pressing the renew button now, this time not, as the cPanel certificate had 2 days left, shortly after finishing the migration the system automatically renewed for 90 days, I just had to add SAN for mail and webmail and .... it works! Now it works well.
Yes, between my testing and last night, there was a CWP update but ... I highly doubt it.
Maybe it was that clicking on renew now affects something? I don't think so, but it's the only thing different ...
The rest of the process, identical, first I changed dns in cloudflare pointing to the new server in the mx record and mail.mydomain.com, I saw that everything was going well and .... I already changed the rest of the dns's.
For my part there is no explanation, I do not know if when there is a transfer it is different than if we create an account of 0 in CWP, for example, if I go to Domains-List Domains none appears (and I have two active accounts-domains ), but to think something ...
For the rest, what I mentioned (and this post is long) I did the process exactly the same as the other time, except that I let the certificate renew itself and ... it worked ...
Greetings!
