• If you are still using CentOS 7.9, it's time to convert to Alma 8 with the free centos2alma tool by Plesk or Plesk Migrator. Please let us know your experiences or concerns in this thread:
    CentOS2Alma discussion

Plesk 12.5.30 update fail! neeed help

PIRITEL

New Pleskian
Hi my server config
CentOS Linux 7.0.1406 (Core)‬
Plesk12.5.30 Actualización #10, última actualización 14/Nov/2015 00:57

yesterday was updated from 12.0.18 with latest microupdates to 12.5.30
all was fine during upgrade with exception one warning

Updating: 1:fail2ban-0.9.2-centos7.15103017.noarch [47/213]
warning: /etc/fail2ban/fail2ban.conf saved as /etc/fail2ban/fail2ban.conf.rpmsave
Job for fail2ban.service failed. See 'systemctl status fail2ban.service' and 'journalctl -xn' for details.
warning: %post(fail2ban-1:0.9.2-centos7.15103017.noarch) scriptlet failed, exit status 1
Updating: wpb-designs-20-12.5.29-43615.15081314.noarch [48/213]

and a problem that postmix mail mta wont start
plesk panel says:
ERROR: PleskUtilException
mailmng-service failed:

and a panel.log says

[2015-11-14 01:01:26] ERR [panel] mailmng-service failed: :
0: /usr/local/psa/admin/plib/Service/Agent.php:217
Service_Agent->execAndGetResponse(string 'mailmng-service', array, NULL null, array)
1: /usr/local/psa/admin/plib/Service/Driver/Mail/Plesk/Server/Abstract.php:250
Service_Driver_Mail_Plesk_Server_Abstract->_callMailMng(string 'service', array, NULL null, array)
2: /usr/local/psa/admin/plib/Service/Driver/Mail/Plesk/Server/Abstract.php:306
Service_Driver_Mail_Plesk_Server_Abstract->_callMailMngService(array)
3: /usr/local/psa/admin/plib/Service/Driver/Mail/Plesk/Server/Unix.php:377
Service_Driver_Mail_Plesk_Server_Unix->_serviceControl(string 'start', string 'smtp')
4: /usr/local/psa/admin/plib/Service/Driver/Mail/Plesk/Server/Unix.php:335
Service_Driver_Mail_Plesk_Server_Unix->startService(string 'smtp')
5: /usr/local/psa/admin/plib/Mail/Facade.php:251
Mail_Facade->serviceAction(string 'start', string 'smtp')
6: /usr/local/psa/admin/plib/ServiceControl/Unix.php:521
ServiceControl_Unix->action()
7: /usr/local/psa/admin/htdocs/server/restart_services.php:26
require(string '/usr/local/psa/admin/htdocs/server/restart_services.php')
8: /usr/local/psa/admin/htdocs/plesk.php:44
[2015-11-14 01:01:26] ERR [1] mailmng-service failed:

I change mta to qmail and reinstalled to postfix for reinstall but same results.

Many thanks.
 
yes with no luck result!!
changed mta to qmail and rechanged other time to postmail with same result and same error.
When instaled qmail the users can send mail, ask me credentials all time.

the update was not good result
finally need to restore a previus snapshot backup with 12.0.18 version because a server is functional.

until next weekend can test a new upgrade

Thanks
 
Follow the below steps.

1) Run the autoinstaller and remove fail2ban

or do the below


rpm -qa | grep fail2ban

remove all fail2ban using the below command.


rpm -e <package name>

2) Install all microupdate

/usr/local/psa/admin/sbin/autoinstaller --select-product-id plesk --select-release-current --reinstall-patch --install-component base

3) cd /etc/yum.repos.d/

make sure only centos repos are there. Move epel and atomice repos to another location

4) Upgrade now. I can assure that it will upgrade without any issue
 
Hi upgrade with your recomendations but same results
but the command #plesk repair installation works for postfix error
now the postfix is starting but mails are on queue always
not sending to users mailbox.

in maillog wont see any problem

ov 22 00:54:39 cloud dovecot: config: Warning: Killed with signal 15 (by pid=1 uid=0 code=kill)
Nov 22 00:54:39 cloud dovecot: master: Dovecot v2.2.18 starting up for imap, pop3, sieve (core dumps disabled)
Nov 22 00:54:40 cloud postfix/master[8084]: warning: process /usr/libexec/postfix/qmgr pid 8086 exit status 1
Nov 22 00:54:40 cloud postfix/master[8084]: warning: /usr/libexec/postfix/qmgr: bad command startup -- throttling
Nov 22 00:54:43 cloud /usr/lib64/plesk-9.0/psa-pc-remote[673]: PleskFilter: mi_stop=1

Nov 22 00:55:41 cloud postfix/qmgr[8865]: fatal: parameter maximal_queue_lifetime: bad time value or unit: 1dnon_smtpd_milters =
Nov 22 00:55:41 cloud postfix/cleanup[8864]: F36CC1B2B1D: message-id=<[email protected]>
Nov 22 00:55:42 cloud postfix/master[8084]: warning: process /usr/libexec/postfix/qmgr pid 8865 exit status 1
Nov 22 00:55:42 cloud postfix/master[8084]: warning: /usr/libexec/postfix/qmgr: bad command startup -- throttling
Nov 22 00:55:42 cloud postfix/pickup[8085]: 06D851B2AE5: uid=89 from=<[email protected]> orig_id=62F751B2AF2

i dont know a reason.

at momment the mail service wont work

thanks
 
Hi PIRITEL,

please use the following commands:

sudo postsuper -r ALL && service postfix restart && service dovecot restart && service pc-remote restart


If this doesn't already solved your issue, please change your mail-software from postfix to qmail and backwards to postfix again:

/usr/local/psa/admin/bin/autoinstaller --select-product-id plesk --select-release-current --install-component qmail
/usr/local/psa/admin/bin/autoinstaller --select-product-id plesk --select-release-current --install-component postfix


If you still experience issues, please use the Plesk repair utility with the "mail" option:

plesk repair mail -y

Please report back every stage and possible errors from the command line, and/or your mail.log for further investigations.
 
Thanks for quick reply
Finally was restored snapshot
This is a production server amd cant stop many time
Next weekend test your recomendation but sincerely think that have other problem
With plesk 12.0.18 all is working fine
Next weekeng 3er round

I post in forum results

Thanks
 
Hello Piritel,
I have the same config than you, and I was wondering about the upgrade.
I already have had problems with Fail2ban on my production server and had to downgrade it to 0.8 version.

I am looking for a safe way to upgrade to 12.5, so if you've got some clue about how to do that safely, I'll be very interested.

Thanks in advance for your answers, and good luck for the upgrade.

Dear Plesk developpers, it doesn't seem so easy to upgrade ...

Thanks
Best Regards
 
Hello Piritel,
I have the same config than you, and I was wondering about the upgrade.
I already have had problems with Fail2ban on my production server and had to downgrade it to 0.8 version.

I am looking for a safe way to upgrade to 12.5, so if you've got some clue about how to do that safely, I'll be very interested.

Thanks in advance for your answers, and good luck for the upgrade.

Dear Plesk developpers, it doesn't seem so easy to upgrade ...

Thanks
Best Regards


1) took psa, apsc and mysql database:

mysqldump -uadmin -p`cat /etc/psa/.psa.shadow` apsc > apscorg1.sql; mysqldump -uadmin -p`cat /etc/psa/.psa.shadow` psa > psaorg1.sql; mysqldump -uadmin -p`cat /etc/psa/.psa.shadow` mysql > mysqlorg1.sql

2) move all third party repositories to another location


/etc/yum.repos.d

3) Install microupdate

/usr/local/psa/admin/sbin/autoinstaller --select-product-id plesk --select-release-current --reinstall-patch --install-component base


4) Upgrade :

/usr/local/psa/admin/sbin/autoinstaller

If any error occurs. update the error here.
 
Hello Christy,

Thank you for your explanations. I'll do that when I'll be more available.

I'll explain here every step, I think in the next weeks
Thanks again.

Best Regards
 
@PIRITEL,

You stated:

Updating: 1:fail2ban-0.9.2-centos7.15103017.noarch [47/213]
warning: /etc/fail2ban/fail2ban.conf saved as /etc/fail2ban/fail2ban.conf.rpmsave
Job for fail2ban.service failed. See 'systemctl status fail2ban.service' and 'journalctl -xn' for details.
warning: %post(fail2ban-1:0.9.2-centos7.15103017.noarch) scriptlet failed, exit status 1

and you probably have EPEL installed and/or a default fail2ban package on your CentOS system.

Simply remove the original installation of fail2ban AND the plesk provided fail2ban package.

Re-install the Plesk provided fail2ban package and you are good to go.

If that did not help, just redo all steps, but ascertain (very carefully) that all (old) packages and config files were removed.

Furthermore, you stated:

and a problem that postmix mail mta wont start
plesk panel says:

and I must admit, it has been some time ago that I tackled this issue.

If I am not mistaken, it is a courier related issue: switch to dovecot.

If you want full certainty that everything goes well, just revert from postfix + courier to qmail + dovecot AND (afterwards) qmail + dovecot to postfix + dovecot.

Note that activating dovecot should be sufficient, but adding "postfix > qmail > postfix" effectuates a clean postfix install (and that is always good).

Regards.....
 
Back
Top