Issue 502 bad gateway error after migrating to alma linux and updating plesk

[axeronb]

Server operating system version

Alama Linux 8.6

Plesk version and microupdate number

Plesk Obsidian Web Pro Edition Version 18.0.46 Update #1

I had migrated from cent os to alma linux
and then updated plesk ...everything is up to date
now I am getting
502 bad gateway error
I am getting this error while repairing

The operation failed with the following error: {"id":1,"cmd":"plesk repair installation -y","type":"progress","message":"Reconfiguring the Plesk installation","level":0} {"id":1,"cmd":"plesk repair installation -y","type":"failed","message":"Reconfiguring the Plesk installation","level":0} {"id":1,"cmd":"plesk repair installation -y","type":"failed","message":"Reconfiguring the Plesk installation","level":0,"details":"Started bootstrapper repair procedure. This may take a while.\nCertain actions may be skipped if not applicable.\n\n Finishing up upgrade procedures and rerunning previously\nfailed upgrade actions...\n===> Cumulative Plesk database upgrade and repair (revertable\nstage) has been started.\n===> Preparing Plesk database upgrade (revertable stage).\n===> Cumulative upgrade and repair of Plesk database\n(revertable stage) has been completed.\n===> Plesk database scheme upgrade has been started.\nApplying migrations from:\n\/usr\/local\/psa\/bootstrapper\/pp18.0.46-bootstrapper\/migrations\/\n===> Plesk database scheme upgrade has been completed.\n===> Cumulative Plesk upgrade and repair (final stage) has been\nstarted.\n===> Preparing Plesk upgrade (final stage).\n===> Cumulative upgrade and repair of Plesk (final stage) has\nbeen completed.\n Reconfiguring mail subsystem...\n Reconfiguring Apache web server...\n Reconfiguring ProFTPD FTP server...\n===> Configuring ProFTPD server\n Reconfiguring AWStats web statistics...\n Reconfiguring WatchDog...\n Restoring SELinux contexts...\nSynchronizing state of sw-cp-server.service with SysV service\nscript with \/usr\/lib\/systemd\/systemd-sysv-install.\nExecuting: \/usr\/lib\/systemd\/systemd-sysv-install enable\nsw-cp-server\nSynchronizing state of sw-engine.service with SysV service\nscript with \/usr\/lib\/systemd\/systemd-sysv-install.\nExecuting: \/usr\/lib\/systemd\/systemd-sysv-install enable\nsw-engine\nSynchronizing state of dovecot.service with SysV service script\nwith \/usr\/lib\/systemd\/systemd-sysv-install.\nExecuting: \/usr\/lib\/systemd\/systemd-sysv-install enable dovecot\nCreated symlink\n\/etc\/systemd\/system\/multi-user.target.wants\/spamassassin.service\n-> \/usr\/lib\/systemd\/system\/spamassassin.service.\nSynchronizing state of nginx.service with SysV service script\nwith \/usr\/lib\/systemd\/systemd-sysv-install.\nExecuting: \/usr\/lib\/systemd\/systemd-sysv-install enable nginx\nSynchronizing state of psa.service with SysV service script\nwith \/usr\/lib\/systemd\/systemd-sysv-install.\nExecuting: \/usr\/lib\/systemd\/systemd-sysv-install enable psa\n Reconfiguring SSL ciphers and protocols...\n Regenerating web servers' configuration files...\n Cleaning active Panel sessions...\n Fixing permissions on Panel packages files\nERROR:[Errno 2] No such file or directory:\n'\/opt\/plesk\/node\/12\/lib\/node_modules\/npm\/appveyor.yml'\nChanging permissions on \/var\/run\/dovecot\/login to 0o755\n\nBootstrapper repair finished.\nErrors occurred while performing the following actions: fix\nPlesk packages permissions.\nCheck '\/var\/log\/plesk\/install\/plesk_18.0.46_repair.log' and\n'\/var\/log\/plesk\/install\/plesk_18.0.46_repair_problems.log' for\ndetails.\nIf you can't resolve the issue on your own, please address\nPlesk support.\n"}

 

[Bitpalast]

Could this have something to do with

https://support.plesk.com/hc/en-us/articles/4406125133842-Unable-to-use-NPM-install-for-Node-js-application-in-Plesk-usr-bin-env-node-No-such-file-or-directory-

 

[axeronb]

When I restart it works but within some minutes its goes down again ..saying bad gateway..
but I can access the Plesk CP @Peter Debik

 

[Bitpalast]

That symptom is typical for Fail2Ban. Have you checked that your server's public IP address(es) are whitelisted in your Fail2Ban configuration?

 

[axeronb]

Yes it is listed as trusted IP address @Peter Debik
Either I have to restart the server or check/ repair the server to get the websites up again.

 

[Bitpalast]

Have you checked /var/log/fail2ban.log for details? Do you see local IP address(es) listed there? For example 127.0.0.1 or your public IP addresses? If not, another not so likely cause could be either a crash of a PHP-FPM service or exceeding the max number of allowed children of such a service. That could be seen in a # service plesk-php-<number>-fpm status or the corresponding logs.

 

[axeronb]

Nope, not listed there.
I even disabled Fail2Ban and checked
the server shuts down showing 504 error after some 12-20 mins after running Plesk repair or restarting the server
I disabled unsused FPM too... can you please help!

 

[axeronb]

systemd [4540] System daemons [3]

Failed to canonicalize path '/usr/local/psa/admin/.config/systemd/user/systemd-exit.service.d': Permission denied

getting this error... @Peter Debik

 

[Bitpalast]

So we are back to my first guess that the Installation of Node.js is faulty.
Same issue described here Issue - Strange systemd messages piling up
however, the workaround described is only a cure for the symptoms, not the cause. The cause seem to be missing file access permissions. I am personally unexperienced with Node.js at this time, so I cannot provide the correct solution. But probably someone else who knows more about Node.js can.