Php-fpm fails to start up (502 Bad Gateway, status=226/NAMESPACE)

ℹ️ Support
1

[details=“Support intro”]

Nextcloud version (eg, 20.0.5): 22.2.3
Operating system and version (eg, Ubuntu 20.04): Manjaro
Apache or nginx version (eg, Apache 2.4.25): nginx 1.20.2
PHP version (eg, 7.4): 8.0.13

Hi there,

Not a nextcloud issue by itself, but after updating today I’m facing some strange issues. After each reboot I see a 502 when trying to access my nextcloud instance. Checking the status of php-fpm shows then the following error:
´´´
Jan 03 22:31:17 servername systemd[395]: php-fpm.service: Failed to set up mount namespacing: /run/systemd/unit-root/media/data/nextcloud: No su>
Jan 03 22:31:17 servername systemd[395]: php-fpm.service: Failed at step NAMESPACE spawning /usr/bin/php-fpm: No such file or directory
Jan 03 22:31:17 servername systemd[1]: php-fpm.service: Main process exited, code=exited, status=226/NAMESPACE
Jan 03 22:31:17 servername systemd[1]: php-fpm.service: Failed with result ‘exit-code’.
Jan 03 22:31:17 servername systemd[1]: Failed to start The PHP FastCGI Process Manager.
´´´
In order to fix this is tried restarting first php-fpm and the nginx, which both change their status then to active and running. I can then see at least a nextcloud site, but mentioning an internal server error and cryptic technical details, which I can neither find in my nginx access nor error logs.

Do you have an idea or a hint, where to start in order to solve the issue? Did e.g. anything change that would require an adapted config for php8?

Thanks for your help!

2

This should give a hint. Is php-fpm installed correctly for php8.0? I don’t know Manjaro, in Debian the executables contain the version number. I have /usr/sbin/php-fpm8.0 installed.

Also check where your php-fpm config file is are located. Check the configuration.

3

Sorry i do not use Manjaro.
But you can read perhaps this at position Install PHP Extensions.

4

Also, Manjaro seems to be mainly desktop oriented. You might want to install Debian or Ubuntu on the server. This works very reliable, and you get more community support.

5

Thank you, @eehmke and @devnull , for your hints. I guess I was able to fix the “nextcloud-related” part of my problem, so I have a working instance again.

Based on your advice and a hunch I tested whether php was working properly (after restarting) by running occ maintenance:repair. Here some errors appeared:
"An exception occurred while executing a query: SQLSTATE[HY000]: General error: 4047 InnoDB refuses to write tables with ROW_FORMAT=COMPRESSED or KEY_BLOCK_SIZE."
Here I found someone with the same error message by nextcloud and how to fix it.

Now everything works “fine”, except that after a reboot, php-fpm and nginx show the error mentioned in my original post, which goes away by simply restarting both services. Here I am still wondering what is the cause of this error.

@eehmke : Yes, you are absolutely right, it’s not server distro. As most of my other computers are now running Manjaro, I started it as an experiment and then stuck with it. I guess it’s the “safe” way of having a rolling release for a dummy like me, who is afraid of going arch. :slight_smile:

1 Like