Implement occ status command via return codes

[leegarrett]

How to use GitHub

• Please use the 👍 reaction to show that you are interested into the same feature.
• Please don't comment if you have no relevant information to add. It's just extra noise for everyone subscribed to this issue.
• Subscribe to receive notifications on status change and new comments.

---

I propose the --check-if-disabled flag to occ maintenance:mode. It doesn't write any output, it just returns 0 if maintenance mode is off, and 1 if it's on.

This allows to programmatically check for maintenance mode, which makes automation easier. It allows us to use it as an conditional for the cron job (e.g. ExecCondition in the systemd unit), so the cron job will be skipped when maintenance mode is running. Currently it generates an admin email every 5 minutes during maintenance mode.

[kesselb]

Hey, sounds good to me.

https://github.com/nextcloud/server/blob/master/core/Command/Maintenance/Mode.php is the file to modify.

Would you mind sending a pull request? I guess everything you need is already in the file ;)

[solracsf]

if [ "$(occ config:system:get maintenance)" == 'false' ]; then
    //fire
fi

[luddinho]

Or the same with this occ command combined with grep:

if [ "$(occ maintenance:mode | grep -o 'enabled\|disabled')" == "disabled" ]; then
    //fire
fi

[leegarrett]

Hey, sounds good to me.

https://github.com/nextcloud/server/blob/master/core/Command/Maintenance/Mode.php is the file to modify.

Would you mind sending a pull request? I guess everything you need is already in the file ;)

Thanks, and done in PR #35830. I took a look at Core/Command/Maintenance/ModeTest.php but didn't understand the code well enough to add test coverage, too.

[MichaIng]

The two above mentioned minimally more complex methods to programmatically check for maintenance mode already now do not make this redundant?

[leegarrett]

The two above mentioned minimally more complex methods to programmatically check for maintenance mode already now do not make this redundant?

I'm aware that you can do this over a shell call. However:

• It's not always possible nor desirable to spawn a shell
• Having to spawn a shell with e.g. sh -c makes the test hard to read with double enclosure quoting, due to the need to escape inside quotes and newlines, and easy for the test to silently fail
• Depending on textual output is flawed as it's not guaranteed to be stable
• Those examples are not covered by tests, which this ideally should have.

[leegarrett]

I've done a second PR which I think is overall cleaner and better thought out. It's a parameter to ./occ status instead of maintenance:mode, and doesn't solely decide on maintenance mode. It's also possible to expand it in the future to return other (non-zero) exit codes for other critical issues that prevent normal operation.

[leegarrett]

The new PR also avoids the negation that the old one was plagued with.