Having no tries option in the supervisor causes unlimited tries

[shivampawstickee]

• Horizon Version: 5.7.13
• Laravel Version: 8.49.2
• PHP Version: 7.4

Description:

When there is no tries option set the SupervisorOptions class defaults $maxTries to 0 which causes an infinite loop when jobs fail.

Is this intended? My hunch is it isn't because in Illuminate\Queue\WorkerOptions the default is 1

I know that tries is set in the config file by default, but it still feels incorrect to me to have unlimited retries allowed?

Happy to create a PR if it is agreed this needs addressing.

[driesvints]

Hi there,

Thanks for reporting but it looks like this is a question which can be asked on a support channel. Please only use this issue tracker for reporting bugs with the library itself. If you have a question on how to use functionality provided by this repo you can try one of the following channels:

• Laracasts Forums
• Laravel.io Forums
• StackOverflow
• Discord
• Larachat
• IRC

However, this issue will not be locked and everyone is still free to discuss solutions to your problem!

Thanks.

[shivampawstickee]

I don't see how this isn't a bug? The current behaviour means leaving the tries field out of the supervisor config causes an infinite amount of attempts for a failed job. @driesvints

[matheusb-comp]

@driesvints, I also think this is a bug.
In the Queue module of the framework, the default changed from 0 to 1 since version 6.0.

Also, in Horizon since version 3.0 the default tries in the config file changed to 1.
It just seems very strange to have a different default in Horizon only for this attribute.

[driesvints]

Hey all. Sorry for neglecting this a bit. What do you recommend for fixing this? Having a better fallback? In general we don't recommend removing the tries field or setting it to zero.

[matheusb-comp]

Having a better fallback?

I think just changing the $maxTries default value from 0 to 1, to match the $maxTries default value in Illuminate/Queue/WorkerOptions.php.

Since all the other default values are the same in both files, as @shivampawstickee said, it just does not seem intended to be 0.

[driesvints]

If you could PR that (probably to master because it'll be a breaking change) with a thorough explanation for Taylor, he could have a look 👍