Update readme.md about maxDuration

[ForsakenRei]

I think we can remove this warning section since that value in .json has been changed to 10 by default nowadays?

Also removed related section in self deployment guide.

[vercel]

@ForsakenRei is attempting to deploy a commit to the github readme stats Team on Vercel.

A member of the Team first needs to authorize it.

[codecov]

Codecov Report

Patch and project coverage have no change.

Comparison is base (38c8a7a) 97.59% compared to head (3ea030a) 97.59%.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #2792   +/-   ##
=======================================
  Coverage   97.59%   97.59%           
=======================================
  Files          24       24           
  Lines        4814     4823    +9     
  Branches      447      453    +6     
=======================================
+ Hits         4698     4707    +9     
  Misses        115      115           
  Partials        1        1           

see 3 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[qwerty541]

@ForsakenRei Looks like you are right and we no longer need to keep this warning section, but i think that probably it can be useful to add inverse warning for pro plan users to remind them about ability to increase maximum execution time. What do you think @rickstaa?

[ForsakenRei]

That's a good point. Though I'm not sure if there is any benefit of increasing the max execution time, like more frequently data refresh?

[qwerty541]

That's a good point. Though I'm not sure if there is any benefit of increasing the max execution time, like more frequently data refresh?

Parameter maxDuration is an integer defining how long your serverless function should be allowed to run on every request in seconds. I think that increasing max execution time make sense when your vercel instance experiencing high load and you are using several different GitHub API tokens. Currently the application works the following way: it iterates through provided API tokens array and try to fetch card data until one of requests succeed. When most of your API tokens hit the limiter it start taking a while.

[ForsakenRei]

That's a good point. Though I'm not sure if there is any benefit of increasing the max execution time, like more frequently data refresh?

Parameter maxDuration is an integer defining how long your serverless function should be allowed to run on every request in seconds. I think that increasing max execution time make sense when your vercel instance experiencing high load and you are using several different GitHub API tokens. Currently the application works the following way: it iterates through provided API tokens array and try to fetch card data until one of requests succeed. When most of your API tokens hit the limiter it start taking a while.

I see, I'm open to this suggestion for sure.

[rickstaa]

@ForsakenRei and @qwerty541 the reason we used to have a maxDuration of 30 on the public instance was the following:

Open the vercel.json file and change the maxDuration to 10.

It's not a bug, it was intentionally reduced to 30s because by default on PRO plan it sets to 1min which is too high and was causing too much memory usage. but for non-pro users the duration can only be <10s thus 30s fails.

Originally posted by @anuraghazra in #1416 (comment)

This behaviour was simplified by @Zo-Bro-23 who created a GitHub action that changes the maxDuration on our public instance during deployment. This was done since more people are on the free Vercel plan than the Pro Vercel plan.

So, you're right @ForsakenRei, the message in the readme should be changed or removed.

[rickstaa]

@ForsakenRei and @qwerty541 the reason we used to have a maxDuration of 30 on the public instance was the following:

Open the vercel.json file and change the maxDuration to 10.
It's not a bug, it was intentionally reduced to 30s because by default on PRO plan it sets to 1min which is too high and was causing too much memory usage. but for non-pro users the duration can only be <10s thus 30s fails.
Originally posted by @anuraghazra in #1416 (comment)

This behaviour was simplified by @Zo-Bro-23 who created a GitHub action that changes the maxDuration on our public instance during deployment. This was done since more people are on the free Vercel plan than the Pro Vercel plan.

So, you're right @ForsakenRei, the message in the readme should be changed or removed.

I updated this message to remove the suggested warning, as it was incorrect. I, however, would change it to something along the line of:

Note
If you are on the Pro (i.e. paid) Vercel plan, the maxDuration value found in the Vercel.json can be increased when your Vercel instance frequently times out during the card request. You are advised to keep this value lower than 30 seconds to prevent high memory usage.

@qwerty541 I added this suggestion to this pull request. Please review 🙏🏻.