Is your feature request related to a problem?

I have a test scenario where I need to access specific url's, e.g.: ['url01.com', 'url02.com', 'url03.com']

The wdio workers should share these URL's and never clash. Once a worker is finished, it "frees" up that url for another worker.

Possibly a better way to handle this is through network/load balancing, however I wanted to try this with shared-data service, setting up the url sharing between workers:

// worker A
await getValue("*"); // { 'url01.com': true, 'url02.com': true } booleans represent isUrlAvailable

// worker B
await getValue("*"); // { 'url01.com': true, 'url02.com': true }

// worker A chooses `url01.com`
await setValue("url01.com", false)
// worker B chooses `url01.com`
await setValue("url01.com", false)

// both A + B workers now have the same URL
// clash !

This is basically a shared queue management problem between node worker threads

Describe the solution you'd like.

I could implement in shared-store another server operation '/compareAndSet' to avoid collisions when writing data

compareAndSet(key, expectedValue, newValue) only sets newValue if expectedValue is the current value in the store

// worker A
await getValue("*"); // { 'url01.com': true, 'url02.com': true }

// worker B
await getValue("*"); // { 'url01.com': true, 'url02.com': true }

// worker A chooses `url01.com`
await compareAndSet("url01.com", true, false)
// worker B chooses `url01.com`
await compareAndSet("url01.com", true, false) // throws error, 'url01.com' is not `=== true` anymore
// worker B must retry another value

Describe alternatives you've considered.

I thought about a different service where we just configure a queue of resources to be shared [1, 2, 3, 4] and the operations would just be /take a resource and /return a resource, this is easier on the beforeSession for each worker since the server would automatically dispense the value the worker needs and it doesn't need to retry until a value is free

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct