This message was deleted.

• Is 2 minutes and 34 seconds for storing configuration too long? Is there something wrong?

For the pure storing, probably. But I guess you have a big project and the 2.5 minutes are the actual configuration time that is then saved by reusing the configuration cache. Or how long is configuration time without configuration cache?

• Is the behavior of not recognizing a configuration cache between the sync and the run intended, or is there something wrong?

Nothing is wrong. The configuration cache is stored per requested set of tasks. So if you run

gw assemble

and then again

gw assemble

, the cache entry is reused. If you run

gw assemble

and then

gw assemble build

, the set of requested tasks is different and thus there is not cache entry that could be reused.

@Vampire the time only of storing configuration cache, configuration time previously toke 18s, so is it normal ?

Given your observations are correct, I don't think an incresase of 855% is normal, no. You should probably post a link to this conversation to #C013WEPGQF9 to maybe get some more targeted attention. If you can provide a reproducer that shows this difference it would be optimal and maybe could be posted to an issue too.

@Vampire thank you for your fast response, so I can share the scan report for configuration cache enabled and the before is configuration cache disabled? is that what you meant of showing the diff?

That's certainly one option, yes. Not an MCVE, but better than nothing and might already provide valuable information. Depending on how low-level you want to get you can also use the

gradle-profiler

to find out where exactly the time is spent.