Fix p2p test timeout on coverage

[azarovh]

This is the fix for the failed job that Boris reported earlier:
https://github.com/mintlayer/mintlayer-core/actions/runs/6425530789/job/17448253640?pr=1262

I see no other reason for the failure except that the syncing of 1000 blocks took more than 60s that is the standard timeout for the p2p tests. I suggest decreasing the number of processing blocks to 500 which won't make a difference for the quality of the test but allows to keep timeout the same.

[TheQuantumPhysicist]

This error has nothing to do with the 60 seconds you're seeing first. That's just rust testing framework warning that a test took more than 60 seconds.

[TheQuantumPhysicist]

The error can be seen here:

2023-10-05T23:48:04.7806169Z 2023-10-05T23:48:02.104855Z  INFO initial_download_unexpected_disconnect: mempool::pool: new tip: block Id<Block>{0x3331e96a5d554fc6ae3e7e8e873e0864d4a7b4e777b50bf5d526d8342eff92bf} height BlockHeight(895)    
2023-10-05T23:48:04.7806521Z thread 'tokio-runtime-worker' panicked at 'Failed to receive value in time: Elapsed(())', p2p/src/sync/tests/helpers/mod.rs:501:9
2023-10-05T23:48:04.7806616Z stack backtrace:
2023-10-05T23:48:04.7807188Z 2023-10-05T23:48:02.205146Z  INFO initial_download_unexpected_disconnect: mempool::pool: mempool: Processing chainstate event NewTip(Id<Block>{0x62ef6e3e55101b0e071bd2374c87081e569a69192a1cbd33f43ed67e75da5644}, BlockHeight(896))    

[azarovh]

The error can be seen here:

That's exactly the error that fires on timeout. Every test calls expect_future_val on SyncingEventReceiver with 60s timeout.

That panic happens exactly 60s from the start. Also you can see that only block at BlockHeight(895) was processed.

[TheQuantumPhysicist]

The error can be seen here:

That's exactly the error that fires on timeout. Every test calls expect_future_val on SyncingEventReceiver with 60s timeout.

That panic happens exactly 60s from the start. Also you can see that only block at BlockHeight(895) was processed.

Alrighty. I don't like the idea of reducing the test coverage just to appease a timeout. Maybe we should increase the timeout to 120 seconds. I think that's better.

[azarovh]

Alrighty. I don't like the idea of reducing the test coverage just to appease a timeout. Maybe we should increase the timeout to 120 seconds. I think that's better.

Ok if you think so