Test swarm:multiple_addresses_err fails

[kpp]

rust-libp2p/swarm/src/lib.rs

Line 2465 in b8c3b28

assert_eq!(

$ git log -n1
commit b8c3b282eee5d6387a9cf2f64af0253e764c1172 (HEAD -> master, origin/master, origin/HEAD)
Author: Alexander Shishenko <alex@shishenko.com>
Date:   Mon Sep 5 07:31:13 2022 +0300

rust-libp2p/swarm$ while true; do RUST_BACKTRACE=1 cargo t multiple_addresses_err --all-features || break; done

...

test tests::multiple_addresses_err ... FAILED

failures:

---- tests::multiple_addresses_err stdout ----
thread 'tests::multiple_addresses_err' panicked at 'assertion failed: `(left == right)`
  left: `["/udp/16140/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/ip4/0.0.0.0/tcp/7391/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/ip4/0.0.0.0/tcp/12545/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/58510/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/5146/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/24313/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/29274/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48"]`,
 right: `["/udp/16140/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/ip4/0.0.0.0/tcp/7391/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/ip4/0.0.0.0/tcp/12545/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/ip4/0.0.0.0/tcp/12545/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/58510/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/5146/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/24313/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48", "/udp/29274/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48"]`', swarm/src/lib.rs:2465:25
stack backtrace:
   0: rust_begin_unwind
             at /rustc/4b91a6ea7258a947e59c6522cd5898e7c0a6a88f/library/std/src/panicking.rs:584:5
   1: core::panicking::panic_fmt
             at /rustc/4b91a6ea7258a947e59c6522cd5898e7c0a6a88f/library/core/src/panicking.rs:142:14
   2: core::panicking::assert_failed_inner
   3: core::panicking::assert_failed
             at /rustc/4b91a6ea7258a947e59c6522cd5898e7c0a6a88f/library/core/src/panicking.rs:181:5
   4: libp2p_swarm::tests::multiple_addresses_err::{{closure}}
             at ./src/lib.rs:2465:25
   5: <futures_util::future::poll_fn::PollFn<F> as core::future::future::Future>::poll
             at /home/kpp/.cargo/registry/src/github.com-1ecc6299db9ec823/futures-util-0.3.24/src/future/poll_fn.rs:56:9
   6: futures_executor::local_pool::block_on::{{closure}}
             at /home/kpp/.cargo/registry/src/github.com-1ecc6299db9ec823/futures-executor-0.3.24/src/local_pool.rs:317:23
   7: futures_executor::local_pool::run_executor::{{closure}}
             at /home/kpp/.cargo/registry/src/github.com-1ecc6299db9ec823/futures-executor-0.3.24/src/local_pool.rs:90:37
   8: std::thread::local::LocalKey<T>::try_with
             at /rustc/4b91a6ea7258a947e59c6522cd5898e7c0a6a88f/library/std/src/thread/local.rs:445:16
   9: std::thread::local::LocalKey<T>::with
             at /rustc/4b91a6ea7258a947e59c6522cd5898e7c0a6a88f/library/std/src/thread/local.rs:421:9
  10: futures_executor::local_pool::run_executor
             at /home/kpp/.cargo/registry/src/github.com-1ecc6299db9ec823/futures-executor-0.3.24/src/local_pool.rs:86:5
  11: futures_executor::local_pool::block_on
             at /home/kpp/.cargo/registry/src/github.com-1ecc6299db9ec823/futures-executor-0.3.24/src/local_pool.rs:317:5
  12: libp2p_swarm::tests::multiple_addresses_err
             at ./src/lib.rs:2453:9
  13: libp2p_swarm::tests::multiple_addresses_err::{{closure}}
             at ./src/lib.rs:2429:5
  14: core::ops::function::FnOnce::call_once
             at /rustc/4b91a6ea7258a947e59c6522cd5898e7c0a6a88f/library/core/src/ops/function.rs:248:5
  15: core::ops::function::FnOnce::call_once
             at /rustc/4b91a6ea7258a947e59c6522cd5898e7c0a6a88f/library/core/src/ops/function.rs:248:5
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.


failures:
    tests::multiple_addresses_err

[kpp]

Formatting left and right gives:

left: `["/udp/16140/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
  
"/ip4/0.0.0.0/tcp/7391/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
"/ip4/0.0.0.0/tcp/12545/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
"/udp/58510/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
"/udp/5146/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
"/udp/24313/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
"/udp/29274/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48"
]`,

right: `["/udp/16140/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
 
 "/ip4/0.0.0.0/tcp/7391/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
 "/ip4/0.0.0.0/tcp/12545/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
 "/ip4/0.0.0.0/tcp/12545/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
 "/udp/58510/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
 "/udp/5146/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
 "/udp/24313/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48",
 "/udp/29274/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48"
]`

So there is a duplicate /ip4/0.0.0.0/tcp/12545/p2p/1AdgJ5v3F2GQFjzyqSqgMA6UGs26X7WXA4m2nfrf1b9R48 on the right.

[thomaseizinger]

This is strange.

I see you ran the test in a loop for it to fail. How many times did you have to run it for it to break?

[thomaseizinger]

I just tried it on my computer and it took 11m for it to finally fail.

[thomaseizinger]

Ha, so this is a funny one! We are generating a total of 8 addresses to dial. If - by chance - we are generating the same port twice, we will dial the same address twice but only receive one dial error back. This is what is happening here. We can see that because "left" are the failed addresses (only 7) and right are 8 addresses, including a duplicate one.

The odd thing about the test is that the expected and actual values are IMO flipped. I'd expect (no pun intended) that "expected" is left and "actual" is right.

In any case, to make this rather rare test failure go away, we simply need to use a HashSet instead of a Vec. I am gonna prepare a PR.

[kpp]

I just tried it on my computer and it took 11m for it to finally fail.

I would have never caught this one on my own but the CI caught it right after I merged master into my branch =)

[mxinden]

Thanks for debugging here!