Test failing in local environment #4110
-
|
There is a test that is consistently failing on my local machine against the latest master branch (though it passes in Github Actions). `rustc --version --verbose``cargo test --package "libp2p-allow-block-list" --all-features` |
Beta Was this translation helpful? Give feedback.
Answered by
thomaseizinger
Jun 26, 2023
Replies: 2 comments 10 replies
-
|
Can you add a logger to the test, e.g. |
Beta Was this translation helpful? Give feedback.
5 replies
-
`RUST_LOG=libp2p_swarm_test=trace cargo test tests::not_allowed_peer_cannot_dial_us --all-features` |
Beta Was this translation helpful? Give feedback.
-
|
Can you run it with |
Beta Was this translation helpful? Give feedback.
-
`RUST_BACKTRACE=1 cargo test tests::not_allowed_peer_cannot_dial_us --all-features` |
Beta Was this translation helpful? Give feedback.
-
|
Ha, this an interesting one. From your rustc output, I am guessing this is an Apple machine and you only have two network interfaces (loopback + WiFi)? What is happening here is that we wait for a certain number of events and because we weren't considerate enough when we wrote this test, it implicitly requires that we get 3 We'll have to rewrite this test to be resilient to the number of network interfaces of the machine it is running on. |
Beta Was this translation helpful? Give feedback.
Answer selected by
thomaseizinger
-
|
I've pushed a fix here: #4122 |
Beta Was this translation helpful? Give feedback.
-
|
@thomaseizinger thanks for your support on this, the test is now passing for me! `test_discovery_tokio_ipv6``test_discovery_async_std_ipv6`Also |
Beta Was this translation helpful? Give feedback.
5 replies
-
|
Would you mind pushing fixes for those along the same lines as #4122? |
Beta Was this translation helpful? Give feedback.
-
|
I haven't been able to solve this, I think that these fail because simply I don't have an ipv6 interface. Or at least, not one that doesn't start with |
Beta Was this translation helpful? Give feedback.
-
|
What machine are you running that doesn't have IPv6 in 2023? 😮 |
Beta Was this translation helpful? Give feedback.
-
I see. That is odd. Well, I am not sure we can easily fix this. These tests require an IPv6 interface 🤷♂️ |
Beta Was this translation helpful? Give feedback.
-
|
I'm using a Macbook M1 Pro, I don't believe it's too niche. I realised that the above was entirely my fault - I had ipv6 set to Theres a slightly new problem, I'll continue investigating: |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Ha, this an interesting one. From your rustc output, I am guessing this is an Apple machine and you only have two network interfaces (loopback + WiFi)?
What is happening here is that we wait for a certain number of events and because we weren't considerate enough when we wrote this test, it implicitly requires that we get 3
NewListenAddressevents. If you only have two interfaces, we'll only get two of those which is one event to few for the test to succeed.We'll have to rewrite this test to be resilient to the number of network interfaces of the machine it is running on.