[bug] Cannot connect to substrate node after 20 retries

[lexnv]

The issue is causing the cargo clippy to fail.

First observed by the PR #516 when it was merged to master, but the PR is unrelated to the issue.

Details

/home/runner/.cargo/bin/cargo clippy --all-targets -- -D warnings
...

    Compiling test-runtime v0.20.0 (/home/runner/work/subxt/subxt/test-runtime)
error: failed to run custom build command for `test-runtime v0.20.0 (/home/runner/work/subxt/subxt/test-runtime)`

Caused by:
  process didn't exit successfully: `/home/runner/work/subxt/subxt/target/debug/build/test-runtime-b77627502b6263a0/build-script-build` (exit status: 101)
  --- stderr
  2022-04-26 08:30:47 Substrate Node    
  2022-04-26 08:30:47 ✌️  version 3.0.0-dev-59a2150    
  2022-04-26 08:30:47 ❤️  by Parity Technologies <admin@parity.io>, 2017-2022    
  2022-04-26 08:30:47 📋 Chain specification: Development    
  2022-04-26 08:30:47 🏷  Node name: evanescent-tax-4468    
  2022-04-26 08:30:47 👤 Role: AUTHORITY    
  2022-04-26 08:30:47 💾 Database: RocksDb at /tmp/substratef7L8xN/chains/dev/db/full    
  2022-04-26 08:30:47 ⛓  Native runtime: node-268 (substrate-node-0.tx2.au10)    
  2022-04-26 08:30:50 [0] 💸 generated 1 npos voters, 1 from validators and 0 nominators    
  2022-04-26 08:30:58 🔨 Initializing Genesis block/state (state: 0x0e34…b8bb, header-hash: 0xaa3a…58c1)    
  2022-04-26 08:30:58 👴 Loading GRANDPA authority set from genesis on what appears to be first startup.    
  2022-04-26 08:31:07 👶 Creating empty BABE epoch changes on what appears to be first startup.    
  2022-04-26 08:31:07 Using default protocol ID "sup" because none is configured in the chain specs    
  2022-04-26 08:31:07 🏷  Local node identity is: 12D3KooWJHiiuWNRad43dntQrJGqA36ywtKvse5GvfJGpnZuQwtx    
  2022-04-26 08:31:07 💻 Operating system: linux    
  2022-04-26 08:31:07 💻 CPU architecture: x86_64    
  2022-04-26 08:31:07 💻 Target environment: gnu    
  2022-04-26 08:31:07 💻 CPU: Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz    
  2022-04-26 08:31:07 💻 CPU cores: 2    
  2022-04-26 08:31:07 💻 Memory: 6946MB    
  2022-04-26 08:31:07 💻 Kernel: 5.13.0-1021-azure    
  2022-04-26 08:31:07 💻 Linux distribution: Ubuntu 20.04.4 LTS    
  2022-04-26 08:31:07 💻 Virtual machine: yes    
  2022-04-26 08:31:07 📦 Highest known block at #0    
  2022-04-26 08:31:07 〽️ Prometheus exporter started at 127.0.0.1:9615    
  2022-04-26 08:31:07 Listening for new connections on 127.0.0.1:45423.    
  2022-04-26 08:31:07 🏁 CPU score: 518MB/s    
  2022-04-26 08:31:07 🏁 Memory score: 7205MB/s    
  2022-04-26 08:31:07 🏁 Disk score (seq. writes): 227MB/s    
  2022-04-26 08:31:07 🏁 Disk score (rand. writes): 104MB/s    
  2022-04-26 08:31:07 👶 Starting BABE Authorship worker    
  thread 'main' panicked at 'Cannot connect to substrate node after 20 retries', test-runtime/build.rs:72:17
  note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
warning: build failed, waiting for other jobs to finish...
error: build failed
Error: The process '/home/runner/.cargo/bin/cargo' failed with exit code 101

[lexnv]

Investigation

It seems that the substrate node took around ~ 30s to spawn the RPC server.

08:30:47 Substrate Node 
...
08:31:07 Listening for new connections on 127.0.0.1:45423.  

However, on a local machine, the RPC server is usually spawned in less than 2 seconds.

The test-runtime crate performs 20 retires with 1-second sleep in between.

The issue is caused most probably by an extensive load on the CI machine.

Resolution

• Extend the retry time for the test-runtime crate.