Skip to content

Nightly -- Full Network Tests #24

Nightly -- Full Network Tests

Nightly -- Full Network Tests #24

Workflow file for this run

name: Nightly -- Full Network Tests
on:
schedule:
- cron: '0 0 * * *'
workflow_dispatch:
env:
CARGO_INCREMENTAL: 0 # bookkeeping for incremental builds has overhead, not useful in CI.
WORKFLOW_URL: https://github.com/maidsafe/stableset_net/actions/runs
jobs:
e2e:
name: E2E tests
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: [ubuntu-latest, windows-latest, macos-latest]
steps:
- uses: actions/checkout@v3
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
toolchain: stable
- uses: Swatinem/rust-cache@v2
continue-on-error: true
- name: Build node and client
run: cargo build --release --bin safenode --bin safe --bin faucet
timeout-minutes: 30
- name: Start a local network
uses: maidsafe/sn-local-testnet-action@main
with:
action: start
interval: 2000
node-path: target/release/safenode
faucet-path: target/release/faucet
platform: ${{ matrix.os }}
- name: Check contact peer
shell: bash
run: echo "Peer is $SAFE_PEERS"
- name: Create and fund a wallet to pay for files storage
run: |
cargo run --bin faucet --release -- --log-output-dest=data-dir send 1000000 $(cargo run --bin safe --release -- --log-output-dest=data-dir wallet address | tail -n 1) | tail -n 1 > dbc_hex
cat dbc_hex | cargo run --bin safe --release -- --log-output-dest=data-dir wallet deposit --stdin
env:
SN_LOG: "all"
timeout-minutes: 2
- name: Start a client to pay for files storage
run: cargo run --bin safe --release -- --log-output-dest=data-dir wallet pay "./resources"
env:
SN_LOG: "all"
timeout-minutes: 2
- name: Start a client to carry out chunk actions
run: cargo run --bin safe --release -- --log-output-dest=data-dir files upload "./resources"
env:
SN_LOG: "all"
timeout-minutes: 2
- name: Start a client to create a register
run: cargo run --bin safe --release -- --log-output-dest=data-dir register create baobao
env:
SN_LOG: "all"
timeout-minutes: 2
- name: Start a client to get a register
run: cargo run --bin safe --release -- --log-output-dest=data-dir register get -n baobao
env:
SN_LOG: "all"
timeout-minutes: 2
- name: Start a client to edit a register
run: cargo run --bin safe --release -- --log-output-dest=data-dir register edit -n baobao wood
env:
SN_LOG: "all"
timeout-minutes: 2
- name: Stop the local network and upload logs
if: always()
uses: maidsafe/sn-local-testnet-action@main
with:
action: stop
log_file_prefix: safe_test_logs_e2e
platform: ${{ matrix.os }}
- name: post notification to slack on failure
if: ${{ failure() }}
uses: bryannice/[email protected]
env:
SLACK_INCOMING_WEBHOOK: ${{ secrets.SLACK_GH_ACTIONS_WEBHOOK_URL }}
SLACK_MESSAGE: "Please check the logs for the run at ${{ env.WORKFLOW_URL }}/${{ github.run_id }}"
SLACK_TITLE: "Nightly E2E Test Run Failed"
full_unit:
name: Full Unit Tests (including proptests)
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: [ubuntu-latest, windows-latest, macos-latest]
steps:
- uses: actions/checkout@v3
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
toolchain: stable
- uses: Swatinem/rust-cache@v2
continue-on-error: true
- name: Build unit tests before running
run: cargo test --release --lib --bins --no-run
timeout-minutes: 30
- name: Run testnet tests
timeout-minutes: 25
run: cargo test --release --package sn_testnet
- name: Run network tests
timeout-minutes: 25
run: cargo test --release -p sn_networking
- name: Run protocol tests
timeout-minutes: 25
run: cargo test --release -p sn_protocol
- name: Run transfers tests
timeout-minutes: 25
run: cargo test --release --package sn_transfers
- name: Run register tests
shell: bash
timeout-minutes: 50
env:
PROPTEST_CASES: 512
run: cargo test --release -p sn_registers
- name: post notification to slack on failure
if: ${{ failure() }}
uses: bryannice/[email protected]
env:
SLACK_INCOMING_WEBHOOK: ${{ secrets.SLACK_GH_ACTIONS_WEBHOOK_URL }}
SLACK_MESSAGE: "Please check the logs for the run at ${{ env.WORKFLOW_URL }}/${{ github.run_id }}"
SLACK_TITLE: "Nightly Unit Test Run Failed"
spend_test:
name: dbc spend tests against network
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: [ubuntu-latest, windows-latest, macos-latest]
steps:
- uses: actions/checkout@v3
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
toolchain: stable
- uses: Swatinem/rust-cache@v2
continue-on-error: true
- name: Build safe bins
run: cargo build --release --features local-discovery --bin safenode --bin faucet
timeout-minutes: 30
- name: Build testing executable
run: cargo test --release -p sn_node --features=local-discovery --test sequential_transfers --test storage_payments --no-run
timeout-minutes: 30
- name: Start a local network
uses: maidsafe/sn-local-testnet-action@main
with:
action: start
interval: 2000
node-path: target/release/safenode
faucet-path: target/release/faucet
platform: ${{ matrix.os }}
- name: execute the dbc spend test
run: cargo test --release --features="local-discovery" --test sequential_transfers -- --nocapture
env:
SN_LOG: "all"
timeout-minutes: 10
- name: execute the storage payment tests
run: cargo test --release --features="local-discovery" --test storage_payments -- --nocapture
env:
SN_LOG: "all"
timeout-minutes: 10
- name: Stop the local network and upload logs
if: always()
uses: maidsafe/sn-local-testnet-action@main
with:
action: stop
log_file_prefix: safe_test_logs_spend
platform: ${{ matrix.os }}
- name: post notification to slack on failure
if: ${{ failure() }}
uses: bryannice/[email protected]
env:
SLACK_INCOMING_WEBHOOK: ${{ secrets.SLACK_GH_ACTIONS_WEBHOOK_URL }}
SLACK_MESSAGE: "Please check the logs for the run at ${{ env.WORKFLOW_URL }}/${{ github.run_id }}"
SLACK_TITLE: "Nightly Spend Test Run Failed"
churn:
name: Network churning tests
runs-on: ${{ matrix.os }}
strategy:
matrix:
include:
- os: ubuntu-latest
node_data_path: /home/runner/.local/share/safe/node
- os: windows-latest
node_data_path: C:\\Users\\runneradmin\\AppData\\Roaming\\safe\\node
- os: macos-latest
node_data_path: /Users/runner/Library/Application Support/safe/node
steps:
- uses: actions/checkout@v3
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
toolchain: stable
- uses: Swatinem/rust-cache@v2
continue-on-error: true
- name: Build safe bins
run: cargo build --release --features local-discovery --bin safenode --bin faucet
timeout-minutes: 30
- name: Build churn tests
run: cargo test --release -p sn_node --features=local-discovery --test data_with_churn --no-run
timeout-minutes: 30
- name: Start a local network
uses: maidsafe/sn-local-testnet-action@main
with:
action: start
interval: 2000
node-path: target/release/safenode
faucet-path: target/release/faucet
platform: ${{ matrix.os }}
- name: Chunks data integrity during nodes churn (during 10min) (in theory)
run: cargo test --release -p sn_node --features="local-discovery" --test data_with_churn -- --nocapture
env:
TEST_DURATION_MINS: 60
TEST_CHURN_CYCLES: 6
SN_LOG: "all"
timeout-minutes: 70
- name: Verify restart of nodes using rg
shell: bash
timeout-minutes: 1
# get the counts, then the specific line, and then the digit count only
# then check we have an expected level of restarts
# TODO: make this use an env var, or relate to testnet size
run : |
restart_count=$(rg "Node is restarting in" "${{ matrix.node_data_path }}" -c --stats | \
rg "(\d+) matches" | rg "\d+" -o)
echo "Restart $restart_count nodes"
peer_removed=$(rg "PeerRemovedFromRoutingTable" "${{ matrix.node_data_path }}" -c --stats | \
rg "(\d+) matches" | rg "\d+" -o)
echo "PeerRemovedFromRoutingTable $peer_removed times"
if [ $peer_removed -lt $restart_count ]; then
echo "PeerRemovedFromRoutingTable times of: $peer_removed is less than the restart count of: $restart_count"
exit 1
fi
node_count=$(ls "${{ matrix.node_data_path }}" | wc -l)
echo "Node dir count is $node_count"
# TODO: reenable this once the testnet dir creation is tidied up to avoid a large count here
# if [ $restart_count -lt $node_count ]; then
# echo "Restart count of: $restart_count is less than the node count of: $node_count"
# exit 1
# fi
- name: Verify data replication using rg
shell: bash
timeout-minutes: 1
# get the counts, then the specific line, and then the digit count only
# then check we have an expected level of replication
# TODO: make this use an env var, or relate to testnet size
run : |
fetching_attempt_count=$(rg "FetchingKeysForReplication" "${{ matrix.node_data_path }}" -c --stats | \
rg "(\d+) matches" | rg "\d+" -o)
echo "Carried out $fetching_attempt_count fetching attempts"
node_count=$(ls "${{ matrix.node_data_path }}" | wc -l)
if [ $fetching_attempt_count -lt $node_count ]; then
echo "Replication fetching attempts of: $fetching_attempt_count is less than the node count of: $node_count"
exit 1
fi
- name: Stop the local network and upload logs
if: always()
uses: maidsafe/sn-local-testnet-action@main
with:
action: stop
log_file_prefix: safe_test_logs_churn
platform: ${{ matrix.os }}
- name: post notification to slack on failure
if: ${{ failure() }}
uses: bryannice/[email protected]
env:
SLACK_INCOMING_WEBHOOK: ${{ secrets.SLACK_GH_ACTIONS_WEBHOOK_URL }}
SLACK_MESSAGE: "Please check the logs for the run at ${{ env.WORKFLOW_URL }}/${{ github.run_id }}"
SLACK_TITLE: "Nightly Churn Test Run Failed"
verify_data_location:
name: Verify data location
runs-on: ${{ matrix.os }}
strategy:
matrix:
include:
- os: ubuntu-latest
node_data_path: /home/runner/.local/share/safe/node
- os: windows-latest
node_data_path: C:\\Users\\runneradmin\\AppData\\Roaming\\safe\\node
- os: macos-latest
node_data_path: /Users/runner/Library/Application Support/safe/node
steps:
- uses: actions/checkout@v3
- name: Install Rust
uses: dtolnay/rust-toolchain@stable
with:
toolchain: stable
- uses: Swatinem/rust-cache@v2
continue-on-error: true
- name: Build safe bins
run: cargo build --release --features local-discovery --bin safenode --bin faucet
timeout-minutes: 30
- name: Build data location test
run: cargo test --release -p sn_node --features=local-discovery --test verify_data_location --no-run
timeout-minutes: 30
- name: Start a local network
uses: maidsafe/sn-local-testnet-action@main
with:
action: start
interval: 2000
node-path: target/release/safenode
faucet-path: target/release/faucet
platform: ${{ matrix.os }}
- name: Verify the location of the data on the network (approx 12 * 5 mins)
run: cargo test --release -p sn_node --features="local-discovery" --test verify_data_location -- --nocapture
env:
CHURN_COUNT: 12
SN_LOG: "all"
timeout-minutes: 70
- name: Verify restart of nodes using rg
shell: bash
timeout-minutes: 1
# get the counts, then the specific line, and then the digit count only
# then check we have an expected level of restarts
# TODO: make this use an env var, or relate to testnet size
run : |
restart_count=$(rg "Node is restarting in" "${{ matrix.node_data_path }}" -c --stats | \
rg "(\d+) matches" | rg "\d+" -o)
echo "Restart $restart_count nodes"
peer_removed=$(rg "PeerRemovedFromRoutingTable" "${{ matrix.node_data_path }}" -c --stats | \
rg "(\d+) matches" | rg "\d+" -o)
echo "PeerRemovedFromRoutingTable $peer_removed times"
if [ $peer_removed -lt $restart_count ]; then
echo "PeerRemovedFromRoutingTable times of: $peer_removed is less than the restart count of: $restart_count"
exit 1
fi
node_count=$(ls "${{ matrix.node_data_path }}" | wc -l)
echo "Node dir count is $node_count"
- name: Stop the local network and upload logs
if: always()
uses: maidsafe/sn-local-testnet-action@main
with:
action: stop
log_file_prefix: safe_test_logs_data_location
platform: ${{ matrix.os }}
- name: post notification to slack on failure
if: ${{ failure() }}
uses: bryannice/[email protected]
env:
SLACK_INCOMING_WEBHOOK: ${{ secrets.SLACK_GH_ACTIONS_WEBHOOK_URL }}
SLACK_MESSAGE: "Please check the logs for the run at ${{ env.WORKFLOW_URL }}/${{ github.run_id }}"
SLACK_TITLE: "Nightly Data Location Test Run Failed"