version upgrade catchup test

.github/scripts/check_finalization.sh

@@ -1,12 +1,12 @@
 #!/bin/bash
 
-RPC_HOST=127.0.0.1
-RPC_PORT=9933
+RPC_HOST=${RPC_HOST:-127.0.0.1}
+RPC_PORT=${RPC_PORT:-9933}
 LAST_FINALIZED=""
-VALIDATOR=damian
+VALIDATOR=${VALIDATOR:-damian}
 
 while [[ "$LAST_FINALIZED" =~ "0x0" ]] || [[ -z "$LAST_FINALIZED" ]]; do
-  block_hash=$(docker run --network container:$VALIDATOR appropriate/curl:latest \
+  block_hash=$(docker run --rm --network container:$VALIDATOR appropriate/curl:latest \
                       -H "Content-Type: application/json" \
                       -d '{"id":1, "jsonrpc":"2.0", "method": "chain_getFinalizedHead"}' http://$RPC_HOST:$RPC_PORT | jq '.result')
   ret_val=$?
@@ -15,7 +15,7 @@ while [[ "$LAST_FINALIZED" =~ "0x0" ]] || [[ -z "$LAST_FINALIZED" ]]; do
     continue
   fi
 
-  finalized_block=$(docker run --network container:$VALIDATOR appropriate/curl:latest \
+  finalized_block=$(docker run --rm --network container:$VALIDATOR appropriate/curl:latest \
                            -H "Content-Type: application/json" \
                            -d '{"id":1, "jsonrpc":"2.0", "method": "chain_getBlock", "params": ['$block_hash']}' http://$RPC_HOST:$RPC_PORT | jq '.result.block.header.number')
 
@@ -25,9 +25,9 @@ while [[ "$LAST_FINALIZED" =~ "0x0" ]] || [[ -z "$LAST_FINALIZED" ]]; do
     continue
   else
     LAST_FINALIZED=$finalized_block
-    echo "Last finalized block number: $LAST_FINALIZED"
   fi
 
 done
 
+echo "Last finalized block number: $LAST_FINALIZED"
 exit $?

.github/scripts/run_consensus.sh

@@ -6,6 +6,7 @@ set -euo pipefail
 # change when increasing the number of node containers
 NODE_COUNT=5
 MIN_VALIDATOR_COUNT=4
+DOCKER_COMPOSE=${DOCKER_COMPOSE:-"docker/docker-compose.yml"}
 OVERRIDE_DOCKER_COMPOSE=${OVERRIDE_DOCKER_COMPOSE:-""}
 
 # default minimum validator count
@@ -83,19 +84,20 @@ function generate_bootnode_peer_id {
 
 function run_containers {
   local authorities_count="$1"
-  local override_file="$2"
+  local docker_compose_file="$2"
+  local override_file="$3"
 
   echo "Running ${authorities_count} containers..."
   if [[ -z ${override_file} ]]; then
-      docker-compose -f docker/docker-compose.yml up -d
+      docker-compose -f "${docker_compose_file}" up -d
   else
-      docker-compose -f docker/docker-compose.yml -f "${override_file}" up -d
+      docker-compose -f "${docker_compose_file}" -f "${override_file}" up -d
   fi
 }
 
 authorities=$(generate_authorities ${NODE_COUNT})
 generate_chainspec "${authorities[@]}" "${MIN_VALIDATOR_COUNT}"
 generate_bootnode_peer_id ${authorities[0]}
-run_containers ${NODE_COUNT} "${OVERRIDE_DOCKER_COMPOSE}"
+run_containers ${NODE_COUNT} "${DOCKER_COMPOSE}" "${OVERRIDE_DOCKER_COMPOSE}"
 
 exit $?

.github/workflows/e2e-tests-main-devnet.yml

@@ -54,6 +54,36 @@ jobs:
           retention-days: 7
 
 
+  build-cliain-image:
+    name: Build docker image for cliain
+    runs-on: ubuntu-20.04
+    steps:
+      - name: GIT | Checkout source code
+        uses: actions/checkout@v2
+
+      - name: Install Rust toolchain
+        uses: actions-rs/toolchain@v1
+
+      - name: Cargo | Build release binary
+        run: |
+          cd ./bin/cliain && cargo build --release
+
+      - name: Build docker image
+        run: |
+          cd ./bin/cliain
+          chmod +x target/release/cliain
+          docker build --tag cliain:latest -f ./Dockerfile .
+          docker save -o cliain.tar cliain:latest
+
+      - name: Upload test docker image
+        uses: actions/upload-artifact@v2
+        with:
+          name: cliain-docker
+          path: ./bin/cliain/cliain.tar
+          if-no-files-found: error
+          retention-days: 7
+
+
   check-determinism:
     needs: [build-new-node]
     name: Verify runtime build determinism
@@ -279,6 +309,7 @@ jobs:
           test-case: fee_calculation
         timeout-minutes: 2
 
+
   run-e2e-validators-rotate:
     needs: [build-test-docker, build-test-client]
     name: Run validators rotation test
@@ -485,6 +516,130 @@ jobs:
           follow-up-finalization-check: true
         timeout-minutes: 15
 
+  run-e2e-version-upgrade-catchup-one-node-late:
+    needs: [build-test-docker, build-cliain-image]
+    name: Run a test where one of the nodes learns of version-upgrade during catchup
+    runs-on: ubuntu-20.04
+    steps:
+      - name: Checkout source code
+        uses: actions/checkout@v2
+
+      - name: Download artifact with docker image for aleph-node
+        uses: actions/download-artifact@v2
+        with:
+          name: aleph-test-docker
+
+      - name: Load node docker image
+        shell: bash
+        run: docker load -i aleph-node.tar
+
+      - name: Download artifact with docker image for cliain
+        uses: actions/download-artifact@v2
+        with:
+          name: cliain-docker
+
+      - name: Load cliain docker image
+        shell: bash
+        run: docker load -i cliain.tar
+
+      - name: Call catchup_test.sh
+        run: |
+          chmod +x ./scripts/catchup_version_upgrade_test.sh
+          UPGRADE_ROUND=31 NODES="Node1" PORTS="9934" EXT_STATUS="finalized" DOCKER_COMPOSE=docker/docker-compose.bridged.yml ./scripts/catchup_version_upgrade_test.sh
+
+  run-e2e-version-upgrade-catchup-two-nodes-late:
+    needs: [build-test-docker, build-cliain-image]
+    name: Run a test where two nodes learn of version-upgrade during catchup
+    runs-on: ubuntu-20.04
+    steps:
+      - name: Checkout source code
+        uses: actions/checkout@v2
+
+      - name: Download artifact with docker image for aleph-node
+        uses: actions/download-artifact@v2
+        with:
+          name: aleph-test-docker
+
+      - name: Load node docker image
+        shell: bash
+        run: docker load -i aleph-node.tar
+
+      - name: Download artifact with docker image for cliain
+        uses: actions/download-artifact@v2
+        with:
+          name: cliain-docker
+
+      - name: Load cliain docker image
+        shell: bash
+        run: docker load -i cliain.tar
+
+      - name: Call catchup_test.sh
+        run: |
+          chmod +x ./scripts/catchup_version_upgrade_test.sh
+          UPGRADE_ROUND=31 NODES="Node1:Node2" PORTS="9934:9935" EXT_STATUS="in-block" DOCKER_COMPOSE=docker/docker-compose.bridged.yml ./scripts/catchup_version_upgrade_test.sh
+
+  run-e2e-version-upgrade-early-catchup-one-node-late:
+    needs: [build-test-docker, build-cliain-image]
+    name: Run a test where one of the nodes learns of version-upgrade before catchup
+    runs-on: ubuntu-20.04
+    steps:
+      - name: Checkout source code
+        uses: actions/checkout@v2
+
+      - name: Download artifact with docker image for aleph-node
+        uses: actions/download-artifact@v2
+        with:
+          name: aleph-test-docker
+
+      - name: Load node docker image
+        shell: bash
+        run: docker load -i aleph-node.tar
+
+      - name: Download artifact with docker image for cliain
+        uses: actions/download-artifact@v2
+        with:
+          name: cliain-docker
+
+      - name: Load cliain docker image
+        shell: bash
+        run: docker load -i cliain.tar
+
+      - name: Call catchup_test.sh
+        run: |
+          chmod +x ./scripts/catchup_version_upgrade_test.sh
+          UPGRADE_BEFORE_DISABLE=true UPGRADE_ROUND=31 NODES="Node1" PORTS="9934" EXT_STATUS="finalized" DOCKER_COMPOSE=docker/docker-compose.bridged.yml ./scripts/catchup_version_upgrade_test.sh
+
+  run-e2e-version-upgrade-early-catchup-two-nodes-late:
+    needs: [build-test-docker, build-cliain-image]
+    name: Run a test where two nodes learn of version-upgrade before catchup
+    runs-on: ubuntu-20.04
+    steps:
+      - name: Checkout source code
+        uses: actions/checkout@v2
+
+      - name: Download artifact with docker image for aleph-node
+        uses: actions/download-artifact@v2
+        with:
+          name: aleph-test-docker
+
+      - name: Load node docker image
+        shell: bash
+        run: docker load -i aleph-node.tar
+
+      - name: Download artifact with docker image for cliain
+        uses: actions/download-artifact@v2
+        with:
+          name: cliain-docker
+
+      - name: Load cliain docker image
+        shell: bash
+        run: docker load -i cliain.tar
+
+      - name: Call catchup_test.sh
+        run: |
+          chmod +x ./scripts/catchup_version_upgrade_test.sh
+          UPGRADE_BEFORE_DISABLE=true UPGRADE_ROUND=31 NODES="Node1:Node2" PORTS="9934:9935" EXT_STATUS="finalized" DOCKER_COMPOSE=docker/docker-compose.bridged.yml ./scripts/catchup_version_upgrade_test.sh
+
   check-e2e-test-suite-completion:
     needs: [
       run-e2e-finalization-test,
@@ -510,6 +665,10 @@ jobs:
       run-e2e-ban-manual,
       run-e2e-version-upgrade,
       run-e2e-failing-version-upgrade,
+      run-e2e-version-upgrade-catchup-one-node-late,
+      run-e2e-version-upgrade-catchup-two-nodes-late,
+      run-e2e-version-upgrade-early-catchup-one-node-late,
+      run-e2e-version-upgrade-early-catchup-two-nodes-late,
     ]
     name: Check e2e test suite completion
     runs-on: ubuntu-20.04

aleph-client/src/version_upgrade.rs

@@ -10,6 +10,7 @@ pub fn schedule_upgrade(
     connection: &RootConnection,
     version: Version,
     session: SessionIndex,
+    state: XtStatus,
 ) -> Result<(), ApiClientError> {
     let connection = connection.as_connection();
     let upgrade_call = compose_call!(
@@ -30,7 +31,7 @@ pub fn schedule_upgrade(
         &connection,
         xt,
         Some("schedule finality version change"),
-        XtStatus::Finalized,
+        state,
     )
     .map(|_| ())
 }

bin/cliain/src/commands.rs

@@ -3,9 +3,9 @@ use std::{
     path::{Path, PathBuf},
 };
 
-use aleph_client::BlockNumber;
-use clap::{Args, Subcommand};
-use primitives::{Balance, CommitteeSeats};
+use aleph_client::{BlockNumber, XtStatus};
+use clap::{clap_derive::ValueEnum, Args, Subcommand};
+use primitives::{Balance, CommitteeSeats, SessionIndex};
 use serde::{Deserialize, Serialize};
 use sp_core::H256;
 use substrate_api_client::AccountId;
@@ -111,6 +111,8 @@ pub struct ChangeValidatorArgs {
     pub committee_size: Option<CommitteeSeats>,
 }
 
+pub type Version = u32;
+
 impl std::str::FromStr for ChangeValidatorArgs {
     type Err = serde_json::Error;
 
@@ -124,6 +126,21 @@ impl std::str::FromStr for ChangeValidatorArgs {
     }
 }
 
+#[derive(Debug, Clone, ValueEnum)]
+pub enum ExtrinsicState {
+    InBlock,
+    Finalized,
+}
+
+impl From<ExtrinsicState> for XtStatus {
+    fn from(state: ExtrinsicState) -> Self {
+        match state {
+            ExtrinsicState::InBlock => XtStatus::InBlock,
+            ExtrinsicState::Finalized => XtStatus::Finalized,
+        }
+    }
+}
+
 #[derive(Debug, Clone, Subcommand)]
 pub enum Command {
     /// Staking call to bond stash with controller
@@ -336,4 +353,16 @@ pub enum Command {
     /// Code can only be removed by its original uploader (its owner) and only if it is not used by any contract.
     /// API signature: https://polkadot.js.org/docs/substrate/extrinsics/#removecodecode_hash-h256
     ContractRemoveCode(ContractRemoveCode),
+
+    /// Schedules a version upgrade of the network.
+    VersionUpgradeSchedule {
+        #[clap(long)]
+        version: Version,
+
+        #[clap(long)]
+        session: SessionIndex,
+
+        #[clap(long, value_enum, default_value_t=ExtrinsicState::Finalized)]
+        expected_state: ExtrinsicState,
+    },
 }

bin/cliain/src/lib.rs

@@ -8,6 +8,7 @@ mod staking;
 mod transfer;
 mod treasury;
 mod validators;
+mod version_upgrade;
 mod vesting;
 
 use aleph_client::{
@@ -27,6 +28,7 @@ pub use treasury::{
     approve as treasury_approve, propose as treasury_propose, reject as treasury_reject,
 };
 pub use validators::change_validators;
+pub use version_upgrade::schedule_upgrade;
 pub use vesting::{vest, vest_other, vested_transfer};
 
 pub struct ConnectionConfig {

bin/cliain/src/main.rs

@@ -8,9 +8,9 @@ use clap::Parser;
 use cliain::{
     bond, call, change_validators, finalize, force_new_era, instantiate, instantiate_with_code,
     next_session_keys, nominate, owner_info, prepare_keys, prompt_password_hidden, remove_code,
-    rotate_keys, set_emergency_finalizer, set_keys, set_staking_limits, transfer, treasury_approve,
-    treasury_propose, treasury_reject, update_runtime, upload_code, validate, vest, vest_other,
-    vested_transfer, Command, ConnectionConfig,
+    rotate_keys, schedule_upgrade, set_emergency_finalizer, set_keys, set_staking_limits, transfer,
+    treasury_approve, treasury_propose, treasury_reject, update_runtime, upload_code, validate,
+    vest, vest_other, vested_transfer, Command, ConnectionConfig,
 };
 use log::{error, info};
 use sp_core::Pair;
@@ -182,6 +182,14 @@ fn main() {
             Ok(result) => println!("{:?}", result),
             Err(why) => error!("Contract remove code failed {:?}", why),
         },
+        Command::VersionUpgradeSchedule {
+            version,
+            session: session_for_upgrade,
+            expected_state,
+        } => match schedule_upgrade(cfg.into(), version, session_for_upgrade, expected_state) {
+            Ok(_) => {}
+            Err(why) => error!("Unable to schedule an upgrade {:?}", why),
+        },
     }
 }
 

bin/cliain/src/version_upgrade.rs

@@ -0,0 +1,20 @@
+use aleph_client::RootConnection;
+use anyhow::Error;
+use primitives::SessionIndex;
+
+use crate::commands::{ExtrinsicState, Version};
+
+pub fn schedule_upgrade(
+    connection: RootConnection,
+    version: Version,
+    session_for_upgrade: SessionIndex,
+    expected_state: ExtrinsicState,
+) -> anyhow::Result<()> {
+    aleph_client::schedule_upgrade(
+        &connection,
+        version,
+        session_for_upgrade,
+        expected_state.into(),
+    )
+    .map_err(Error::new)
+}