Fix zombienet-cumulus-0007-full_node_warp_sync

[DenzelPenzel]

Is there an existing issue?

• I have searched the existing issues

Experiencing problems? Have you tried our Stack Exchange first?

• This is not a support question.

Description of bug

Summary of failure:

• Parachain full node four never syncs and panics because its relay-chain RPC endpoint is unreachable. It repeatedly tries ws://127.0.0.1:36445 (dave) then stops:

Retrying to connect to any external relaychain node failed… No valid RPC url found. Essential task relay-chain-rpc-worker failed. … panicked … WorkerCommunicationError(“RPC worker channel closed…”)” @cumulus/zombienet/zombienet-sdk/tests/zombie_ci/tmp/four.log#5-18

• The relay validator dave (the only RPC target for four) warp-syncs, fails state sync, then panics inside chain sync: Failed to import target block with state: Cancelled. State sync failed. Falling back to full sync.” followed by assertion failure in chain_sync.rs:588, causing the node to crash @cumulus/zombienet/zombienet-sdk/tests/zombie_ci/tmp/dave.log#57-198.

• Other nodes (two/three) sync normally; the test fails on the wait_metric_with_timeout for four because its relay RPC source died

Last ~90 runs of this test failed 3 times:

• https://github.com/paritytech/polkadot-sdk/actions/runs/20025760102/job/57423826266
• https://github.com/paritytech/polkadot-sdk/actions/runs/20030524808/job/57440164245
• https://github.com/paritytech/polkadot-sdk/actions/runs/20034962328/job/57462717181

Steps to reproduce

trigger-zombienet-cumulus / zombienet-cumulus-0007-full_node_warp_sync

.github/scripts/dispatch-zombienet-workflow.sh \ -w zombienet_cumulus.yml \ -b "master" \ -p "zombienet-cumulus-0007-full_node_warp_sync"

cc @pepoviola

[bkchr]

There is actually a second issue:

2025-12-08 11:20:14.542 ERROR tokio-runtime-worker sync: [Relaychain] Failed to import target block with state: Cancelled.    
2025-12-08 11:20:14.543 ERROR tokio-runtime-worker sync: [Relaychain] State sync failed. Falling back to full sync.    
2025-12-08 11:20:14.543 DEBUG tokio-runtime-worker sync: [Relaychain] Restarting sync with client info Info { best_hash: 0x5b8c7f8a7a09479fcd48f70df5032a7f9045ab57d5ecff6ae63e87467598b0dc, best_number: 0, genesis_hash: 0x5b8c7f8a7a09479fcd48f70df5032a7f9045ab57d5ecff6ae63e87467598b0dc, finalized_hash: 0x5b8c7f8a7a09479fcd48f70df5032a7f9045ab57d5ecff6ae63e87467598b0dc, finalized_number: 0, finalized_state: None, number_leaves: 1, block_gap: None }    
2025-12-08 11:20:14.543 DEBUG tokio-runtime-worker sync: [Relaychain] Restarting from genesis    
2025-12-08 11:20:14.543 DEBUG tokio-runtime-worker sync: [Relaychain] New peer 12D3KooWRkZhiRhsqmrQ28rt73K7V3aCBpqKrLGSXmZ99PTcTZby with best hash 0x90fd…2a1b (1170).    
2025-12-08 11:20:14.543 DEBUG tokio-runtime-worker sync: [Relaychain] New peer 12D3KooWJpC9Ku9RNAcDKRHCtndiFQaunu37DcHziag9DeKpBP2N with best hash 0x223a…7795 (1167).    
2025-12-08 11:20:14.543 DEBUG tokio-runtime-worker sync: [Relaychain] New peer 12D3KooWQCkBm1BYtkHpocxCwMgR8yjitEeHGx8spzcDLGt2gkBm with best hash 0x90fd…2a1b (1170).    
2025-12-08 11:20:14.543 DEBUG tokio-runtime-worker sync: [Relaychain] New peer 12D3KooWPKzmmE2uYgF3z13xjpbFTp63g9dZFag8pG6MgnpSLF4S with best hash 0x90fd…2a1b (1170).    

thread 'tokio-runtime-worker' panicked at /__w/polkadot-sdk/polkadot-sdk/substrate/client/network/sync/src/strategy/chain_sync.rs:588:4:
assertion failed: false
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

[bkchr]

Or maybe not "second". Looks like the same as Dave, but happened for two.

[bkchr]

We need more logging to pinpoint the issue: #10586

@DenzelPenzel after this pr is merged, please continue observing the runs and report back when one failed again.

[pepoviola]

There are also this WARNs in the nodes

(dave)

2025-12-08 17:56:14.922  WARN tokio-runtime-worker sync: Unexpected generic response strategy key StrategyKey("State"), protocol /5b8c7f8a7a09479fcd48f70df5032a7f9045ab57d5ecff6ae63e87467598b0dc/state/2    

(two)

2025-12-08 11:20:14.618  WARN tokio-runtime-worker sync: [Relaychain] Unexpected generic response strategy key StrategyKey("State"), protocol /5b8c7f8a7a09479fcd48f70df5032a7f9045ab57d5ecff6ae63e87467598b0dc/state/2 

But ones the pr is merged we will run a batch of this test and add more logs here 👍

[bkchr]

Yeah all these issues are kind of related. Maybe there is some kind of race condition.