Replymessage unavailable
Yes — with 5 kaswallets you're running 5 workers (frontend-w5), and all 5 are meant to be funded and to share the load. Traefik round-robins requests across the workers, and each worker pays L1 KAS from its own kaswallet. Wallet 0 doing ~100% is NOT normal for a 5-worker node — it means your attestor is submitting to one worker directly (the old default http://rpc-provider-0:8535) instead of the Traefik endpoint (https://<your-domain>:8545), which bypasses the load balancer.
IMPORTANT — do this BEFORE fixing the routing:
Check W1..W4_WALLET_TO_ADDRESS in your .env. That's where each wallet's transaction CHANGE is returned, and the env template ships the same placeholder address for all five workers — if it's still a placeholder, the wallet loses its funds on its FIRST transaction. Wallet 0 surviving 3,000+ txs proves W0 is set correctly.
Then, in order:
- Check W1_KASWALLET_PASSWORD matches keys.kaswallet-1.json — "unsigned transactions" usually means the daemon can't sign (wrong password) or the wallet isn't synced; check kaswallet-1 logs and scripts/debug/wallet-status.sh.
- Update the attestor to the latest version and set its RPC_URL to https://<your-domain>:8545 (NOT rpc-provider-0:8535), then restart. Attestation txs will round-robin across all 5 wallets.
Alternatively, if you only want to attest and not run a public 5-worker RPC, just run frontend-w1 and fund only Wallet 0. Both are valid — the thing to avoid is 5 workers with the attestor pinned to one, or unpinning it while wallets 1–4 still have placeholder change addresses.