Backup pools: configure and test failover
Keep a primary connection and a verified fallback without sending work to the wrong account.
ASIC.tools · Reviewed
Define what the backup should protect
Decide whether you need another endpoint of the same pool or a different pool operator. Another regional endpoint can help with a route problem, but does not remove every dependency on the original service. A second operator requires its own account and payout setup. Record the objective before adding addresses, so a backup field does not create a false sense of resilience.
Prepare complete connection details
Use each operator’s current documentation for the hostname, transport, port and worker format. A hostname copied from an old forum can be obsolete. Configure and verify the destination account before using it as a fallback. Keep the worker identity traceable to the same physical device, even if account prefixes differ. Never use example account names from a tutorial on production equipment.
Check the firmware switching rules
Read how the installed firmware interprets multiple pool entries. Depending on the product and configuration, entries may represent priority failover, groups, quotas or other work distribution. Do not assume three filled fields mean equal distribution or automatic redundancy. Note retry behavior, recovery behavior and any developer fee. Keep a copy of the working configuration before changing one test miner.
Run a controlled test
Schedule the test on a single noncritical device. Temporarily make only its primary route unavailable using a reversible, authorized network rule. Observe whether the miner submits accepted work to the intended backup and record the interruption. Restore the route and confirm the documented recovery behavior. Do not disable the whole farm connection simply to prove that one miner can change endpoints.
Review the financial trail
After the test, inspect both operator dashboards and reconcile the time spent on each. Small balances on a backup pool may remain below a payout threshold; include this in the operating plan. Recheck fallback credentials after changing accounts or wallets. A successful test is a recorded switch with accepted work and a verified destination, not just an interface showing that a backup address exists.
