Reading Rejected Share Logs on Long-Running Rigs
How to distinguish normal pool variance from hardware degradation when your ASIC fleet has been running 24/7 for over a year at 0.05 PH/s or higher.
Rejected shares are a normal part of pool mining. A rate below 1% on most pools is unremarkable. But when rejections climb above 2% on hardware that previously held steady, the cause is worth investigating before hash boards degrade further.
Network vs Hardware Rejects
Pool-side rejects often spike during ISP routing changes or when your facility switches upstream providers. These tend to affect all workers simultaneously and resolve within hours.
Hardware-related rejects show up on individual workers first. One Antminer in a rack of forty starts climbing while neighbours stay flat. That pattern points to a failing hash board, loose ribbon cable, or overheating on a single unit.
What to Log Weekly
We ask retainer clients to export weekly CSV reports from their pool dashboard:
- Reject rate per worker
- Stale share percentage
- Average hash rate per worker (24-hour rolling)
Comparing week-over-week on the same worker ID reveals trends that daily glances miss.
When to Pull a Unit
If a single worker shows reject rate above 3% for three consecutive days while hash rate drops more than 5% from its baseline, schedule a bench inspection. Continuing to run a degraded board increases the risk of damaging adjacent components on the same hash board assembly.
Firmware Considerations
Some firmware versions introduced timing bugs that inflated reject rates on specific pool protocols. Before replacing hardware, verify your firmware version against the manufacturer’s release notes. We maintain a private log of known issues by model and firmware build — contact us if you want us to check your fleet’s versions during an assessment.