Skip to content
Small team, full backlog, zero orders dropped. Support replies are slower than we’d like. Read our status update → Zero orders dropped. Status → 📬 Check your spam folder — most of our replies land there. We do answer. Status update → 📬 Check your spam folder. Status →
1325 Rue Bergar, Laval, QC support@d-central.tech 1-855-753-9997 Mon-Fri: 8:30 - 18:00

Pool Disconnects or Shares Rejected — Troubleshooting

Quick answer

If the hardware is healthy but you're losing shares, the problem is between the miner and the pool: credentials, latency, or an unstable tune.

Sort credentials and a stable backup pool first; a small reject % is normal.

Full diagnostic flow

Does the pool show the worker connecting at all?
  • Worker never authorises → Authorisation failure — wrong worker name, password, or wallet. Fix the worker credentials in the miner config. [Fix it yourself] Error-code guide →
  • Connects then drops → Add a reliable backup pool and watch. Does it still drop on both?
Add a reliable backup pool and watch. Does it still drop on both?
  • Drops on every pool → Network-side: DNS, NAT/firewall, or an unstable connection. Try the pool IP directly and a solid DNS (1.1.1.1). [Fix it yourself] Error-code guide →
  • Stable now, but high rejects → Are rejected/stale shares more than a few percent?
Are rejected/stale shares more than a few percent?
  • High reject rate → Latency or an aggressive overclock producing stale/HW-error shares. Use a closer pool and back the tune off to stock to confirm. [DIY with care] Error-code guide →
  • Only a small reject % → A low single-digit reject rate is normal and not worth chasing. You're fine. [Fix it yourself]

This points at board-level damage. You can attempt it, but the realistic call for most owners is to send it to a repair bench.

Fix it yourself

Work the matching symptom decision tree and error-code guide step by step. Most cable, cooling, network and firmware faults are solvable on the bench you already own.

Send it to D-Central

Cracked solder, dead domains, shorted chips and EEPROM faults need a real repair bench. See transparent repair pricing or get a free repair quote — we diagnose and repair 38+ ASIC models in Montreal, board-level, since 2016.

More tools: Kernel-Log Decoder · ASIC Fault Finder (650 codes) · All symptom trees

Find the last successful connection layer

“Not connecting” can describe several different failures. Test from the network upward and stop at the first layer that does not produce its expected signal.

LayerTestSuccess signalIf it fails
Name resolutionResolve the configured host from the miner’s networkOne or more current IP addressesCorrect the hostname or DNS path; do not permanently substitute an unverified IP
TransportConnect to the documented host and portTCP opens without an immediate resetCheck the port, firewall, egress policy, and whether the endpoint expects TLS
ProtocolObserve the first mining exchangeA valid response to the connection/setup requestConfirm the protocol version and TLS mode; an open web port is not a mining endpoint
AuthorizationInspect miner and pool logsThe worker or payout identity is acceptedRe-enter the documented user/address format and remove hidden spaces or stale credentials
Work deliveryWatch for a current jobNew work references the current chain tipEscalate template or node health; reconnect loops cannot repair missing current work
Share submissionCompare accepted/rejected counters and exact reasonsValid shares are acknowledged for the current jobClassify stale, duplicate, low-difficulty, unauthorized, and malformed submissions separately

An open port is not mining readiness

A successful TCP connection proves only that something accepted a socket. It does not prove that the service recognized Stratum, authorized the worker, delivered current work, or accepted a share. Miner and server logs should identify the last completed exchange.

On Windows, Resolve-DnsName example-pool.invalid and Test-NetConnection example-pool.invalid -Port 3333 cover the first two layers. Replace the placeholder only with a pool’s current documentation.

Classify the rejection before changing settings

  • Stale/job not found: old work, delayed jobs, reconnects, or timing instability.
  • Duplicate: a repeated solution, retry, or duplicated worker process.
  • Low difficulty: the result missed the assigned share target.
  • Unauthorized: the connection exists, but the worker identity was refused.
  • Malformed/unknown method: protocol, framing, version, or extension mismatch.

Privacy-safe support packet

  1. UTC start and failure time, including timezone.
  2. Miner model, firmware version, and recent update history.
  3. Pool host, port, protocol, and TLS setting—without passwords or private URLs.
  4. The exact error and preceding protocol step.
  5. Time of the last new job and accepted share, if available.
  6. Whether a documented fallback works from the same network.
  7. A short redacted log window that preserves event order.

Use the Stratum protocol reference for message flow, the mining-share explainer for counters, and the Bitaxe setup guide for AxeOS fields. Never post a payout address, Wi-Fi password, or authenticated URL in a public support thread.

Related products, repair, and setup paths

Last reviewed June 12, 2026.