I’m running a Public Pool service on Start9 (via StartOS), mining on Bitcoin mainnet. Recently I’ve been observing recurring events in the logs that I’d like help understanding and troubleshooting.
Log excerpts:
Deleted 6 old statistics
Deleted 0 old clients
Killing dead clients
getblocktemplate tx count: 2162
getblocktemplate tx count: 2354
ZMQ Unable to connect, Retrying (multiple times)
ZMQ Connected
getblocktemplate tx count: 2404 …
Killing dead clients
New Block → getblocktemplate tx count drop …
Issues I’m observing:
Frequent “Killing dead clients” entries – what’s the expected cause and threshold for clients flagged as “dead”?
ZMQ connection loops – logs show “Unable to connect, Retrying” repeatedly before eventually reconnecting. I suspect configuration mismatches or port issues.
getblocktemplate timeouts (ESOCKETTIMEDOUT) in past logs (though I’ve since increased rpcservertimeoutand tuned rpcthreads/rpcworkqueue).
What I’ve done already:
Set rpcservertimeout=300
Configured rpcthreads=64, rpcworkqueue=256
Public Pool .env has BITCOIN_RPC_TIMEOUT=20000
What I’d love feedback on:
Could the repeated client kills indicate worker instability, Share timeouts, or misconfigured Keep‑alive? How long before a client is dropped?
For ZMQ: common pitfalls in bitcoin.conf (e.g. correct ports for zmqpubhashblock, zmqpubhashtx) or Docker/BIND settings?
Any additional bitcoind or pool config tweaks I’ve overlooked?
If anyone has solved similar log patterns or recommends specific config tuning, I’d deeply appreciate the guidance. I can also share snippets of my bitcoin.conf and .env (without sensitive credentials) if that’s helpful.
the “killing dead clients” is a normal message from public-pool, which periodically checks for any miners that are not connected anymore.
rpc timeouts and zmq connection issues usually indicate that bitcoin cannot handle the load. What are your hardware specs and what other services have you running on the same server (which may or may not use bitcoin, but can effect the performance of your server)?
Running only: Bitcoin Core and Public Pool; no other services or applications on this server
Solo mining setup: using the Bitaxe Gamma 601 ASIC miner with a single BM1370 chip, providing approximately 1.2 TH/s hash power while consuming around 15–17 W.
That should be ok, I run a vm with similar hardware, on pretty much default bitcoin.conf settings. Do the connection errors happen often? It could happen once in a while, but should be rare.
the “killing dead clients” message is just a notification in the code of public-pool when it checks for unresponsive clients, even if there aren’t any.
Log excerpts:
Issues I’m observing:
What I’ve done already:
What I’d love feedback on:
