-
Notifications
You must be signed in to change notification settings - Fork 2.3k
Handle response failed: Broken pipe #171
Comments
ethminer --help shows : -SC, --stratum-client Stratum client version. Defaults to 1 (async client). Use 2 to use the new synchronous client. |
So ... above is with -SP 1. This happens with -SP 0 -SC 2 after about 2h:
And it is now stuck here for almost 1h. |
Same happens with -SP 2 and nicehash. After about 2h I get "Unhandled signal in divisionErrorHandler()" |
-SP 2 with -SC 2 to nicehash also deadlocks after a while ... |
Using lsof and netstat I see that ethminer process has a connection established to pool stratum server. At the moment I start seeing issues in its output, I see this connection in close_wait and then it goes away completely. But ethminer is not aware of that and is still trying to write to the same socket, which netstat shows as "can't identify protocol". Also ethminer doesn't try to reconnect or to switch to backup server, defined by -FS. |
If you guys like i can hand out a temporary Solution based on the error message, so your miner would restart - permanent fix should be done in miner or pool end of course. |
Well, thats an stratum error, not an gpu error, so its not handled with the exit param. |
Using 0.11. Mining works fine, but in the past week or so I started seeing these more and more:
"Broken pipe" pile up for some hours and at the end miner crashes with "Unhandled signal in divisionErrorHandler()".
I assume this is somehow related to pools and their ability to process requests. I'm seeing this with ethermine.org and ethpool.org.
The text was updated successfully, but these errors were encountered: