My managed profit miners are going "offline" now and then, this seem to happen when pools fail (seen it on nicehash and zpool so far, with bminer, gminer and z-enemy). Even that the miner is flagged as offline, it is still connected to awesome miner manager and I can just start it again. Is there a way to force the miner to not stop/go to the next pool/algo, or any ideas on how should I troubleshoot this?
On options->Profit switching I have this:
Ignore pools with no accepted shares in minutes: checked 6
Ignore minutes: 50
Ignore pools after too many failed miner restart attemps: checked
Thanks
Is the scenario that once you run into this problem, even after > 6 minutes, Awesome Miner has still not marked the pool as failed and moved on to the next pool? It just keeps running the same mining software that is unable to do its work - while it display "Interface offline"?
I usually can't remote to the computers very fast so when I remote to them they are just not doing anything, I can confirm the "Interface offline" only for Equihash 150.5+NiceHash+Gminer CUDA Equihash Miner (it happens more often, I'm not 100% sure about the others).
Edit: If I go to right click->view details (even after a few hours) I have several timestamps (as if the mining didn't stop) and when I click the timestamp after the error (i.e. 12 minutes after the error), it shows the message that some pool is being ignored and the button to reset. If I click a more recent timestamp there isn't any ignored pool message.
Thanks for the update.
The fact that you do see the "Failed" for some pool in the View Details dialog indicates that Awesome Miner did detect this pool failure. The profit switcher will then not use this pool for a while, but the profit switching decision might be made a number of minutes later, depending on how often the profit switcher is configured to run.
Right now you have an ignore time of only 50 minutes, so Awesome Miner will start using the pool a while later and this is why you no longer see the "Failed" indicator. If the pool fails again a little later, Awesome Miner will again mark it as failed and ignore it.
The idea is that Awesome Miner shouldn't be stuck on a failed pool for many hours or days, but a number of minutes of mining will always be lost before Awesome Miner will consider it to be a failure. For this reason it can be good to ignore the pool for a longer period of time after a failure.
If the specific combination for Eq150,5 + Nicehash + Gminer causes these issues frequently, it could be time to consider if this combination should be removed from the profit switcher.
Sorry I wasn't clear, the miners stop working until I manually start them, the other day I went to sleep and it failed after 1-2 hours, I received the offline email but it didn't wake me up(that's ok). Another day the same thing happened, I didn't see/check my email until noon, miners were idle for around 11 hours until I started them.
I just wanted to know if there was a way to troubleshoot this because it started happening with more than 1 algo/pool/software. It doesn't happen that often but I'd rather it kept mining by itself.
When you find the miners in this start, they are showing as "Interface offline" as you pointed out earlier? Not "Service Offline"? And when you look at the miner console window, the mining software is running but do nothing because of pool connection failure?
The Offline e-mail you received, was it from the Offline Detection rule?
Sorry for asking all these questions, I'm just trying to get as detailed understanding as possible of the scenario.
I remember the message being "Interface offline" but now I'm not so sure because I found a screenshot with the miners just saying "Stopped" in red.
I rarely use the console window, I just vnc to the miner and see there isn't any console/mining software running, so I start the miner again.
Yes, the email is from the offline detection rules.
I understand the info I'm giving is probably not good enough, I'll be sure to document it better the next time this happens.
Thanks
Hi danielocdh,
sorry for not reading through all the messages in detail (i'm easily distracted), but if your miner tend to became stopped, you could try setup a rule to start the miners automatically.
I had something that simply starts the miners with the following trigger for example:
Profile Profit(configured for miner) Greater than or equals 0.05 Dollar per day (Trigger if true for 30s)
Actions:
Start Miner
Sometimes miner programs such as BMiner, GMiner if not explicitly configured, will only retry to restart so many times
(edit: the amount of restart depends on your settings in Options -> Mining Settings -> Restart attempts on process crash) before it stopped trying in case of failure to initialize hardware or establish mining with pool. After so many tries, it'll just stop and with no auto start rules after a stop is in place, AM simply takes that stop action as final and won't start your miner anymore in automatic fashion.
Just a thought if this might be a workaround for your case.
***
PS. if this indeed helps, just keep in mind, that sometimes when the above scenario happens and AM auto restarts your miners, you could end up with multiple instances (usually 2) of mining software running (yes, even with explicit command to turn miner watchdogs off) and API port conflict will occur, when this occurs, you'll get the Green "
Interface Offline" situation and need to manually terminate the mining processes. This happens alot with BMiner, GMiner and CryptoDredge in my experience.