P2Pool Beta for Memorycoin

[KillerByte]

This thread has been moved to MMC-Square: https://www.mmc-square.com/index.php/topic,32.0.html

[KillerByte]

I see that the uptime of your pool decrease from 200+ to only 0.3 hrs.  Did you restart ?

I changed the script to provide more logging information... dont know why the uptime is lower now. systemd may have executed an automatic restart. Keeping my eyes peeled.

If it crashes though, I dont have it auto-restarting... so it will be obvious.

EDIT: yep, it crashed. My log was not saved though should not be long now. by the time I am back I will be ready for debugging.

[toonces]

I see that the uptime of your pool decrease from 200+ to only 0.3 hrs.  Did you restart ?

[KillerByte]

KillerByte,

Did http://162.243.241.151:8080/static/ crash ?
the workers crashed several time this past 4 hrs.

workers crashing is normal, I just had it for testing purposes, not reliability. yvg has informed me he is close to finishing YAM and then there will probably be a big switch. I think yvg may be thinking of autobalancing thing... but maby not.

The pool has not crashed on my end, but we should keep going because I am still not sure.

[toonces]

KillerByte,

Did http://162.243.241.151:8080/static/ crash ?
the workers crashed several time this past 4 hrs.

[KillerByte]

I pointed 4 workers to your p2pool.  Hopefully it will capture some useful information.

We will try for 1 day and see if it crashes. Let me know of any downtime so I know when to look.

EDIT: Please pull if you are a prospective pool operator, I updated the root list of nodes so that you can connect to the network, not super important for toonces.

Also, someone is running a P2Pool on an Amazon AWS instance, but does not have any stats to show to MMCExplorer... who is that?

Just in case anyone does not know - you have to mine on your newly created pool a share before its stats show up. After that, you are good.

[KillerByte]

the mmcexplorer.info pool is actually not intended to be mined on, it is just the starting node for the P2Pool scan. You can debug output P2Pool (which would help me if it prints out some signal of what its doing when the error happens) with the --debug option.

You should mine on 162.243.*.*, that is my mining pool (actually it is the XRam Pool). If you crash it, it will help me determine if the crashes are a problem that happens on my server also, since they are very odd.

I cannot mine with my computer at this time.

As far as I understand, mmcexplorer.info P2Pool can be used by miner to discover currently available/running P2Pool instances, isn't it? If so, what is the procedure? Is there some API/link that will get list of current instances in machine readable format (preferably json)?

Yes, http://mmcexplorer.info:8080/peer_addresses, then you just keep iterating through until you find only duplicate addresses, then you must search each p2pool for local stats and fees, if necessary.

[toonces]

I pointed 4 workers to your p2pool.  Hopefully it will capture some useful information.

[yvg1900]

the mmcexplorer.info pool is actually not intended to be mined on, it is just the starting node for the P2Pool scan. You can debug output P2Pool (which would help me if it prints out some signal of what its doing when the error happens) with the --debug option.

You should mine on 162.243.*.*, that is my mining pool (actually it is the XRam Pool). If you crash it, it will help me determine if the crashes are a problem that happens on my server also, since they are very odd.

I cannot mine with my computer at this time.

As far as I understand, mmcexplorer.info P2Pool can be used by miner to discover currently available/running P2Pool instances, isn't it? If so, what is the procedure? Is there some API/link that will get list of current instances in machine readable format (preferably json)?

[KillerByte]

the mmcexplorer.info pool is actually not intended to be mined on, it is just the starting node for the P2Pool scan. You can debug output P2Pool (which would help me if it prints out some signal of what its doing when the error happens) with the --debug option.

You should mine on 162.243.*.*, that is my mining pool (actually it is the XRam Pool). If you crash it, it will help me determine if the crashes are a problem that happens on my server also, since they are very odd.

I cannot mine with my computer at this time.

[toonces]

Do you have a debug version of the p2pool ? I can't find any coredump file. It seems it try to pipe to apport.  I'm running under Ubuntu Server 12.04.4 LTS It seems to crash every 2 hrs with 4 worker pointing at it.  Is the mmcexplorer.info pool working ?  I can just point to it and see if it crash.

Just to point out this is how I start the program under tmux session

Code: [Select]

#!/bin/bash

for i in {1..10}
do
   TIMESTAMP=$(date +%Y%m%d%H%M%S%Z)
   /usr/bin/python /home/ubuntu/p2pool/run_p2pool.py --net memorycoin --fee 1.0 --address MVTEchaqyCuu8RsP5kMYMuTbxRrwZ2jmtQ >> /home/ubuntu/$TIMESTAMP.log 2>&1
   sleep 5
done

Thanks

[KillerByte]

ok toonces, I have made no progress on finding the solution to your errors.

The first 2... there is no error, it just ends.

The last error... if you provide me with the full stacktrace, it would help me quite a bit!

And looking at the tool selection thing... definitely plan to implement that now, hopefully by the time YAM is all ready. We can do even better: Allow the user to implement his address, and then give a command which the user can copy and paste to connect to the pool.

This may take a little bit of time though, the graphs will a database which I must create and manage, and I have to make sure that MMCExplorer is as reliable as possible so that the graphs look correct.

[toonces]

KillerByte,

Look at http://p2pool.jir.dk/dogecoin/ under the tool section.
This make it easier for user to create the miner config.

Maybe this can be added to http://mmcexplorer.info/p2pool

Thanks

[KillerByte]

Great idea. I will make the changes today.

[toonces]

Hi KillerByte,

Could you also add active miners tab like in http://li625-109.members.linode.com:22550/static/
Maybe fork from blixnood instead of forrestv it seem to be more up to date.

Thanks