Alright after going over my logs in extreme detail, this is my findings.
Poolinfo went on forks twice yesterday and I was able to find out a little why
I got this block (Block on fork)
received: block (265 bytes)
received block 000000087c4167eb542e
CBlock(hash=000000087c4167eb542e07968c44e62e068b805c26bde8102fba690d183daa7a, ver=5, hashPrevBlock=d1a892e6d6ed5233589e474950f8bbc34961e6887d573663e7059dd84182b132, hashMerkleRoot=90ce79972cbda34750d9d7094a3d6505974620f1efc398b4d92f490783c79297, nTime=1503401831, nBits=1d0882e7, nNonce=2863360803, vtx=1, vchBlockSig=304502206c818406dcaf76e4c5ddf1766cacb05918c64b86328382bbf8c95b0fdd549798022100bc0a7588471070444c1a23cca1713e69b0bc3d903920803e6a6b2e7653ae4512)
Coinbase(hash=90ce79972c, nTime=1503401715, ver=1, vin.size=1, vout.size=1, nLockTime=0)
CTxIn(COutPoint(0000000000, 4294967295), coinbase 037921160174062f503253482f)
CTxOut(nValue=13.33016327, scriptPubKey=022bd540c7e6ba8a7522ce5cdf6e3fcafc85d79c42b2cd92a80647fde524825593 OP_CHECKSIG)
vMerkleTree:
ComputeNextStakeModifier: prev modifier=0xa377d57fadc587cd time=2017-08-22 11:31:35 UTC
@@PoWII-V2 (nHeight, rDiff, rSubsidy) = (1450356, -0.231851, 13.330163)
trying connection 216.189.144.211 lastseen=0.0hrs
connected 216.189.144.211
send version message: version 71061, blocks=1450360, us=213.32.21.37:8233, them=216.189.144.211:8233, peer=216.189.144.211:8233
sending: version (107 bytes)
socket closed
disconnecting node 216.189.144.211
SetBestChain: new best=000000087c4167eb542e07968c44e62e068b805c26bde8102fba690d183daa7a height=1450361 money supply=7955827 trust=66007579680572 date=08/22/17 11:37:11
and then a couple of seconds got this (Block on the proper chain)
received block 0000000246981327761b
CBlock(hash=0000000246981327761bf9bbff24c8420f4998e1ae3a4b2b24f24ca4b8f60cec, ver=5, hashPrevBlock=d1a892e6d6ed5233589e474950f8bbc34961e6887d573663e7059dd84182b132, hashMerkleRoot=efc38da3cdbac1cda1dbe6d29acbed1315d62e1190a5ccb7b83b1dd923d37fe3, nTime=1503401828, nBits=1d0882e7, nNonce=1431662122, vtx=1, vchBlockSig=304402206dbc0c70aadb8aadef6a5ae3028a069244b59988867914e720a817140d0e7eaf02204a64a8ac08fc68ab955cbf71a58ebc00ce20fb41c13a3f33f42e6a13b21d74ac)
Coinbase(hash=efc38da3cd, nTime=1503401717, ver=1, vin.size=1, vout.size=1, nLockTime=0)
CTxIn(COutPoint(0000000000, 4294967295), coinbase 0379211602fa00062f503253482f)
CTxOut(nValue=13.33016327, scriptPubKey=02b6a8b835d671c026250c024ba5d145ded4599ccde38adecdf769a70e7464066a OP_CHECKSIG)
vMerkleTree:
Misbehaving: 104.128.225.240:8233 (0 -> 100) DISCONNECTING
disconnecting node 104.128.225.240
ERROR: ProcessBlock() : block with too little proof-of-work
At this point I was looking at how many times i got peers disconnected due to ERROR: ProcessBlock() : block with too little proof-of-work.
The proper chain told poolinfo that this is the block 10 times and actually disconnected and banned all 10 peers telling poolinfo about the proper block.
So to test my theory out, I increased banscore in magi.conf to 500 with
The debug log dropped from ~620MB to ~62MB in size (about the same running time +- a few hrs) + was able to recover from when poolinfo got invalid blocks.
I've seen in poolinfo logs block that was accepted but later on ORPHAN because the right chain was presented to poolinfo and thus got back on the right track.
So if you like to test this too, add this to
magi.conf
You might need to resync the chain if you are on a fork.
To test this properly, we are going to need to mine it. Anyone willing to try this out?
