Testnet - Test Proof of Giving

[orbis]

...(IE coin must be 20% older than stated for it to auto-tithe).

Does it mean to us increasing diff by 20%?
Because now we have problems with coinage, not with +20%

[Rob Andrews]

Does it mean to us increasing diff by 20%?
Because now we have problems with coinage, not with +20%

What do you mean?

Edit: You had a problem with coin_age, it was too low in your illegal tithe.

Edit2:  No it will not mean increase in diff - it just means the miner would wait longer before tithing it.
If the age required is 10 days, it would wait until it had found 12 days of coin age before tithing.  Better than complaints that we lost peoples money.

A user can always tithe manually however against the current diff level.

[Rob Andrews]

Putting this in here because I thin it's related.

I setup a prod sanc yesterday, it crashed based on a gobject issue

debug log shows this:   Shouldn't this be disabled in prod currently?

Code Select Expand


NextPogRevenue 50694.078759
NextPogRevenue 50694.078759
NextPogRevenue 50694.078759
NextPogRevenue 50694.078759
NextPogRevenue 50694.078759
NextPogRevenue 50694.078759
NextPogRevenue 50694.078759


Also, watchman crashed:

Code Select Expand


Traceback (most recent call last):
  File "bin/watchman.py", line 244, in <module>
    main()
  File "bin/watchman.py", line 191, in main
    perform_biblepayd_object_sync(biblepayd)
  File "bin/watchman.py", line 24, in perform_biblepayd_object_sync
    GovernanceObject.sync(biblepayd)
  File "lib/models.py", line 78, in sync
    golist = biblepayd.rpc_command('gobject', 'list')
  File "lib/biblepayd.py", line 42, in rpc_command
    return self.rpc_connection.__getattr__(params[0])(*params[1:])
  File "/home/mn2/.biblepaycore/watchman/venv/lib/python2.7/site-packages/bitcoinrpc/authproxy.py", line 139, in __call__
    response = self._get_response()
  File "/home/mn2/.biblepaycore/watchman/venv/lib/python2.7/site-packages/bitcoinrpc/authproxy.py", line 179, in _get_response
    http_response = self.__conn.getresponse()
  File "/usr/lib/python2.7/httplib.py", line 1121, in getresponse
    response.begin()
  File "/usr/lib/python2.7/httplib.py", line 438, in begin
    version, status, reason = self._read_status()
  File "/usr/lib/python2.7/httplib.py", line 402, in _read_status
    raise BadStatusLine(line)
httplib.BadStatusLine: ''



This is 1.1.7.5 from the mainnet PPA

edit:

ran again and got this in the debug.log

Code Select Expand


NextPogRevenue 50694.078759 CGovernanceObject::IsCollateralValid -- Couldn't find opReturn 045038724b047ba1908bc557777a5c5ae04703230449d1707865b6$
    CTxIn(COutPoint(188a03b6449ca0bc55e6080dffe79eb3ce82ab29a8fa058c25cd12eb56ba13a2, 1), scriptSig=4730440220609c612a0d1bc0, nSequence=429496729$
    CTxOut(nValue=5.00000000, scriptPubKey=6a20be8b2d169ab6657870d1490423)
    CTxOut(nValue=350.70471391, scriptPubKey=76a914c9a8aec4fae6853343f451e6)

2019-01-03 18:44:47 MNGOVERNANCEOBJECT -- Governance object is invalid - Couldn't find opReturn 045038724b047ba1908bc557777a5c5ae04703230449d1707$
    CTxIn(COutPoint(188a03b6449ca0bc55e6080dffe79eb3ce82ab29a8fa058c25cd12eb56ba13a2, 1), scriptSig=4730440220609c612a0d1bc0, nSequence=429496729$
    CTxOut(nValue=5.00000000, scriptPubKey=6a20be8b2d169ab6657870d1490423)
    CTxOut(nValue=350.70471391, scriptPubKey=76a914c9a8aec4fae6853343f451e6)


On the POGRevenue, I deliberately added that to print it to the log and not to the screen - just so I could see the value in advance for sanity sake.  I just took out the logging for prod for the next version.

[Rob Andrews]

On the POGRevenue, I deliberately added that to print it to the log and not to the screen - just so I could see the value in advance for sanity sake.  I just took out the logging for prod for the next version.

On watchman, could you try running it again and see if it recovered?  There are a lot of Error conditions in watchman that are picked up on the next run (it runs every 5 mins).

On the Collateral being invalid in prod, that is normal.  Thats due to someone creating a proposal without aged collateral.  That just wont be a valid governance object, so that is OK.

[sunk818]

1.1.7.7 - Leisure Upgrade
Mandatory Upgrade for TestNet


- MIP added WIN exit handler (in getboincmetrics during initialization) - Test to see if this stops crashing in Win10 Home (be sure to set boincmetrics=0 first)
- Moved exec tithe and exec titheinfo to : tithe and titheinfo
- Added a spork to allow us to add or remove POG training wheels (allows devs to attempt to verify forks)
- Added nickname to 'titheinfo' (Allows user to see config is being recognized)
- Modified POG to only look back 16 blocks (prevent forks)
- POG 24 hour tithes is now based on the last 16 blocks extrapolated out to 24 hours

Having hard time finding peers again. No problems on 1175. 1177 only one wallet found peers. Rest at 0 peers for several hours.

[Rob Andrews]

Having hard time finding peers again. No problems on 1175. 1177 only one wallet found peers. Rest at 0 peers for several hours.

Please wait about an hour, 1178 is almost out.

[Rob Andrews]

I shut down all my clients a while ago based on Rob's message earlier.

Sorry for the delay; Its taking a little longer than usual; moving to daily superblocks required a larger than expected change.

Debugging now.

[Rob Andrews]

1.1.7.8
Mandatory Upgrade for TestNet

- Transitioned to Daily POG superblocks
- Ignore POG bankroll denominations in PODC Update (will someone please manually test some exec podcupdates to ensure it will not spend bankroll denominations)
- Enhanced exec pogaudit and added details parameter to 'pogpool' rpc commands


** After upgrading please start with -reindex -zapwallettxes=1 **


PS:  Someone explained reindexing was very slow - FIXED

[Rob Andrews]

Let us also please test a boinc PODC superblock.

[sunk818]

Let us also please test a boinc PODC superblock.

Does this mean masternodes should be fired up on testnet?

[orbis]

Rob is it possible to do something with coin age?
After upgrade to 1178 I've made reindex and zapwallettx, after some time my wallet spontaneously restart and start to reindex again.
Now I'm not able to tithe because my coin age is 0.
Here is example:

Code Select Expand

exec getdimensionalbalance 0 1

{
  "Command": "getdimensionalbalance",
  "Amount 84265.2600": "Age 0.02",
  "Amount 81562.3800": "Age 0.49",
  "Tithe_Balance_Available": 165827
}


But when I look inside my wallet inputs there are many coins with many confirmations. Look at picture.
It will be possible to count coin age from numbers of confirmations?

[Rob Andrews]

Rob is it possible to do something with coin age?
After upgrade to 1178 I've made reindex and zapwallettx, after some time my wallet spontaneously restart and start to reindex again.
Now I'm not able to tithe because my coin age is 0.
Here is example:

Code Select Expand

exec getdimensionalbalance 0 1

{
  "Command": "getdimensionalbalance",
  "Amount 84265.2600": "Age 0.02",
  "Amount 81562.3800": "Age 0.49",
  "Tithe_Balance_Available": 165827
}


But when I look inside my wallet inputs there are many coins with many confirmations. Look at picture.
It will be possible to count coin age from numbers of confirmations?

Yeah, good idea, I might have to do something so we dont have to test forever.

But lets see if anyone online can help us regardless, to get everyone synced.

First, on the spontaneous reboot, I was a little worried about that myself, but I see the condition in which it happens.  The wallet has a 'feature' in it where if it finds a chain with higher work and it thinks chain corruption is possible, then it erases the chain and reboots.  This is a long discussion, but lets just say for now that the chain corruption flag is due to the fact that we have a bad block in the longer chain (it doesnt have a conforming daily superblock),  so we get fooled into resyncing and then we cant go any further than we were to begin with.  So I feel a lot better about this particular point.

However, we have someone on 76.170.58.128 that brought two nodes on that never reindexed  - I think this is the root of that problem.

Ill need to add a wallet rule to force an upgrade and ignore that chain.  Ill also need to give us more granular control on that restart "feature" like being able to disable it etc.

I can also look at the coin - age thing, but still we need to be synced first.

Is 58.128 out there?  Can you reindex your chain please?

[Rob Andrews]

In the mean time, Ill work on Adding a method to count 'some' low tithes in testnet, and adding a block rule to get past our block, and disabling the reboot and adding a key for the reboot.

EDIT:  Please hold off everyone, 1179 will be out after church, then we can just reindex.

[orbis]

Yes, I have 76.170.58.128 in my nodes...
I can ban him for one day...
Now I'll wait for 1179 and then I'll try it again.

[sunk818]

I erased everything except wallet.dat - no peers. It was fine on 1175 yesterday.