v0.10.16.x Testing

[qwizzie]

Evan : i sent you the debug.log + wallet of the misbehaving client, it just received a third 0.1 fee
on a roll.

 

[AjM]

I think it's a side effect (false message) from me setting up 20 masternodes with one transaction.
The new system only allows masternodes to send free transactions once in (some time period), so the new code would disallow using the masternode your wallet selected (even though it is a different masternode, the txn makes it look like the same node.)

I've shut down 19 of them. Once the cache's clear, we'll see if things spring back to life.

Ok, now after 15 min trying, its happening, DS transaction succesfull...

 

[moocowmoo]

Finally my 'hung' wallet got its first mix (90 minutes later) -- am guessing shutting down the 19 'rogue' nodes helped things along.

Evan, will each masternode having a distinct txn hash be a future requirement?
Or is this just a bug to squash when you wake up?

 

[eduffield]

****** PLEASE UPDATE TO v10.16.5 *******

Let me know if things start hanging again, if this doesn't fix it i'll look into it further.

- Removed the rest of the spam ("Masternode is publishing too many free transactions.. ", "vout - X", )
- Added queue matching algorithm, Darksend will now try to figure out how to match with active queues (by using that denom). Should be much faster.
- Fixed DSQ throlling issues
- Fixed "Last darksend was too recent" issue (can still happen right after you start the wallet though)

----

CI-builds for v0.10.16.5

Source:
https://github.com/darkcoin/darkcoin/tree/ds-zero-fee/src

Windows 32bit:
http://bamboo.darkcoin.qa/artifact/DRKW-DEV/JOB1/build-34/gitian-win-darkcoin-bin/32/darkcoin-qt.exe
http://bamboo.darkcoin.qa/artifact/...tian-win-darkcoin-bin/32/daemon/darkcoind.exe
http://bamboo.darkcoin.qa/artifact/...oin-bin/32/darkcoin-0.10.16.5-win32-setup.exe

Mac OS X:
http://bamboo.darkcoin.qa/artifact/...-osx-darkcoin-dist/darkcoin-0.10.16.5-osx.dmg

Linux 32bit:
http://bamboo.darkcoin.qa/artifact/.../gitian-linux-darkcoin-bin/bin/32/darkcoin-qt
http://bamboo.darkcoin.qa/artifact/DRKL-DEV/JOB1/build-35/gitian-linux-darkcoin-bin/bin/32/darkcoind

Linux 64bit:
http://bamboo.darkcoin.qa/artifact/.../gitian-linux-darkcoin-bin/bin/64/darkcoin-qt
http://bamboo.darkcoin.qa/artifact/DRKL-DEV/JOB1/build-35/gitian-linux-darkcoin-bin/bin/64/darkcoind

 

[qwizzie]

****** PLEASE UPDATE TO v10.16.4 ******* should be ****** PLEASE UPDATE TO v10.16.5 ******* but we get it

I will update the client

Edit 1 : after yr done compiling the windows version.. because it doesnt seem to download right now
Edit 2 : damn, my antivirus actually stopped my download .. had to disable it to download this new version. all good now.

 

[AjM]

WTF!

V16.4, stopped mixing and trying to send tDRK:s back to faucet,
balance 663, sent amount 660 -> error not enough funds, see pic.

 

[qwizzie]

try ''no preference''

 

[AjM]

try ''no preference''

Too late, resetted for new ver. v16.5.

And back to sync stuck again.
Deleted wallet.dat and peers.dat, start -> 0 hours behind -> stuck -> restart -> sync OK.

 

[eduffield]

WTF!

V16.4, stopped mixing and trying to send tDRK:s back to faucet,
balance 663, sent amount 660 -> error not enough funds, see pic.

I've seen that error once too. It's really rare, haven't had it since.

 

[eduffield]

Keep in mind, random collateral fee's are OK now. The system charges them instead of the subscription. However, they should be pretty rare, so 3 errors in a row is definitely not correct (unless you're really unlucky).

Beyond that, does everything look like it's working now? I don't see anything else to fix (Anything besides Darksend will be tacked next release).

 

[AjM]

I've seen that error once too. It's really rare, haven't had it since.

Ok, i am trying replicate this.

BTW: Win7 x64.

 

[AjM]

Keep in mind, random collateral fee's are OK now. The system charges them instead of the subscription. However, they should be pretty rare, so 3 errors in a row is definitely not correct (unless you're really unlucky).

Beyond that, does everything look like it's working now? I don't see anything else to fix (Anything besides Darksend will be tacked next release).

Any comment about that sync stuck problem?
Do you need log for this?

 

[eduffield]

Any comment about that sync stuck problem?
Do you need log for this?

It's just something with testnet, I would ignore it.

 

[AjM]

It's just something with testnet, I would ignore it.

Its also in main net, look this thread: https://darkcointalk.org/threads/wallet-not-synchronizing.2780/

 

[oblox]

I'm still not getting in denominations past round 1 on v5.

 

[qwizzie]

to be honest, this latest update is looking good.. all my clients are happily denominating so far
Its just that one client from an older version that had problems which i sent your way.

edit : we could use a bit more mining power on Testnet to clear the blocks faster.

 

[UdjinM6]

Wallet does not respect locked unspent while trying to denominate (I locked 1000 for tMN using Coin Control) and so it will never start to denominate

2014-11-13 20:25:47 DoAutomaticDenominating: Split up large input (justCollateral 0):
2014-11-13 20:25:47  -- nTotalBalance 672600941118
2014-11-13 20:25:47  -- denom 672600941118
2014-11-13 20:25:47 keypool reserve 890
2014-11-13 20:25:47 SplitUpMoney: Error - Unable to locate enough Darksend non-denominated funds for this transaction
2014-11-13 20:25:47 keypool return 890
2014-11-13 20:25:47 Darksend is idle
2014-11-13 20:26:00 received block 000000006a4d71829c3dfb822fd289dd16e81b03cf466cd85c25bc5b8c98e73a peer=4
2014-11-13 20:26:00 Committing 546 changed transactions to coin database...
2014-11-13 20:26:00 SetBestChain: new best=000000006a4d71829c3dfb822fd289dd16e81b03cf466cd85c25bc5b8c98e73a  height=63377  log2_work=45.688534  tx=145555  date=2014-11-13 20:25:57 progress=1.000000
2014-11-13 20:26:00 mnw - winning vote  CTxIn(COutPoint(de664d6b8dee103e8ef4b93df62288a2c836000907a659f568bf2ed0bb7bf217, 13), scriptSig=) Height 63387 bestHeight 63377
2014-11-13 20:26:00 send last getblocks for 000000006a4d71829c3dfb822fd289dd16e81b03cf466cd85c25bc5b8c98e73a peer=2
2014-11-13 20:26:00 ProcessSyncCheckpoint: sync-checkpoint at 0000000010910d440267c9c6cc84c91a932ff047ef940e4bb55da6a32f99ca7a
2014-11-13 20:26:01 send last getblocks for 000000006a4d71829c3dfb822fd289dd16e81b03cf466cd85c25bc5b8c98e73a peer=3
2014-11-13 20:26:01 send last getblocks for 000000006a4d71829c3dfb822fd289dd16e81b03cf466cd85c25bc5b8c98e73a peer=7
2014-11-13 20:26:01 getblocks -1 to 0 limit 500 peer=3
2014-11-13 20:26:01 getblocks -1 to 0 limit 500 peer=7
2014-11-13 20:26:13 CDarkSendPool::UpdateState() == 3 | 3
2014-11-13 20:26:24 SplitUpMoney - Too soon to split up again

 

[bituzer]

yes I also had this sync stuck error repeatedly since few weeks, with newest windows binarys, on main net

edit: i think its windows only problem

 

[oblox]

Ok, i am trying replicate this.

BTW: Win7 x64.

I got the same error (Unable to locate enough darksend) from using normal transaction on the same platform on v5...

 

[AjM]

yes I also had this sync stuck error repeatedly since few weeks, with newest windows binarys, on main net

edit: i think its windows only problem

Same here, and its happening in testnet too.