[TEST RELEASE] Cryptonite binary for linux (NEW: Qt and Windows builds)

[mrvegad]

Ok so I started with a clean db,wallet let it sink up then started to mine, within minutes of mining my blockcount started going down, i wasn't sending out txs and i didn't time my computer time.


Edit: I restarted and let it run without mining, worked. Started up mining and let it run for a bit and had no problems. Not sure what to make of it.

[CoinHoarder]

Congrats on the test release! I am glad you finally found some developers and made this a reality! Props to you putting up the bounty.

Block chain bloat could prove troublesome to Cryptocoins, and even more so for coins that have quicker block speeds.. those blocks will add up over time. Also more and more features are being built on top of coins, and in new coins as well, which will bloat them even more. Due to the network effect transaction rates will go up in a parabolic arc, so I'm not sure everyone understands how big of a problem block chain bloat could be in the future if Cryptocoins are widely adopted.

[bitfreak!]

Edit: I restarted and let it run without mining, worked. Started up mining and let it run for a bit and had no problems. Not sure what to make of it.

Well I have no idea what is causing the issue because no one else seems to be having the same problem. We have to recode a lot of the syncing code anyway when we implement blockchain trimming, if the problem is still persisting after we do that then we'll pay more attention to it, but for now we have to focus on other things.

[catia]

Big news here. Latest release implements block pruning according to MBC scheme. So when you sync the new client it will be retrieving all of the block headers but only the contents of last 1k blocks. This version prints a lot of spam to the console. The driving force behind leaving the spam on for now is that the sync process is kind of involved and stuff like getinfo doesn't make any sense. It's really hard to tell if anything is going on until suddenly it is online. Getinfo also won't respond for a long period of time while the code is checking the downloaded data for consistency. So the spam kind of lets you know it is still working until we come up with improved UI. You can tell it's done syncing when the last line says "Ready for committal"

Restart after killing the client is also much slower now because of increased consistency checking. Again the spam will let you know what is going on.

As usual, expect lots of bugs. Performance enhancements are coming soon.

Mrvegad I am not confident in a fix for disappearing blocks yet. This could get especially interesting since all the blocks don't even exist anymore. I'm looking into it though. Complete rewrite of some temperamental code is on the short list.

Billotronic, yes you have to start with -testnet

http://gpile.it/files/cryptonited MD5 a3398cf9f07ec28705d08081a4dab210

[billotronic]

lol i put a sticky note on my display to remind me.

[mrvegad]

lol,i shoot sticky on my display to remind me.

gross

[mrvegad]

Big news here. Latest release implements block pruning according to MBC scheme. So when you sync the new client it will be retrieving all of the block headers but only the contents of last 1k blocks. This version prints a lot of spam to the console. The driving force behind leaving the spam on for now is that the sync process is kind of involved and stuff like getinfo doesn't make any sense. It's really hard to tell if anything is going on until suddenly it is online. Getinfo also won't respond for a long period of time while the code is checking the downloaded data for consistency. So the spam kind of lets you know it is still working until we come up with improved UI. You can tell it's done syncing when the last line says "Ready for committal"

Restart after killing the client is also much slower now because of increased consistency checking. Again the spam will let you know what is going on.

As usual, expect lots of bugs. Performance enhancements are coming soon.

Mrvegad I am not confident in a fix for disappearing blocks yet. This could get especially interesting since all the blocks don't even exist anymore. I'm looking into it though. Complete rewrite of some temperamental code is on the short list.

Billotronic, yes you have to start with -testnet

http://gpile.it/files/cryptonited MD5 a3398cf9f07ec28705d08081a4dab210

nice! can't wait to break try it 

[mrvegad]

spam accepted here" n1V6XHG5RTwDaw1KymRjxYvsFM93Jbgjq2

[billotronic]

looking to break shit tonight

mhBpjn7iknihVoktgBkBERgPo9s5rXcKnD

[edit] is the testnet down? Not getting any connections

ok, keeps crashing on me while trying to synch

Code:

2014-06-21 00:34:31 AddToWallet a3199904ac62e9490511a5b31d023ceda9dcb545ddc9cf226167159db5fe0a58  
2014-06-21 00:34:31 ERROR: CheckTransaction() : vin empty
2014-06-21 00:34:31 ERROR: AcceptToMemoryPool: : CheckTransaction failed
2014-06-21 00:34:31 UpdateTip()

Three crashes, each the same error

log :: http://pastebin.com/tXEBxA1F

[catia]

looking to break shit tonight

Well you are doing a good job. I am writing code as fast as I can but can't keep it running.

edit: seems the disappearing blocks happen when a block gets orphaned. not quite sure if it's fixed but 7fdf8b6164131867ae46e227a1ba9441 is better.

[mrvegad]

looking to break shit tonight

mhBpjn7iknihVoktgBkBERgPo9s5rXcKnD

[edit] is the testnet down? Not getting any connections

ok, keeps crashing on me while trying to synch

Code:

2014-06-21 00:34:31 AddToWallet a3199904ac62e9490511a5b31d023ceda9dcb545ddc9cf226167159db5fe0a58  
2014-06-21 00:34:31 ERROR: CheckTransaction() : vin empty
2014-06-21 00:34:31 ERROR: AcceptToMemoryPool: : CheckTransaction failed
2014-06-21 00:34:31 UpdateTip()

Three crashes, each the same error

log :: http://pastebin.com/tXEBxA1F

Tough day billo?  Guess breaking this is better then kicking the dog 

[billotronic]

Tough week truthfully but the mrs had work shit to do last night so it was just going to be a romantic evening with the computer and myself... and the recently ported Civ V for linux... So of course this shit breaks on me right from the start.

[mrvegad]

Tough week truthfully but the mrs had work shit to do last night so it was just going to be a romantic evening with the computer and myself... and the recently ported Civ V for linux... So of course this shit breaks on me right from the start.

sorry to hear about your week, i will have to check out the Civ V port.

[mrvegad]

the first block pruning release was working great for me, i am now using the second release but i am not able to get "Ready for committal"  and in my debug.log i see lots of:
2014-06-21 14:50:48 ERROR: AcceptToMemoryPool: : transaction already exists
2014-06-21 14:50:50 connect() to 66.45.239.67:18253 failed after select(): Connection refused

[bitfreak!]

I asked catia about the "Ready for committal" thing and he said it only shows that on the first time. As for that error in the log, it looks like it might just be someone sending you a transaction your already have or something, but we'll have to wait for catia to take a look at it to be sure though. I assume getinfo and other commands are still working for you?

[catia]

There is a serious bug in orphan resolution. Good news is that I can reproduce it easily right now. I'll fix it later tonight, until then primary node is offline and experiment halted.

[Edit] I actually fixed something for once. No more disappearing blocks and orphans working well. d9204e91683f871ab8f9c638427fa71a  Let the games begin when you get a chance. Thanks all.

[billotronic]

mvLpTpus6gKcLRMvs72qUnvjXaoUAebxaV

[mrvegad]

mzWZMszmtp2oScRxvSKCZ8ZSCNFQNZqMzY

[catia]

does this mean you haven't managed to break it yet?

[billotronic]

I didn't monitor it much this weekend but this morning it seemed like the client had froze up. I tried a getinfo and nothing happened before I had to leave for work soooo maybe?