Jump to content

xaminmo

MEMBER
  • Content count

    61
  • Joined

  • Last visited

  • Time Online

    1h 46m 34s

1 Follower

About xaminmo

  • Rank
    Member

Recent Profile Visitors

507 profile views
  1. Lots of good references and discussion there. It feels like the natural follow-on to previous discussions. I might need a nap after replying though. :)
  2. @Mr. Wood, Sorry for the delay. I was on holiday (am still). Operationally, the snapshot2 that Rob linked should get you going. Long-term, it indicates there may still be some invalid blocks from old clients that cause current clients to get hung up. While that should happen less often over time, its still a risk/hole that the devs will want to address (if not already addressed in the dev branches being worked on now). I'll add some list items here for the past few days of posts: * downloadblocks just downlods the zip file of the snapshot. If it failed once, retrying will not usually help. * The snapshot2 that Rob linked is much newer, and should get past the block you are hung on. * Lots of orphans is not necessarily a problem. Any block from a different chain is an orphan, and they get used in reorg. * "socket sending timeout" in rapid succession may be related to the bad block, which may be related to a CPID tally defect. * It looked like you stair-stepped down a long fork. Can you post that with the block hash and number that you were hung on? It may be something in the GitHub list of bad blocks. Today, bad blocks are listed manually in the code to reject them; however, most of the worst causes are now handled automatically. If there is something still in the wild that's a problem, it needs developer attention. GitHub issues are usually the best way for detail/specific issues. * @UnitedNetworkers, repairwallet would not be related to blockchain hangs. The wallet is just the list of personal keys and transactions related to those keys. The things Mr. Wood is listing are block chain related, not really wallet related.* Vehemently rejected" is not super new. It just means that ProcessBlock returned false. The "CheckBlock FAILED" came from ProcessBlock, and is a preliminary check. As we follow the messages backwards, it looks like you staked a block, added mined credit, tried to sign it, and the new block was rejected as invalid. It's possible your beacon is not valid anymore, or something else. Basically though, when you end up on a lonely fork, lots of extra mining of junk blocks occurs locally, and you'll get weird errors like that.
  3. Last message I saw, it looked like you were still at 3.7.8. Is that correct? You would need to be at 3.7.10. You're far enough down-chain on a fork that reorganize will not automatically unlink your fork. You will need to do that manually. You can use "getblockhash" for block numbers backwards until you find a hash that matches one of the block explorers. https://grcexplorer.neuralminer.io or https://gridcoinstats.eu are both good. Once you find a matching block, you can use "reorganize ###" on that hash (not the block number) and it will unlink your block chain after that. Then, it will look for the longest chain, and connect you from there. It still will take a little time for the transaction database to catch up, but it should get you back going. There are other ways to restart entirely, such as "downloadblocks", but that replaces your entire block database with a snapshot, and then downloads the changes since then. That is likely to take a lot longer, and also the snapshot comes from just one server (inherently less secure).
  4. @Rob Halförd - (Gridcoin), thanks for the clarification to us all. I think I understand that @LennStar's main concern is that there was not a big, public discussion about the rate change. Renewal, and the new rate seemed perfectly rational from my perspective. I do not know of the founders and whales and etc had other discussions. I do not know the governance requirements on that either. I do know the dev community earns every bit of that. It's obvious they are humbled and honored by the adjustment. They are excellent resources, and it's good to keep them feeling appreciated. This is just my perspective from a consulting background, and the last 4 months of digging into the GRC community. I definitely have no authority beyond a couple thousand coins. I'm a long way off from providing valuable assistance with the project. Maybe by next year.
  5. You're not actually paying for anything. So far, Rob has been paying out of his wallet just barely more than he earns in interest. Did you read through the entire Developer Compensation thread ? If you look at the discussions, plus all of the developer communications, you'll find it is fairly clear what is being paid for. "mentioned that any increase above $30 needs discussion" is a bit misleading. A very small number of people and magnitude have mentioned their opinion. Based on the polls, only 10% of the voting population agrees with you. Note that this excludes any whales trying to swing the vote. They are staying out of the vote on this so far.
  6. @LennStar, I do not see why you have a problem with this. * The voice of the people is "this pay raise is acceptable." * The voice of the owner of the coins is "This is what I want to pay them with my own coins." Things you might not know: * The foundation wallet moved to S8VRnqQ1JZDuwSRxK4b94MiaArVULjgRUX in 2017 and has not spent any coins since then. * The dev compensation program is being paid for by Rob, from his own wallet * The vote for paying more is primarily YES, with no whales, and has 10x the participation vs your vote. * The devs did not release bad code. They inherited a time bomb. They worked very hard to fix it, and proved their worth. * $60/hr equivalent is a bargain for the core resources we are getting, especially since they under-bill anyway. The other things you complain about, if you really want them fixed, then you need to do the work. * If you think Twitter should have had more updates, then instead of complaining after it's all fixed, you should have offered to make the tweets. * If you think the documentation is so horrible for newbies, then instead of complaining about it, why don't you write up a better document? * If you wish you made $60 per hour, then instead of complaining, who don't you get some code commits merged? * If you think $60/hr is too much for anyone to make, then do the work for less.
  7. Yes. The sync from 0 issue was finally resolved in 3.7.10.0 This is at the standard installer locations. Write-up just posted an hour after your question, so you didn't miss it. It missed you. https://steemit.com/gridcoin/@barton26/gridcoin-leisure-update-3-7-10-0-released
  8. There have been about 90 fixes in the last week. Forking is triggered by a whole lot of issues that were hidden in v8 blocks. Lots of work is happening right now, but there's not a stable checkpoint yet. More people have joined in slack to try to help with some of the coding.
  9. Right now, reliability is still very low. Large amounts of work gets cancelled, fails, runs long, or is otherwise invalid. No credits given for the work done (which beta testing an app should be considered valid work, even if the result is bad). Work availability is spotty, even for the small number of computers assigned to the project. I love the concept of the project, and hope it stabilizes, but they should not have been whitelisted in the first place. They are just not ready. The project admins also believe the same thing, and have for a while. Mods: https://sourcefinder.theskynet.org/duchamp/forum_thread.php?id=269&postid=1245#1245 Project Owner: https://cryptocurrencytalk.com/topic/83717-duchamp-sourcefinder/?tab=comments#comment-426539 No major amounts of WUs for a while: https://sourcefinder.theskynet.org/duchamp/forum_thread.php?id=213#688 (He was expecting a few months from July 2017, but it depended on integration issues overall, which they are still dealing with.) More discussion about whitelisting vs beta here: https://sourcefinder.theskynet.org/duchamp/forum_thread.php?id=269 My guess is, shortly after it's delisted, it will be ready for more compute resources. Murphy's law says so anyway. :) And even though there are still 2 weeks before the poll is over, there's a lot of weight behind delisting.
  10. I agree on the separate stickey. Reddit and Github have them. Not sure who deals with that on CCT. I know there is a PPD issue open 3 days, and the remaining beacon fork issue is open 6 days. The devs have pretty decent code activity going on, but it's hard to guess how long it will take them to unravel it all. It looks like two guys working on the code, and one guy helping with QA. On here, they are Quez, Marco and iFoggz, but there are already so many places for them to have to spend time not developing (or earning money that can pay for housing and food). I've gotten lucky with my wallet, same as all of gridcoin stats, 0.441 difficulty, net wright 2.64m. However, I am paranoid and restart my wallet when I get up in the morning, plus again later in the day. I started that at 3.7.5, and continue it into 3.7.7. I know it's not reasonable, but it *seems* to help.
  11. Did you restart your wallet? That's the key fix for most forks right now. There is not a "final fix" per se. This is an ongoing development project. There are a couple of different issues open right now. For your issue, you'll want to look at debug logs and see if it matches the PPD issue already open. If so, follow that. If not, then open an issue on github and provide debug logs there. The dev notes for 3.7.7.0 were: After reorganize is fixed (hopefully in #902), this problem will not go away completely. If node is on a lesser fork and a beacon is advertised on first the lesser fork and then spreads naturally to the main chain, the node will not accept blocks from the main chain to Index. Reorg only works when both chains are in index, and so the node will get stuck and continue mining the lesser fork.
  12. Mostly okay. Tomas said there is still one fork issue left. If you're on a fork, and receive a beacon first from that same fork, then you won't accept blocks from the main chain after that. https://github.com/gridcoin/Gridcoin-Research/issues/904 However, most of the other bad issues are resolved, at least for those who updated to 3.7.7 and reset with the official snapshot.
  13. Ubuntu and MacOS hotfixes are available from the normal download locations. Version is 3.7.7. Windows and Red Hat packages are not out there yet. There is also a blockchain snapshot from today on the official site: http://download.gridcoin.us/download/downloadstake/signed/ Or if any concerns, just take the URL of the official windows package, and you'll see it's the same directory. The windows MSI is 3.7.5, so look for the 3.7.7 update soon with a date or 2018-02-01 or 2018-02-02.
  14. MacOX and Ubuntu packages are available for 3.7.7 from the official locations. Windows and RPM are not updated yet. Also, in the developer list, there is a blockchain snapshot from 2018-02-01. After updating both, and getting in sync, be sure to run repairwallet.
×

Important Information

By using CRYPTOCURRENCYTALK.COM, you agree to our Terms of Use.