1 2017-09-17 07:09:32 <hkjn0> hey folks, after upgrading to 0.15 I noticed that the getblockchaininfo RPC returns a *decreasing* verificationprogress on my node..
 2 2017-09-17 07:09:54 <hkjn0> the node was fully synced on 0.14, so I interpret this as the node receiving new blocks but for some reason being unable to verify them
 3 2017-09-17 07:11:03 <hkjn0> debug logs don't say much of anything, and while I could remove all the blocks or similar, I wanted to check if someone has hunches of what could be going wrong / what I could check
 4 2017-09-17 07:12:52 <gmaxwell> ugh. no.
 5 2017-09-17 07:13:17 <gmaxwell> 1) questions about core should go to #bitcoin-core-dev this channel is mostly dead.
 6 2017-09-17 07:13:49 <gmaxwell> 2) verificationprogress is a very rough estimate based on the time and height... rpc probably shouldn't return it  IIRC the GUI hides it once its over 90%
 7 2017-09-17 07:21:20 <hkjn0> ah, thanks gmaxwell. I switched to getblockchaininfo due to deprecation notice from getinfo, which I used until 0.14..
 8 2017-09-17 07:23:12 <hkjn0> but it is in the 97% range, so I'll just disregard it then, and try to understand what the number means to build the right intuition here..
 9 2017-09-17 07:24:13 <gmaxwell> it's generally impossible for a bitcoin node to know if it's synced or not, it's always trying to sync more. the verification progress is just a rough hurestic based on expected numbers of blocks vs time
10 2017-09-17 07:25:53 <hkjn0> yes, I get that, was just non-intuitive to see the number slowly dropping :)
11 2017-09-17 07:31:14 <hkjn0> ah, *now* I get what you're saying.. the verificationprogress number is just "how many blocks *should* I have gotten, if they arrived 10 min apart"? that makes perfect sense that it's very slowly decreasing then, because each time I look at  it (with 'watch') a few more seconds have gone by with no new block..
12 2017-09-17 07:32:02 <hkjn0> agreed that it might be more confusing to expose it in the RPC response than not, then.. you have at least one data point of someone being needlessly confused about it, right here!
13 2017-09-17 07:39:35 <gmaxwell> thanks!
14 2017-09-17 18:05:29 <knaccc> anyone know anything about this? https://www.w3.org/blog/wpwg/2017/09/14/payment-request-api-now-being-implemented-in-all-major-browsers-advances-on-the-recommendation-track/
15 2017-09-17 18:07:15 <knaccc> the coinbase article on this says: "The W3C is already working with third-party payments apps to integrate both cryptocurrency and non-credit card forms of payment into the API in a way that can be interpreted by merchants and consumers"
16 2017-09-17 23:31:26 <Krellan> Known bug already? In 0.15, on MacOS, it crashed for me every time during startup, until I ran it again with -resetguisettings option.
17 2017-09-17 23:46:03 <esotericnonsense> Krellan: there are a few issues that seem to be related to that.
18 2017-09-17 23:46:11 <esotericnonsense> https://github.com/bitcoin/bitcoin/issues/11262 is an example.
19 2017-09-17 23:48:09 <Krellan> Thanks - by luck I was running with -debug= when it happened, and also I saved the MacOS crash dialog, if that might help
20 2017-09-17 23:49:03 <esotericnonsense> Krellan: there's a (possible) fix for that specific issue that's already in master
21 2017-09-17 23:49:33 <esotericnonsense> Krellan: so it might not be necessary now
22 2017-09-17 23:49:41 <esotericnonsense> keep it around just in case i guess :)
23 2017-09-17 23:50:23 <Krellan> Good - thanks - have a GitHub link for that?
24 2017-09-17 23:50:41 <esotericnonsense> it's linked at the bottom of the page, 11332
25 2017-09-17 23:51:09 <Krellan> OK - final line in my debug text was this: GUI: TransactionTablePriv::refreshWallet