[08:14:17] My node crashed outright [08:14:33] Closed out because of an assert that was missing a lock [08:14:54] But that was for ther QT wallet, not the daemon [13:45:59] @aguycalled I was able to compile and run PR 557 on osx [13:46:19] Here is what the wallet looks like compiled natively [13:46:20] https://cdn.discordapp.com/attachments/416000318149754881/600654443301568518/splash.jpg [13:46:23] https://cdn.discordapp.com/attachments/416000318149754881/600654460632694835/wallet.jpg [13:46:40] Details for the os and tools [13:46:41] https://cdn.discordapp.com/attachments/416000318149754881/600654532707352577/autoconf.jpg [13:46:42] https://cdn.discordapp.com/attachments/416000318149754881/600654538583834625/clang.jpg [13:46:43] https://cdn.discordapp.com/attachments/416000318149754881/600654543583445057/macOS.jpg [13:46:44] https://cdn.discordapp.com/attachments/416000318149754881/600654546758402048/make.jpg [13:47:19] Commands I ran to build [13:47:22] cd depends make cd .. ./autogen.sh ./configure --prefix=`pwd`/depends/x86_64-apple-darwin17.7.0 make [14:00:41] XCode version [14:01:56] only different is my clang and osx version [14:01:58] https://cdn.discordapp.com/attachments/416000318149754881/600658382512324618/2019-07-16_19-59-16.jpg [14:02:04] $ clang --version Apple LLVM version 10.0.1 (clang-1001.0.46.4) Target: x86_64-apple-darwin18.5.0 Thread model: posix InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin [14:02:16] https://cdn.discordapp.com/attachments/416000318149754881/600658458294878208/unknown.png [14:02:24] I see [14:02:28] What about SDK? [14:02:36] This is using older Xcode [14:02:40] 10.1 [14:02:47] Which is using older SDK [14:03:01] https://cdn.discordapp.com/attachments/416000318149754881/600658648007573515/unknown.png [14:03:07] Hmmmm [14:03:13] Interesting [14:03:20] So the toolchain should work [14:03:27] You are able to compile right? [14:03:35] But only issue is it breaks when running? [14:03:46] Ohh, I think I know why [14:04:16] Are you running a different version of QT in a system installed DIR maybe? [14:04:20] Like maybe in /opt ? [14:05:09] theres 5.12 in 7usr/local/opt [14:05:59] Hmmm [14:06:05] Could that cause it? [14:06:17] The system I tested on did not have a system install of QT [14:06:20] Only the QT in depends [14:07:23] it should not [14:07:29] i'm trying a make clean [14:07:36] Ack [14:07:45] That might help [14:07:54] I've run into issue before with a non clean src dir [14:13:43] ok it works now [14:13:45] @mxaddict [14:15:31] nice [14:15:38] 👌 [14:18:59] it gets hang and unresponsive while syncing on mainnet [14:19:08] devnet behaves correct [14:19:29] Hmmm, Ok, I'll try and debug [14:19:35] if i put the focus out of the window cant go back [14:19:38] But on linux it syncs nicely [14:19:43] cannot move out of the syncing status dialog neither [14:23:40] I'm not sure what you mean by hung? [14:23:49] Like you can't use any of the buttons on the wallet? [14:24:06] And by synching status, you mean the new UI that shows the progress? [14:26:00] shows the progress, but gets frozen at some point and the app is unresponsive [14:26:12] i have to force kill the process [14:26:13] I see [14:26:22] Is that with a clean datadir? [14:26:42] nope [14:26:50] trying with [14:26:54] But sounds like a deadlock [14:27:15] clean datadir goes fine [14:27:28] ill leave it syncing while i prepare some food [14:27:38] Ack [20:51:55] I'm rich! I'm wealthy~ I'm well to do* [20:51:56] https://cdn.discordapp.com/attachments/416000318149754881/600761550453866528/Screenshot_from_2019-07-17_02-51-22.png [20:52:01] Thanks for the testnet coins [21:08:08] @prole @aguycalled [21:08:18] I have a PR that I think should be added in 4.7.0 [21:08:20] https://github.com/navcoin/navcoin-core/pull/571 [21:08:42] It's unlikely to happen often this crash, but still worth adding [21:09:43] i can confirm i had that last week too, probably better if it's included [21:10:09] @salmonskinroll did it happen on the RC node? or a normal 4.6.0 node? [21:10:36] normal 4.7 node i think, i downloaded the binary on github [21:10:42] From what I can tell it was also possible to happen with the old version, but I think more unlikely as it was using a more stable API to load price data [21:11:00] I don't think we have the 4.7 bins in the github releases [21:11:18] bet 1 nav on it? 😄 [21:11:37] https://wetransfer.com/downloads/549e47e9fd40af44d4f2d46d2dbb750d20190714073050/c0587e [21:11:55] well it's not in the releases, but in pr [21:12:05] Ahh, I see 😄 [21:12:09] I when bet 🤣 [21:12:20] nooooooo [21:12:25] Hahahahahahaha [21:53:10] i managed to reindex the latest 4.7.0 (built locally with depends) without encountering the 0000000 cfund payment request issue. i still had to reconsider block 2.7million whatever it is. [21:58:58] I've tried 4 full syncs now [21:59:08] I did not get the 00000 cfund error [22:07:43] @prole If you get a second today, can you check out https://github.com/navcoin/navcoin-core/pull/571 [22:20:00] sure [22:20:13] i put another bug that's happening for me in github [22:20:14] https://github.com/navcoin/navcoin-core/issues/572 [22:20:26] not sure if anyones seen this before [22:21:33] im getting this on the debian node im setting up for navpay [22:24:00] gonna keep testing, could be a local setup issue on my server [22:24:08] but thought id just chuck it there for now [22:26:44] I'll see if I can replicate [22:27:48] I think that issue was in BTC as well [22:27:54] But looks like they patched it out [22:28:24] BTC code: cpp void CConnman::ThreadSocketHandler() { while (!interruptNet) { [22:28:32] so.. if i wipe my chainstate & blocks the daemon starts okay [22:28:39] Ours: cpp void ThreadSocketHandler() { unsigned int nPrevNodeCount = 0; while (true) { [22:28:45] but if i use the data i bootstrapped into the navpay server, this happens [22:29:07] Basically the while(true) might be the cause [22:29:12] dunno why the heck that would effect the banlist loading [22:30:00] The last few messages I said here were about the stuck node when calling stop [22:30:08] ah right [22:31:18] @aguycalled do we have any custom code in src/net.cpp ? [22:31:37] If not we could pull the upstream version to pull in any updates [22:32:12] If we do have custom functionality in that file, we might have to manually patch into an update src/net.cpp from upstream [22:35:09] i would say no, but im not sure [22:35:59] Ok, I'll take a look