05:25:17 sure 05:30:02 done, I think 05:30:53 triggered a build to make sure it works 06:34:50 Yep, looks great. Thanks! 07:28:00 Hi guys, how to re-login multisig wallet? 07:28:02 https://monero.stackexchange.com/questions/5646/how-to-use-monero-multisignature-wallets-2-2-2-3 07:28:40 This post just has brand new creation, not how to login a existing multisig wallet 09:20:26 Guest_48: I don't understand what you mean with "re-login a multisig wallet". Can you please elaborate the situation you are in? 11:56:04 moneromooo: You still looking for testers for the Monero Update thing? If so, which OS is most needed? 11:56:07 (testing wise) 12:03:39 Always :) Windows was a massive pita, iDunk spent hours on it. If you do try windows, it'll crash till a new version of gpgme. 12:03:57 Otherwise, less common ones that still support Qt I guess. 12:04:36 OK, will make a thread on Reddit then later 12:04:46 ty 13:29:34 Which jason parser would you recommend to connect to monero-wallet-rpc via C libcurl? Minimalistic perfered. 13:32:40 json-c 13:33:07 Assuming plain C, that is 13:41:53 Looks promising. 13:54:34 What is your opinion about this library? https://zserge.com/jsmn/ It has a very small footprint. But I'm not sure that's sufficient. 13:55:59 Never worked with it 14:08:51 fair enough. You wouldn't happen to have a very simple example of json-c in use!? e.G. get_version or get_balance from monero-wallet-rpc or so. 14:11:11 is there any reason Monero doesn't use include guards? 14:14:07 nonie_: Sorry, it's been years since I've worked with it. Nowadays I only use C++, so had no use for json-c 14:15:08 FWIW primo uses cJSON to talk to monerod. 14:16:11 peach34: Maybe because "#pragma once" is better and simpler? 14:21:59 cJSON (Ultralightweight JSON parser) nice, nice. 14:24:20 rbrunner I don't see that much where I'm looking either 14:24:34 But yeah I agree 14:32:55 xiphon: stoffu: I added remember-where-a-tx-came-from-to-drop-if-invalid to 5968 and you had reviewed that PR before, you might want to check the addition. 14:33:31 moneromooo: kk 14:37:34 xiphon: also can you check whether you're ok with my last comment in 6103 ? 14:38:58 peach34: Maybe I was lucky! I just randomly grabbed this and found it there: https://github.com/monero-project/monero/blob/master/src/daemon/daemon.h 14:39:45 moneromooo: acked it 14:40:20 ty 14:40:38 sure, np 14:52:47 Anyone with established history wants to add their (long term) signing pubkey to the gitian checker ? You would not have to post sigs every release without fail. 15:16:18 i don't think i've ever posted my sig elsewhere so i dunno if its useful.. .? i made a key once upon a time 15:17:36 does it work with gpg curve25519 keys or does it need to be rsa? 15:18:13 If you want to try to build gitian on releases and PR to gitian.sigs, it'd be useful. 15:18:23 It will work with whatever GPG likes. 15:18:45 i think the version in debian stable supports it 16:19:15 is there anything that could prevent a coredump being written? 16:19:39 monero-wallet-gui segfaulted on me and I didn't get a coredump, I'm currently trying to reproduce it with gdb attached 16:21:39 ulimit, /proc/sys/kernel/core_pattern, read only filesystem, ENOSPC, possibly some capability. 17:49:19 moneromooo: i've just noticed that with v0.15 there is still a burst of about 4MB of data when the wallet initially connects to the server, i seem to recall from a few months ago that it was possible to instruct the server to compress the data stream (via curl back then) and it turned into something like 600KB. do you know what it is about? 17:49:46 Is it a trusted daemon ? 17:49:52 yes 17:50:30 4 MB seems like a lot. I don't recall seeing that. 17:50:56 I'll add to the list to look at. Low priority though. 17:52:46 it's especially noticeable over tor as you might imagine 17:53:29 simple tcpdump should provide enough evidence, i recall back then that the options to compress were already available server-side, but not client-side, and I thought the client woudl eventually land support for compressing and never thought about it again 17:53:48 but i'm back to using onion services now and ran into the same thing, so just fyi 17:53:54 Was your wallet synced btw ? 17:53:58 If not, it syncs. 17:54:05 Just making sure :P 17:54:06 yeah about 1000 blocks behind 17:54:12 and it does sync 17:54:21 just takes a very long time (minutes) to kickstart the process over onion 17:54:40 while the background data transfer is happening, the wallet just says "Starting refresh ..." and is unresponsive 17:54:43 Were those 1000 blocks about 4 MB ? :) 17:54:51 or "Refreshing" .. 17:55:13 I thought about that, but it was also 4MB months ago. too much of a coincidence I think, but I can also check with a synced wallet if you wish 17:59:11 oh 17:59:18 client<->server rpc is encrypted now? 17:59:19 that's new 17:59:51 so much for the manual curl :p 17:59:55 but good development! 18:00:36 i'll look at this with more time tomorrow. should still be possible, but tcpdump is useless in this case. take care! 18:12:16 you can turn off SSL still 18:17:29 curl does ssl auth just fine. 18:28:38 yeah but you have to bend over backwards to tcpdump it 18:30:09 in OpenLDAP we use layered protocol modules and we can stick a "debug" module anywhere into the stack. so we can do plain protocol -> network socket, or plain -> debug -> socket (and when debugging is on, gives hex/alpha dump of in and out traffic) 18:30:19 and plain -> debug -> ssl -> debug -> socket 18:30:30 Sure. I'm not claiming anything about tcpdump. 18:30:38 although usually the debug output below the SSL layer isn't much interesting 18:30:59 I did have a patch to dump traffic if that helps. It's in a closed patch, should be findable with "dump". 18:31:15 but this way we can always see our traffic, even if SSL or GSSAPI or IPsec security layers are in use 18:31:22 I'm not sure whether it dumps SSL or before though... 22:19:17 What do people think of mms rpc ? It's meant to make multisig easier to use, so it doesn't really make sense for rpc I think. 22:20:00 It's not invasive though I guess. 22:20:07 (I've not read the patch yet)