--- Log opened Wed Oct 27 00:00:15 2021
04:12 -!- Netsplit *.net <-> *.split quits: stick
04:15 -!- Netsplit over, joins: stick
04:45 -!- dr-orlovsky [~dr-orlovs@31.14.40.18] has joined #bitcoin-rust
04:52 -!- Netsplit *.net <-> *.split quits: dr_orlovsky
06:02 -!- tibo [~tibo@240d:1a:afa:1d00:74c4:18a5:8b1b:e510] has quit [Remote host closed the connection]
06:55 -!- dr-orlovsky [~dr-orlovs@31.14.40.18] has quit [Ping timeout: 260 seconds]
07:58 -!- davidbak [~DavidBaki@c-174-61-163-5.hsd1.wa.comcast.net] has joined #bitcoin-rust
08:00 -!- DavidBakin [~DavidBaki@c-174-61-163-5.hsd1.wa.comcast.net] has quit [Ping timeout: 264 seconds]
08:00 -!- b10c [uid500648@id-500648.ilkley.irccloud.com] has joined #bitcoin-rust
09:55 < BlueMatt> ariard: ^
10:04 -!- b10c [uid500648@id-500648.ilkley.irccloud.com] has quit [Quit: Connection closed for inactivity]
10:26 -!- shatnerz [~shatnerz@h127209.man.tk-internet.pl] has joined #bitcoin-rust
12:27 -!- jrawsthorne [~jrawsthor@static.235.41.217.95.clients.your-server.de] has quit [Quit: ZNC 1.8.1 - https://znc.in]
12:27 -!- jrawsthorne [~jrawsthor@static.235.41.217.95.clients.your-server.de] has joined #bitcoin-rust
13:50 < ariard> BlueMatt: yep, i'm eager to do so, just not able to catch back a fuzz failure with the fuzz target input provided here : https://github.com/rust-bitcoin/rust-lightning/pull/1054#issuecomment-930644283
13:50 < ariard> though i think the unfixable force-close behavior it still here, just harder to hit with the new buffer
13:51 < BlueMatt> ah, i have a new sample input for you, then
13:51 < BlueMatt> sec
13:52 < ariard> ah awesome happy to try it!
13:52 < BlueMatt> try this https://0bin.net/paste/E01k-El1#i5F7+eilPNEeIGaFbV9qtXIrYdPFAfTQVf8FRVgXi5r
13:57 < ariard> BlueMatt: trying this, just pushed a small nit correction on 1054, apart of the fuzzer detect let me know if I should update other things
13:59 < BlueMatt> ariard: I think its mostly good but tbh haven't had a chance to review it in full, it was mostly good though
13:59 < BlueMatt> ariard: been insane around here, users waiting to do pre-prod-testing on lack of payment retires, but hopefully we finish those up ~tomorrow.
13:59 < BlueMatt> luckily not folks trying to do routing nodes or opening channels with untrusted counterparties :)
14:01 < ariard> yeah using ldk as a prod routing node rn with untrusting counterparties that would be bold!
14:01 < ariard> didn't get a chance to dig into the payment retries yet, mostly adding a module on top of the send_payment call?
14:01 < ariard> and be sure that's stateful when you crash in middle, or what can get wrong
14:04 < BlueMatt> yea, the retries pr itself is now quite small (well, was, we landed it, but one more rebuild on it), we ended up with all the logic in the new ChannelManager::retry_payment
14:04 < BlueMatt> but, yea, should all be crash-safe, even if the channelmanager wasnt persisted after the send started (!)
14:18 < ariard> BlueMatt: goood now i've the crash, a README.md in fuzz/ is really cool to have!
14:18 < ariard> missed we added one :p
14:18 < ariard> for payment retry, if the channelmanager wasn't persisted and the HTLC is still pending isn't the channel gonna be force-closed by counterparty?
14:19 < ariard> or like inconsistency between the payment retry module and channelmanager state
14:20 < BlueMatt> ariard: yea, we handle that!
14:20 < BlueMatt> ariard: and, yea, at that point the channel will go on chain, but the payment is still pending cause its now on chain
14:20 < BlueMatt> the counterparty can still claim it (obviously)
14:20 < BlueMatt> so we have to handle that
14:21 < BlueMatt> ariard: see https://github.com/rust-bitcoin/rust-lightning/blob/main/lightning/src/ln/payment_tests.rs#L465
15:02 < ariard> BlueMatt: okay so added a failure exemption in chanmon_consistency though note i've removed a fuzz-only check in commitment_signed reception
15:02 < ariard> i think it was too zealous and in fact requiring the sender to always have a +2 buffer, even if it was correctly anticipated
15:03 < ariard> like a concurrent inbound HTLC correctly anticipated by the outbound update_fee sender
15:05 < BlueMatt> ariard: cool! will take a look soon...hopefully
15:12 < ariard> BlueMatt: good staying around if needed, looking on the new whole InvoicePayer thing!
15:12 < BlueMatt> ariard: it wont happen tonight, I think, sadly.
15:14 < ariard> Oky, still not hungry for dinner, so still around if it happens :)
15:15 < BlueMatt> heh, well I've gotta leave for dinner soon
15:16 < BlueMatt> ariard: if you want to review https://github.com/rust-bitcoin/rust-lightning/pull/1141 rewrites the "guts" of InvoicePayer fixing them, so you can review that and basically review the InvoicePayer standalone.
15:29 < ariard> BlueMatt: yeah sounds a good start!
17:23 -!- tibo [~tibo@fp5a950abc.tkyc110.ap.nuro.jp] has joined #bitcoin-rust
18:08 -!- tibo [~tibo@fp5a950abc.tkyc110.ap.nuro.jp] has quit []
18:12 -!- tibo [~tibo@240d:1a:afa:1d00:c4b1:ad11:f446:556f] has joined #bitcoin-rust
--- Log closed Thu Oct 28 00:00:16 2021