15:55:37 @rbrunner7:monero.socialwait so youre adding the fix on the polyseed side so that the current bg sync code (without the hotfix) will simply work? 16:28:24 Not exactly. The current fix reloads something from the wallet.keys file to get the Polyseed and optional passphrase set again. Works, but is something like a "brute-force solution". It's a bit wasteful, because the same data was loaded immediately before anyway, Polyseed or not Polyseed, in another place of the code. What I [... too long, see https://mrelay.p2pool.observer/e/n7aoq4QLNzF0Y3Uw ] 16:48:00 I may be unable to make it today's meeting. My update in advance: worked on diagnosing a stressnet-specific chain split, and patching / investigating other observed issues on the stessnet. I'm personally OK with moving forward with @jeffro256's patch for the chain split here: https://github.com/seraphis-migration/monero/pull/391 16:56:24 > <@rbrunner7> About why loading the keys again probably works: The load before that load_keys just a few lines further up is for background sync type BackgroundSyncCustomPassword, while the loading of the keys happens for BackgroundSyncReusePassword. Does not really look like a puzzle ... but anyway, I agree that [... too long, see https://mrelay.p2pool.observer/e/7oCPrIQLOFFsWkhJ ] 16:56:24 Yep, little silly of me there. Too much context switching 16:59:24 Meeting in about 1 hour 18:00:01 Meeting time. Hello! https://github.com/monero-project/meta/issues/1390 18:00:13 Hello 18:00:35 Hi 18:01:46 Howdy 18:01:57 Hi 18:03:05 Alright. Let's start with the reports about last week. 18:03:19 @jberman:monero.social reported before the meeting: "My update in advance: worked on diagnosing a stressnet-specific chain split, and patching / investigating other observed issues on the stessnet. I'm personally OK with moving forward with @jeffro256's patch for the chain split here: https://github.com/seraphis-migration/monero/pull/391" 18:03:24 Hey 18:03:58 Me: Polyseed. To my surprise noticed only yesterday that monero-wallet-rpc also deals with seeds :) 18:04:52 Me: trying to salvage the stressnet chain state, hot/cold stuff, knowledge proof stuff, and RPC improvements 18:05:02 Me: working on updating lws and lwsf patches for some minor changes to upstream code. This includes the curve tree serialization and rename of a carrot function 18:05:11 Me: did some work on the main I2P integration PR; submitted a few patches related to Tor addresses and the build system; got some minor FCMP++ Rust build optimizations merged 18:05:25 I was all over the place, now working on the restore height approximation again 18:05:55 @jeffro256:monero.social: What do you mean with "salvage the stressnet chain state"? 18:06:17 me: multisig wip 18:07:40 @rbrunner7:monero.social: there's a temporary fork in the network , I'm working on a fix which may or may not merge them 18:07:50 Ok :) 18:08:10 You mean right now we have two competing stressnets? 18:08:53 Do they have some consensus related disagreements? 18:10:17 Yep ! 18:10:58 The block weight cache is different b/t them which in turn affects the coinbase tx reward 18:11:22 @vtnerd:monero.social: By the way, is there any estimate about some first "production version" for lws? Or will this be some very gradual process, the software just becoming better and better? 18:11:46 Cool. So interesting case of chain split, interesting to study I guess 18:11:51 @rbrunner7: 1.0 is either out or being prepared 18:12:09 (and lws in use by edge wallet) 18:12:26 Well, if it's out, that was one stealth release :) 18:12:52 Where are the trumpets and the fanfares and the big applause 18:13:26 No, seriously, I see this as quite some milestone worthy of proper attention 18:13:43 @rbrunner7: I think prepared. I dont aee 1.0 branch. Cant rememeber where i read about it - maybe a ccs update? 18:14:14 Anyway, good to hear that it's close. 18:15:00 So if I hop onto stressnet it's more or less random which of the two chains my daemon will chase down, right? 18:15:49 No necessarily 18:15:59 *not 18:16:37 There were 10 competing chains at a specific height, but most nodes that got past that roadblock ended up on the same chain 18:17:02 If you never restart, then you will be on the "main" chain. Or if you pop blocks then sync. If you restart your node tho, then you effectively fork from the network wherever you stopped 18:17:07 The joys of beta testing 18:17:55 Lol yeah except that it an issue which would never appeared on mainnet lol 18:18:28 So the nodes that synced past the roadblock, are on a non-consensus fork? 18:18:43 If you didnt restart, the node just gkt stuck afaict 18:19:11 Not sure. I think moneromooo once tried to chase down a rare problem where daemons did not agree regarding, what was it, longtime difficulty average? 18:20:52 Anyway, that what testing is for. After all, stressnet could just crash and burn. It holds instead - more or less ... 18:21:19 Alright, after these reports, do we have something to discuss today? 18:21:46 @ofrnxmr: There's not necessarily any single roadblock. The road block is every block if you restart your node 18:22:34 A thousand blooming networks, cool 18:23:36 Looks like we are through. Short meeting, but with good attendance, which is nice. Thanks everybody, read you again next week! 18:23:53 thanks everyone, see you 18:24:12 @ofrnxmr: Depends on what you mean by non-consensus. It wasn't an intended consensus rule, but it is what it is. AFAICT, the fork which currently has the most PoW is the fork that all nodes can converge onto if they apply #391, and #391 works as intended 18:25:37 Thanks 18:26:08 Thanks everyone 20:09:19 Yes 1.0 is going to include mempool scanning that is the holdup. It's already used in production environments so hopefully all of the changes don't trigger a regression > <@ofrnxmr> 1.0 is either out or being prepared