00:03 <zsexa> stellar is dead ya?

00:06 <zsexa> i guess so

00:11 <stellar-slack> <jed> Zsexa. Nope

00:39 <stellar-slack> <lab> IMHO, it is hibernating.

07:18 <stellar-slack> <armed10> Are there any circumstances known where a stellar-core returns a transaction with 'PENDING' and the transaction does not make it into the database?

07:24 <stellar-slack> <armed10> When testing the network for speed, roughly 20% of transactions are 'PENDING' but never put into a database.

15:24 <stellar-slack> <scott> @armed10: I’m not sure of your total context, but the pending response from submitting a transaction to stellar core doesn’t imply that the transaction will eventually succeed, only that the stellar-core has checked for well-formedness and added it to the pending transaction list. It’s still possible for the transaction to fail. When you say “never being put into a database” do you mean stellar-c

15:28 <stellar-slack> <scott> I don’t believe there should be a situation where a transaction that was received as pending will never end up in the database, but I wouldn’t be surprised if there was a situation under load where pending transactions are dropped

15:30 <stellar-slack> <scott> The code around (https://github.com/stellar/stellar-core/blob/ffd1ccda4b81e5986346f4ee5b08c970495d642a/src/herder/HerderImpl.cpp#L1324) seems to drop transactions that have become invalid since initially being submitted to the network, although I’m not very familiar with the stellar-core codebase so I could be missing some context

15:55 <stellar-slack> <dchrist22> @armed10: I've got rid of my "lost transactions" while stress testing, by using a bigger ledger size. In my case I used a ledger size of 3000 and all transactions showed up in the database, since I didn´t fill the hole ledger with transactions any more. Do you maybe also have full ledgers while performing your test?

16:10 <stellar-slack> <jed> dchrist22 you mean the max txset size?

16:15 <stellar-slack> <dchrist22> @jed: DESIRED_MAX_TX_PER_LEDGER in the config file