💬 maflcko commented on issue "Intermittent issue in test/ipc_tests.cpp Fatal glibc error: pthread_mutex_lock.c:450 (__pthread_mutex_lock_full): assertion failed: e != ESRCH || !robust":
(https://github.com/bitcoin/bitcoin/issues/29889#issuecomment-2854218017)
Is this still an issue with a recent version of Bitcoin Core? If yes, what are the steps to reproduce?

As mentioned above, running in `rr` could be the most useful.

✅ maflcko closed an issue: "Intermittent failure in rpc_bind.py: in get_socket_inodes FileNotFoundError: [Errno 2] No such file or directory: '/proc/34007/fd/23'"
(https://github.com/bitcoin/bitcoin/issues/29643)

💬 maflcko commented on issue "Intermittent failure in rpc_bind.py: in get_socket_inodes FileNotFoundError: [Errno 2] No such file or directory: '/proc/34007/fd/23'":
(https://github.com/bitcoin/bitcoin/issues/29643#issuecomment-2854221586)
b2e9fdc00f5c40c241a37739f7b73b74c2181e39

💬 Sjors commented on pull request "rpc, rest: Improve block rpc error handling, check header before attempting to read block data.":
(https://github.com/bitcoin/bitcoin/pull/30410#issuecomment-2854226908)
It would be nice to have unique (new) RPC error codes for this, because e.g. with ForkMonitor I was parsing the string that comes with `RPC_MISC_ERROR`, which this PR slightly changes.

https://github.com/jonathanbier/forkmonitor/commit/2d0e9f48cbd95b5ce52416ef497d1691ea3e24f0

💬 stickies-v commented on pull request "validation: stricter internal handling of invalid blocks":
(https://github.com/bitcoin/bitcoin/pull/31405#discussion_r2075270458)
> I think that's why sometimes one or the other is used.

Right, that makes sense.

I think in the two other cases where `CBlockIndexWorkComparator` is used in this PR, comparing directly on `nChainwork` would be more readable and more correct?
- https://github.com/bitcoin/bitcoin/pull/31405/files#diff-97c3a52bc5fad452d82670a7fd291800bae20c7bc35bb82686c2c0a4ea7b5b98R3704: "that have at least as much work as where we expect the new tip to end up."
- https://github.com/bitcoin/bitcoin/pull/3
...

💬 murrayn commented on pull request "Remove arbitrary limits on OP_Return (datacarrier) outputs":
(https://github.com/bitcoin/bitcoin/pull/32359#issuecomment-2854249655)
Concept NACK

I speak as a core contributor. I find this whole situation terrifying. There is clearly enough resistance to this PR that it should either be dismissed outright, or its rationale suitably justified to the whole community, whose judgement has been proven correct several times in the past. Yes, this may mean utilising X or Youtube instead of the mailing list or moderated github comments, as uncomfortable as that may be for some, but if there is a solid rationale for this change it
...

💬 0xmatt11 commented on pull request "Remove arbitrary limits on OP_Return (datacarrier) outputs":
(https://github.com/bitcoin/bitcoin/pull/32359#issuecomment-2854263883)
Removing the OP_RETURN limit is a sensible step. The 83-byte cap no longer serves its original purpose and only encourages inefficient workarounds. Larger, provably unspendable outputs don’t burden the UTXO set and open the door for legitimate use cases like sidechains and timestamping. Bitcoin should enable innovation, not restrict it arbitrarily.

💬 laanwj commented on pull request "rpc: Undeprecate rpcuser/rpcpassword, store all credentials hashed in memory":
(https://github.com/bitcoin/bitcoin/pull/32423#discussion_r2075287790)
i intentionally kept this as-is (eg converting to use `Split` slightly changes semantics), but thinking of it, calling `user_pass.find(':')`three times in a row is maybe a bit much.

💬 maflcko commented on issue "wallet: Data race in GetOrCreateLegacyScriptPubKeyMan vs IsMine":
(https://github.com/bitcoin/bitcoin/issues/27354#issuecomment-2854265066)
For reference, BDB removal https://github.com/bitcoin/bitcoin/pull/31250 was merged (for 30.0)

However, this issue may still exist, as `m_spk_managers` still exists.

💬 l0rinc commented on pull request "validation: periodically flush dbcache during reindex-chainstate":
(https://github.com/bitcoin/bitcoin/pull/32414#issuecomment-2854283411)
Finished measuring `reindex-chainstate` performance for max dbcache (best case scenario), where this PR shines most (i.e. to avoid the worst-case scenario of doing everything in-memory and crashing at the end without anything saved).

Measured it on an HDD with i7 processor, so disk writes are expensive here - would likely be better on SSD.
Unfortunately this change results in a 11% slowdown for scenarios that were hoarding memory, since it has to flush intermediary results regularly, so the
...

💬 instagibbs commented on pull request "policy: uncap datacarrier by default":
(https://github.com/bitcoin/bitcoin/pull/32406#discussion_r2075306983)
still fumbling this. In `mempool_datacarrier.py` I'm not explicitly stopping or restarting the nodes but allowing test teardown to do it. Should I just be stopping the node myself with the optional arg?

💬 instagibbs commented on pull request "policy: uncap datacarrier by default":
(https://github.com/bitcoin/bitcoin/pull/32406#discussion_r2075314466)
Documentation for this is in `src/kernel/mempool_options.h`, it can be further spelled out in policy.h for `IsStandardTx` if that helps?

👍 ismaelsadeeq approved a pull request: "fees: document non-monotonic estimation edge case"
(https://github.com/bitcoin/bitcoin/pull/31080#pullrequestreview-2818014449)
Code review ACK 1e0de7a6ba926487c8a075856b74af2a3a0eb8ef

💬 instagibbs commented on pull request "policy: uncap datacarrier by default":
(https://github.com/bitcoin/bitcoin/pull/32406#discussion_r2075329397)
done

💬 instagibbs commented on pull request "policy: uncap datacarrier by default":
(https://github.com/bitcoin/bitcoin/pull/32406#discussion_r2075329807)
done

💬 Sjors commented on pull request "rpc: Undeprecate rpcuser/rpcpassword, store all credentials hashed in memory":
(https://github.com/bitcoin/bitcoin/pull/32423#issuecomment-2854351055)
Hah, I was planning on doing this undeprecating this morning, but couldn't find the deprecation message, so I figured someone already did it.

Concept ACK

💬 ryanofsky commented on issue "Intermittent issue in test/ipc_tests.cpp Fatal glibc error: pthread_mutex_lock.c:450 (__pthread_mutex_lock_full): assertion failed: e != ESRCH || !robust":
(https://github.com/bitcoin/bitcoin/issues/29889#issuecomment-2854354693)
This seems a lot like issue https://github.com/bitcoin-core/libmultiprocess/issues/154 which was fixed by https://github.com/bitcoin-core/libmultiprocess/pull/159

✅ Sjors closed an issue: "Make -stopatheight work with background sync"
(https://github.com/bitcoin/bitcoin/issues/28809)

💬 Sjors commented on issue "Make -stopatheight work with background sync":
(https://github.com/bitcoin/bitcoin/issues/28809#issuecomment-2854361655)
I stopped caring about this :-)

💬 adamandrews1 commented on pull request "fees: document non-monotonic estimation edge case":
(https://github.com/bitcoin/bitcoin/pull/31080#issuecomment-2854362619)
Code review ACK 1e0de7a