💬 mzumsande commented on issue "net: Tor service target port collides when running multiple nodes, making bitcoind error out":
(https://github.com/bitcoin/bitcoin/issues/31133#issuecomment-2429776658)
I do think that keeping `-port` around could be helpful, especially for users less knowledgeable about networking, who aren't sure which address to put in an equivalent `-bind` command.
I also like option 2, provided that the behavior will be well-documented, so that users specifying `-port` values for multiple nodes don't pick consecutive ones by accident (which would then again lead to collisions). Otherwise I don't see a downside.
I would also be ok with option 6, if I specify a port li
...
(https://github.com/bitcoin/bitcoin/issues/31133#issuecomment-2429776658)
I do think that keeping `-port` around could be helpful, especially for users less knowledgeable about networking, who aren't sure which address to put in an equivalent `-bind` command.
I also like option 2, provided that the behavior will be well-documented, so that users specifying `-port` values for multiple nodes don't pick consecutive ones by accident (which would then again lead to collisions). Otherwise I don't see a downside.
I would also be ok with option 6, if I specify a port li
...
💬 sdaftuar commented on pull request "cluster mempool: Implement changeset interface for mempool":
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811072822)
I tried to make this code analogous to the old code, which repeated information for each transaction. I'm open to changing the logging but I am always a little afraid of breaking user's scripts or habits if people are used to grepping for a transaction ID and seeing more information? (Though I guess this PR already is breaking the logging for package RBF... so maybe not a big concern.)
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811072822)
I tried to make this code analogous to the old code, which repeated information for each transaction. I'm open to changing the logging but I am always a little afraid of breaking user's scripts or habits if people are used to grepping for a transaction ID and seeing more information? (Though I guess this PR already is breaking the logging for package RBF... so maybe not a big concern.)
💬 sdaftuar commented on pull request "cluster mempool: Implement changeset interface for mempool":
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811074008)
Not sure if the tracing API is well-specified for package based replacement. I'm open to doing something different than what I propose here though, suggestions?
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811074008)
Not sure if the tracing API is well-specified for package based replacement. I'm open to doing something different than what I propose here though, suggestions?
💬 Christewart commented on issue "net: Tor service target port collides when running multiple nodes, making bitcoind error out":
(https://github.com/bitcoin/bitcoin/issues/31133#issuecomment-2429789160)
I'm in the camp of deprecating and then removing the `-port` setting. It seems we should should just get people to migrate to bind. Is `-port` strictly a subset of `-bind` functionality? If so it seems simplifying the config options would make sense.
(https://github.com/bitcoin/bitcoin/issues/31133#issuecomment-2429789160)
I'm in the camp of deprecating and then removing the `-port` setting. It seems we should should just get people to migrate to bind. Is `-port` strictly a subset of `-bind` functionality? If so it seems simplifying the config options would make sense.
💬 laanwj commented on pull request "rest: Support transaction broadcast in REST interface":
(https://github.com/bitcoin/bitcoin/pull/31065#issuecomment-2429793735)
Concept NACK from me, sorry
i'd prefer to keep REST as read-only interface. As an unauthenticated interface, allowing any kind of posting exposes bitcoind to cross-site scripting attacks through web-browsers, for example.
There's also a scope-creep issue. If we expose more and more of the RPC interface on REST, we end up having to maintain and test two interfaces for everything. And they're not even so different.
> This would improve UX (no need to authenticate) and security (only a sma
...
(https://github.com/bitcoin/bitcoin/pull/31065#issuecomment-2429793735)
Concept NACK from me, sorry
i'd prefer to keep REST as read-only interface. As an unauthenticated interface, allowing any kind of posting exposes bitcoind to cross-site scripting attacks through web-browsers, for example.
There's also a scope-creep issue. If we expose more and more of the RPC interface on REST, we end up having to maintain and test two interfaces for everything. And they're not even so different.
> This would improve UX (no need to authenticate) and security (only a sma
...
💬 sdaftuar commented on pull request "cluster mempool: Implement changeset interface for mempool":
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811083143)
Agreed, done.
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811083143)
Agreed, done.
💬 sdaftuar commented on pull request "cluster mempool: Implement changeset interface for mempool":
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811083287)
Agreed, done.
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811083287)
Agreed, done.
💬 sdaftuar commented on pull request "cluster mempool: Implement changeset interface for mempool":
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811084319)
Thanks that is much better; the only reason I had introduced `GetAggregateFeeRate()` was for access in this one place, so now that is gone.
(https://github.com/bitcoin/bitcoin/pull/31122#discussion_r1811084319)
Thanks that is much better; the only reason I had introduced `GetAggregateFeeRate()` was for access in this one place, so now that is gone.
💬 sdaftuar commented on pull request "cluster mempool: Implement changeset interface for mempool":
(https://github.com/bitcoin/bitcoin/pull/31122#issuecomment-2429804570)
Thanks @glozow and @ismaelsadeeq for the review; I've incorporated some of the comments. If you have suggestions for how the logging can be improved in the package RBF case (and how the USDT tracing messages can be improved, similarly) please let me know your thoughts.
Also, I wanted to flag that the locking is a bit of a mess in this branch. Initially, I tried to add lock annotations to the `CTxMemPoolChangeSet` functions to guarantee that the mempool itself was locked whenever any of thos
...
(https://github.com/bitcoin/bitcoin/pull/31122#issuecomment-2429804570)
Thanks @glozow and @ismaelsadeeq for the review; I've incorporated some of the comments. If you have suggestions for how the logging can be improved in the package RBF case (and how the USDT tracing messages can be improved, similarly) please let me know your thoughts.
Also, I wanted to flag that the locking is a bit of a mess in this branch. Initially, I tried to add lock annotations to the `CTxMemPoolChangeSet` functions to guarantee that the mempool itself was locked whenever any of thos
...
💬 laanwj commented on pull request "cleanse: switch to SecureZeroMemory for Windows cross-compile":
(https://github.com/bitcoin/bitcoin/pull/26950#issuecomment-2429812545)
i think the goal here was to make Windows code consistent between MSVC and mingw. MSVC is windows' official toolchain so it's leading for that. Using something else for Mingw could be considered a workaround.
(https://github.com/bitcoin/bitcoin/pull/26950#issuecomment-2429812545)
i think the goal here was to make Windows code consistent between MSVC and mingw. MSVC is windows' official toolchain so it's leading for that. Using something else for Mingw could be considered a workaround.
💬 jonatack commented on pull request "rpc, cli: return "verificationprogress" of 1 when up to date":
(https://github.com/bitcoin/bitcoin/pull/31135#discussion_r1811126433)
Note to self, using the "validated" field isn't a good check here, best to use blockheight equals headers as in getblockchaininfo.
(https://github.com/bitcoin/bitcoin/pull/31135#discussion_r1811126433)
Note to self, using the "validated" field isn't a good check here, best to use blockheight equals headers as in getblockchaininfo.
📝 jonatack converted_to_draft a pull request: "rpc, cli: return "verificationprogress" of 1 when up to date"
(https://github.com/bitcoin/bitcoin/pull/31135)
in getblockchaininfo/-getinfo and getchainstates, as requested in issues https://github.com/bitcoin/bitcoin/issues/31127 and https://github.com/bitcoin/bitcoin/issues/26433. Verification progress estimates in the debug logging remain unchanged.
(https://github.com/bitcoin/bitcoin/pull/31135)
in getblockchaininfo/-getinfo and getchainstates, as requested in issues https://github.com/bitcoin/bitcoin/issues/31127 and https://github.com/bitcoin/bitcoin/issues/26433. Verification progress estimates in the debug logging remain unchanged.
💬 Kasucode101 commented on pull request "kernel: Introduce initial C header API":
(https://github.com/bitcoin/bitcoin/pull/30595#issuecomment-2429867912)
In coding see the difference between natural coding and human coding beyond AI coding to measure code performance in Financial technology use case in pursuit of transition from bad money to good money in money market regardless of block chain/cryptocurrency or Quantum financial technology not only in creation/source of money but also in spending/allocation of money with developed/preferred/conscious plan rather than less plan with more budget leading to wasting resources even terrible ideas beyo
...
(https://github.com/bitcoin/bitcoin/pull/30595#issuecomment-2429867912)
In coding see the difference between natural coding and human coding beyond AI coding to measure code performance in Financial technology use case in pursuit of transition from bad money to good money in money market regardless of block chain/cryptocurrency or Quantum financial technology not only in creation/source of money but also in spending/allocation of money with developed/preferred/conscious plan rather than less plan with more budget leading to wasting resources even terrible ideas beyo
...
💬 TheBlueMatt commented on issue "Mining Interface doesn't allow for Bitcoin Core to create blocks when it wants":
(https://github.com/bitcoin/bitcoin/issues/31109#issuecomment-2429943359)
> But aren't we holding cs_main throughout the process?
Doesn't mean we can't build a new block. There's several ways we might build a new block here - (a) we might precalculate an oversized next-next-block in advance and just walk through and quickly remove conflicts to get us mining on a non-empty template as fast as possible, (b) we might delay updating the mempool entirely waiting on new block building - if we're mining the most important thing is the new work (and relay might work withou
...
(https://github.com/bitcoin/bitcoin/issues/31109#issuecomment-2429943359)
> But aren't we holding cs_main throughout the process?
Doesn't mean we can't build a new block. There's several ways we might build a new block here - (a) we might precalculate an oversized next-next-block in advance and just walk through and quickly remove conflicts to get us mining on a non-empty template as fast as possible, (b) we might delay updating the mempool entirely waiting on new block building - if we're mining the most important thing is the new work (and relay might work withou
...
💬 wonder75 commented on issue "LevelDB read failure: Corruption: block checksum mismatch":
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430050545)
> I haven't seen any data corruption since Sept 12, so at this point I can't really give you any information to duplicate the problem.
>
Hallo @apulsifer,
could you please provide information on how to fix this problem when it occurs? I suffer from frequent levelDB corruptions too and found no way to fix the problem other than redoing the Initial Block Download. After that it might work for a week or a month until the next corruption. Would really like to know a way to manuly fix this issu
...
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430050545)
> I haven't seen any data corruption since Sept 12, so at this point I can't really give you any information to duplicate the problem.
>
Hallo @apulsifer,
could you please provide information on how to fix this problem when it occurs? I suffer from frequent levelDB corruptions too and found no way to fix the problem other than redoing the Initial Block Download. After that it might work for a week or a month until the next corruption. Would really like to know a way to manuly fix this issu
...
💬 apulsifer commented on issue "LevelDB read failure: Corruption: block checksum mismatch":
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430072059)
I'm running on Amazon EC2 (Elastic Compute), and have multiple servers running bitcoind. The servers all have the bitcoin data directory stored on a dedicated EBS (Elastic Block Store) volume. If one fails, I have been snapshot'ing the volume on a working server and copying it over to the failed server. Time to repair is about 30 minutes. (Reindex also does not work for me, but I'm running pruned mode, and reindex is documented to not work in prune mode.)
The only thing I could suggest for yo
...
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430072059)
I'm running on Amazon EC2 (Elastic Compute), and have multiple servers running bitcoind. The servers all have the bitcoin data directory stored on a dedicated EBS (Elastic Block Store) volume. If one fails, I have been snapshot'ing the volume on a working server and copying it over to the failed server. Time to repair is about 30 minutes. (Reindex also does not work for me, but I'm running pruned mode, and reindex is documented to not work in prune mode.)
The only thing I could suggest for yo
...
💬 instagibbs commented on pull request "Package validation: accept packages of size 1":
(https://github.com/bitcoin/bitcoin/pull/31096#discussion_r1811269497)
done
(https://github.com/bitcoin/bitcoin/pull/31096#discussion_r1811269497)
done
💬 instagibbs commented on pull request "Package validation: accept packages of size 1":
(https://github.com/bitcoin/bitcoin/pull/31096#discussion_r1811270698)
hmm, it's a fair bit of code difference because we essentially need to get `BroadcastTransaction` grafted into place, then also return the proper `PackageMempoolAcceptResult` details vs the coarse details returned by it. My attempt required quite a bit of churn and additional testing required for a small feature like this, so I'm keeping as-is.
I'm also thinking now that gating the call via `package-not-child-with-unconfirmed-parents` check is the right thing to do. The further we stray from
...
(https://github.com/bitcoin/bitcoin/pull/31096#discussion_r1811270698)
hmm, it's a fair bit of code difference because we essentially need to get `BroadcastTransaction` grafted into place, then also return the proper `PackageMempoolAcceptResult` details vs the coarse details returned by it. My attempt required quite a bit of churn and additional testing required for a small feature like this, so I'm keeping as-is.
I'm also thinking now that gating the call via `package-not-child-with-unconfirmed-parents` check is the right thing to do. The further we stray from
...
💬 apulsifer commented on issue "LevelDB read failure: Corruption: block checksum mismatch":
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430081120)
Suggestion for bitcoind code maintainers-- As I've mentioned, I believe this data corruption happens but is not detected until a few days later when bitcoind attempts to read an older block (for whatever reason, idk). It would be very handy if there were some utility or bitcoind RPC command or command line option that read and checked every file on the disk, or all the files containing data from the prior X days or X blocks. That would help users ensure that they have a known good copy/backup of
...
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430081120)
Suggestion for bitcoind code maintainers-- As I've mentioned, I believe this data corruption happens but is not detected until a few days later when bitcoind attempts to read an older block (for whatever reason, idk). It would be very handy if there were some utility or bitcoind RPC command or command line option that read and checked every file on the disk, or all the files containing data from the prior X days or X blocks. That would help users ensure that they have a known good copy/backup of
...
💬 sipa commented on issue "LevelDB read failure: Corruption: block checksum mismatch":
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430085267)
`gettxoutsetinfo` should read through all the chainstate LevelDB files.
(https://github.com/bitcoin/bitcoin/issues/30159#issuecomment-2430085267)
`gettxoutsetinfo` should read through all the chainstate LevelDB files.