💬 fjahr commented on issue "bitcoin core crashes when too many rpc calls are made":
(https://github.com/bitcoin/bitcoin/issues/11368#issuecomment-1560253321)
> 2.2.1-alpha is [now out](https://github.com/libevent/libevent/releases/tag/release-2.2.1-alpha) 👀

I have opened a draft PR #27731 that uses the proposed solution with the max connection feature that was added there. But currently I still have a hard time reproducing the issue in the same way I was able to back in 2020 described above: https://github.com/bitcoin/bitcoin/issues/11368#issuecomment-638770701 I have a much faster machine now but there may also be other reasons.

I also tried
...

⚠️ fjahr opened an issue: "Creating too many wallets exhausts file descriptor limit and leads to crash"
(https://github.com/bitcoin/bitcoin/issues/27732)
### Is there an existing issue for this?

- [X] I have searched the existing issues

### Current behaviour

This behavior was first noted by @EthanHeilman here: https://github.com/bitcoin/bitcoin/issues/11368#issuecomment-751574945 but it was conflated with the RPC issue there while I now think this is a separate issue.

When a lot of wallets are created in quick succession, sqlite will exhaust all the available file descriptors and the node crashes the next time it tries to create file for un
...

💬 fjahr commented on issue "bitcoin core crashes when too many rpc calls are made":
(https://github.com/bitcoin/bitcoin/issues/11368#issuecomment-1560260664)
See #27732 for the wallet issue, I that makes sense to keep these things separate.

📝 theStack opened a pull request: "test: refactor: introduce `generate_keypair` helper with WIF support"
(https://github.com/bitcoin/bitcoin/pull/27733)
In functional tests it is a quite common scenario to generate fresh elliptic curve keypairs, which is currently a bit cumbersome as it involves multiple steps, e.g.:

privkey = ECKey()
privkey.generate()
privkey_wif = bytes_to_wif(privkey.get_bytes())
pubkey = privkey.get_pubkey().get_bytes()

Simplify this by providing a new `generate_keypair` helper function that returns the private key either as `ECKey` object or as WIF-string (depending on the boolean `wif` parameter)
...

⚠️ Sokhanetaze80 opened an issue: "Is there any documentation on PROTOCOL_VERSION I can refer to please? How is this decided and how does it cope with branches?"
(https://github.com/bitcoin/bitcoin/issues/27734)
Is there any documentation on PROTOCOL_VERSION I can refer to please? How is this decided and how does it cope with branches?

_Originally posted by @rebroad in https://github.com/bitcoin/bitcoin/pull/932#discussion_r580883_

✅ Sokhanetaze80 closed an issue: "Is there any documentation on PROTOCOL_VERSION I can refer to please? How is this decided and how does it cope with branches?"
(https://github.com/bitcoin/bitcoin/issues/27734)

:lock: achow101 locked an issue: "."
(https://github.com/bitcoin/bitcoin/issues/27734)

💬 ishaanam commented on pull request "wallet: improve IBD sync time by skipping block scanning prior birth time":
(https://github.com/bitcoin/bitcoin/pull/27469#issuecomment-1560345015)
cr-ACK 76396ca376188631ba46bd47b134881efcc6f755

💬 fanquake commented on pull request "init: Improve file descriptor limit handling":
(https://github.com/bitcoin/bitcoin/pull/27730#issuecomment-1560455205)
Duplicate of #27539?

🤔 MarcoFalke reviewed a pull request: "wallet: improve IBD sync time by skipping block scanning prior birth time"
(https://github.com/bitcoin/bitcoin/pull/27469#pullrequestreview-1440982489)
left some comments/nits/questions. Feel free to ignore.

💬 MarcoFalke commented on pull request "wallet: improve IBD sync time by skipping block scanning prior birth time":
(https://github.com/bitcoin/bitcoin/pull/27469#discussion_r1203481523)
first commit: This would lead to issues in the rare case that more than 2 hours worth of blocks are produced, as they end up being in the future.

Is there a reason for the first commit?

💬 MarcoFalke commented on pull request "wallet: improve IBD sync time by skipping block scanning prior birth time":
(https://github.com/bitcoin/bitcoin/pull/27469#discussion_r1203482881)
first commit: Any reason this is changed?

💬 MarcoFalke commented on pull request "wallet: improve IBD sync time by skipping block scanning prior birth time":
(https://github.com/bitcoin/bitcoin/pull/27469#discussion_r1203482268)
Also, it is a source of non-determinism, but I guess this is fine.

📝 MarcoFalke opened a pull request: "test: Disable legacy wallet for mempool_packages.py"
(https://github.com/bitcoin/bitcoin/pull/27735)
This fixes a bug.

On master:

```
$ ./test/functional/mempool_packages.py --legacy-wallet
File "./test/functional/mempool_packages.py", line 52, in run_test
self.nodes[0].importaddress(self.wallet.get_address())
test_framework.authproxy.JSONRPCException: Bech32m addresses cannot be imported into legacy wallets (-5)
```

On this pull:

```
$ ./test/functional/mempool_packages.py --legacy-wallet
usage: mempool_packages.py [options]
mempool_packages.py: error: unrecognize
...

💬 MarcoFalke commented on pull request "test: Disable legacy wallet for mempool_packages.py":
(https://github.com/bitcoin/bitcoin/pull/27735#issuecomment-1560544086)
(An alternative to test this would be to compile with bdb only)

💬 MarcoFalke commented on issue "Intermittent failures in interface_usdt_mempool.py":
(https://github.com/bitcoin/bitcoin/issues/27380#issuecomment-1560583591)
https://cirrus-ci.com/task/6382990889058304?logs=ci#L3266

💬 fjahr commented on pull request "init: Improve file descriptor limit handling":
(https://github.com/bitcoin/bitcoin/pull/27730#issuecomment-1560647386)
> Duplicate of #27539?

Ugh, mostly yeah.

✅ fjahr closed a pull request: "init: Improve file descriptor limit handling"
(https://github.com/bitcoin/bitcoin/pull/27730)

🤔 fjahr reviewed a pull request: "init: Fixes for file descriptor accounting"
(https://github.com/bitcoin/bitcoin/pull/27539#pullrequestreview-1441209058)
Concept ACK

I think it would be very helpful to add comments to explain what is going on. See my accidentally duplicate PR for some suggestions: #27730.

💬 fjahr commented on pull request "init: Fixes for file descriptor accounting":
(https://github.com/bitcoin/bitcoin/pull/27539#discussion_r1203666492)
Needs an explanation of why it is ok to take `FD_SETSIZE` completely out of the equation in the commit description.