💬 davidgumberg commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183868035)
Thanks for this suggestion, I've implemented this approach in https://github.com/bitcoin/bitcoin/pull/32273/commits/55e60bd9c76ea74a108cd5c7275e26554b822b4c, and I've reused `migrate_and_get_rpc()` in the other tests added here (except [`test_failed_migration_cleanup_relative_path()`](https://github.com/bitcoin/bitcoin/blob/6b2e3fbf64b19b6065768ef85155f986859fcfb9/test/functional/wallet_migration.py#L1038) where it's not feasible)

💬 davidgumberg commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183873334)
As suggested by @furszy [above](https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2181204274) I moved this logic to `migrate_and_get_rpc()` in https://github.com/bitcoin/bitcoin/commit/55e60bd9c76ea74a108cd5c7275e26554b822b4c:

https://github.com/bitcoin/bitcoin/blob/6b2e3fbf64b19b6065768ef85155f986859fcfb9/test/functional/wallet_migration.py#L125-L147

And I've reused `migrate_and_get_rpc()` in the tests added in this PR.

💬 davidgumberg commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#issuecomment-3033891866)
> I don't think belt and suspenders checks are a great idea in general, but even when they are appropriate I think they need to be labeled with "// This condition is never expected to trigger" or similar comments to avoid confusing people reading the code and trying understand why the checks are there.

OK, in lieu of an assert/assume, I've gone to the simplified form suggested, and I've also used the lambda+ternary over if+else form as you suggested. I've also taken advantage of the fact that
...

💬 achow101 commented on pull request "test: fix an incorrect `feature_fee_estimation.py` subtest":
(https://github.com/bitcoin/bitcoin/pull/32463#issuecomment-3033949761)
ACK 9b75cfda4d62a0a3bde402503244dd57e1621a12

✅ achow101 closed an issue: "test: `feature_fee_estimation` failure after duplicate coinbase tx weight reservation fix [AssertionError: Estimated fee (0.00923427) out of range]"
(https://github.com/bitcoin/bitcoin/issues/32461)

🚀 achow101 merged a pull request: "test: fix an incorrect `feature_fee_estimation.py` subtest"
(https://github.com/bitcoin/bitcoin/pull/32463)

💬 furszy commented on pull request "fees: prevent redundant estimates flushes":
(https://github.com/bitcoin/bitcoin/pull/32748#discussion_r2183951988)
could write this in a single line instead.
```c++
if (WITH_LOCK(m_cs_fee_estimator, return !MaxUsableEstimate())) return;
```

Still, it would be nice to explain why `MaxUsableEstimate` returning 0 means "nothing to do". It is not immediately clear to me without checking the `MaxUsableEstimate` internals.

💬 davidgumberg commented on pull request "p2p: Relax BlockRequestAllowed to respond to advertised blocks":
(https://github.com/bitcoin/bitcoin/pull/32869#discussion_r2183963160)
```suggestion
assert {"height": 13, "hash": block.hash, "branchlen": 1, "status": "headers-only"} in self.nodes[1].getchaintips()
```

💬 davidgumberg commented on pull request "p2p: Relax BlockRequestAllowed to respond to advertised blocks":
(https://github.com/bitcoin/bitcoin/pull/32869#discussion_r2183963930)
nit:

```suggestion
node.bumpmocktime(200 * 60 * 2)
```

💬 achow101 commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183965661)
In 146b133697f07c8dda004d60087505b2e1ec14b9 "test: Migration of a wallet ending in `/`"

Could add a check for the expected backup path.

💬 achow101 commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183964782)
In e7fec273593d825e0e7dbec46a026aacc58c479f "test: Migration fail recovery w/ `../` in path"

I think `failed_watchonly` is the wrong wallet name. Probably best to do `f"{wallet_name}_watchonly"`

💬 achow101 commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183965792)
In 9fa5480fc4c46fa11345c45fbe4dd21c127e3e05 "test: Migration of a wallet ending in `../`"

Could add a check for the expected backup path.

💬 davidgumberg commented on pull request "p2p: Relax BlockRequestAllowed to respond to advertised blocks":
(https://github.com/bitcoin/bitcoin/pull/32869#issuecomment-3034004122)
Maybe a dumb question, but why not relax the other side here? I don't know what the new criteria would be, but in an ideal world, nodes should not be sending block announcements for blocks they have identified as invalid just to avoid upsetting their peers, so maybe the constraints set by `PeerManagerImpl::SendMessages()` are too rigid?

https://github.com/bitcoin/bitcoin/blob/e3f416dbf7633b2fb19c933e5508bd231cc7e9cf/src/net_processing.cpp#L5865-L5873

💬 davidgumberg commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183974514)
I add this check to `migrate_and_get_rpc()` in 55e60bd9c76:

https://github.com/bitcoin/bitcoin/blob/9fa5480fc4c46fa11345c45fbe4dd21c127e3e05/test/functional/wallet_migration.py#L138-L146

💬 davidgumberg commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183974714)
See above: https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183974514

💬 achow101 commented on pull request "p2p: protect addnode peers during IBD":
(https://github.com/bitcoin/bitcoin/pull/32051#issuecomment-3034010351)
> @achow101 re: slow preferred peer "causing congestion" -- wouldn't this only affect the single user who explicitly set the preferred peer?

Yes, however I have seen people post on various forums suggestions of specific nodes that they can addnode, and users may be addnode'ing such nodes without realizing that there are potential adverse consequences of doing so. Generally, I would prefer to have things that change how we treat nodes be limited to the whitelist options rather than addnode als
...

💬 davidgumberg commented on pull request "wallet: Fix relative path backup during migration.":
(https://github.com/bitcoin/bitcoin/pull/32273#discussion_r2183977261)
Thanks for catching this, I've applied your fix in f67d6fe811

💬 yuvicc commented on pull request "test: Fix reorg patterns in tests to use proper fork-based approach":
(https://github.com/bitcoin/bitcoin/pull/32587#issuecomment-3034576513)
Addressed @instagibbs [suggestion](https://github.com/bitcoin/bitcoin/pull/32587#discussion_r2180262495) to use time-based timelock instead of n block method.

💬 Sjors commented on pull request "rpc: Handle -named argument parsing where '=' character is used":
(https://github.com/bitcoin/bitcoin/pull/32821#issuecomment-3034701850)
From CI:

```
assert not "non_witness_utxo" in mining_node.decodepsbt(signed_psbt["psbt"])["inputs"][0]
```

I didn't check the code, but that's usually a cryptic way of saying that signing the psbt failed.

💬 b-l-u-e commented on pull request "net: Fix Discover() not running when using -bind=0.0.0.0:port":
(https://github.com/bitcoin/bitcoin/pull/32757#issuecomment-3034729829)
```
@vasild Thanks for the suggested improvements! I have a question about the `self.extra_args` configuration.

In your patch, node 2 only has `-whitebind=0.0.0.0` without any onion bind:
```python
['-discover', f'-whitebind=0.0.0.0:{self.bind_ports[2]}'], # Explicit -whitebind=0.0.0.0
```

However, when I tried this configuration, I encountered port conflicts because the test framework automatically adds an onion bind to node 2, which conflicts with the port already used by node 1.

...