5fd4836019 init: change shutdown order of load block thread and scheduler (Martin Zumsande)
Pull request description:
This avoids situations during a reindex, in which the shutdown doesn't finish since `LimitValidationInterfaceQueue()` is called by the load block thread when the scheduler is already stopped, in which case it would block indefinitely. This can lead to intermittent failures in `feature_reindex.py` (#30424), which I could locally reproduce with
```diff
diff --git a/src/validation.cpp b/src/validation.cpp
index 74f0e4975c..be1706fdaf 100644
--- a/src/validation.cpp
+++ b/src/validation.cpp
@@ -3446,6 +3446,7 @@ static void LimitValidationInterfaceQueue(ValidationSignals& signals) LOCKS_EXCL
AssertLockNotHeld(cs_main);
if (signals.CallbacksPending() > 10) {
+ std::this_thread::sleep_for(std::chrono::milliseconds(50));
signals.SyncWithValidationInterfaceQueue();
}
}
```
It has also been reported by users running `reindex-chainstate` (#23234).
I thought for a bit about potential downsides of changing this order, but couldn't find any.
Fixes#30424Fixes#23234
ACKs for top commit:
maflcko:
review ACK 5fd4836019
hebasto:
re-ACK 5fd4836019.
tdb3:
ACK 5fd4836019
BrandonOdiwuor:
Code Review ACK 5fd4836019
Tree-SHA512: 3b8894e99551c5d4392b55eaa718eee05841a7287aeef2978699e1d633d5234399fa2f5a3e71eac1508d97845906bd33e0e63e5351855139e7be04c421359b36
7cb9367157 rpc: keep .cookie if it was not generated (Roman Zeyde)
Pull request description:
Otherwise, starting bitcoind twice may cause the `.cookie` file generated by the first instance to be deleted by the second instance shutdown (after failing to obtain a lock).
ACKs for top commit:
willcl-ark:
re-ACK 7cb9367157
achow101:
ACK 7cb9367157
kristapsk:
re-ACK 7cb9367157
stickies-v:
ACK 7cb9367157
Tree-SHA512: 0960dbc457975b0e0535f3d814824a879d7f85c9f1191537415b3fc253429a316a8e4badde56c8bc139778f132392983cec5fbe03891fb15ff61d3bc3f6e681b
This test would cause a crash in bitcoind (see #26274) if the fix given in the
previous commit was not applied.
Github-Pull: #26275
Rebased-From: 9153ff3e27
The external input test with specifying input weight would make a
pessimistic estimate of the input weight. However this would result in a
test failure as it is sometimes too pessimistic when an ECDSA signature
ends up being smaller than usual. To correct this, we can calculate the
input weight more accurately.
Github-Pull: #24454
Rebased-From: 8a04a386f7
Thanks to Jon Griffiths for picking up that the amount and nonce discount
should be multiplied by the witness scaling factor as they form part of
the base transaction.
