-
### What was wrong
The current state network spec has the stored payload for trie nodes be un-anchored. The OFFER payload contains a proof and reference to a block hash, canonically anchoring the …
-
```
18:03:43.917 [xtdb.compactor-1] ERROR xtdb.compactor - Error running compaction job.
java.lang.IllegalArgumentException: Could not load buffers for field foo: Union(Dense, [0, 1, 2, 3]) not nu…
-
I currently have some logic that I hacked in to dump all tries before and after txns. It's been useful enough times that I think we should probably get this back into `main` that is enabled with a pro…
-
In no particular order...
- [x] `Cargo.lock`
We definitely want this for reproducibility, especially when we're not pinned to an `alloy` revision.
No-one wants to bisect versions.
See also…
-
We store symbol names for scip documents in a table called `codeintel_scip_symbol_names` and in order to save space we encode them as a prefix-trie inside the database.
When storing larger scip indic…
-
> Build failure seen for ["Test: Language Modeling Layer (Common)"](https://build.palaso.org/buildConfiguration/Keyman_Common_LMLayer_TestPullRequests/473674):
```
15:56:57
LMLayer using the…
-
**Description**
On one of my nodes I have noticed that right after State Sync finished Recover mode had fired and recovered one missing trie node.
Good that mechanism worked and recovered me a node …
-
# Motivation
- #93
# Previous work
- #206
- #198
# Background
## Terminology
- Type 1 🥇 is full compatibility [^1] (there should be some more words here, but I don't know what to write)
…
-
*question*: It seems a shame that one must repeat most of the fields between `parse_options` and `parse_options_view`. Shall we open an issue to discuss whether these should be shared with something l…
vyasr updated
1 month ago
-
Our trie update logic is currently tied directly to a database write transaction. This means that updating the state trie for a given block can only occur in the same transaction that is inserting the…