Skip to content

fix: SocialRegistry and sql cache#858

Merged
andyoknen merged 21 commits into0.22from
fix/socialregistry
Apr 23, 2025
Merged

fix: SocialRegistry and sql cache#858
andyoknen merged 21 commits into0.22from
fix/socialregistry

Conversation

@andyoknen
Copy link
Collaborator

Calling the SocialRegistry table rebuild function from the main thread causes the logic to fail if the height of the ChainActive internal object does not match the height of the sqlite database.

This leads to the creation of incorrect blocks.

One of the reasons for node failures may be the obscenely high memory usage of the sqlite database. The reason for the high memory consumption is the incorrect use of the cache_size pragma. This pragma is set for each database connection (which can be quite a lot when the public interface is enabled - api=1).

These changes change the behavior of using the pragma so that the specified cache size value is not multiplied by the number of database connections, but distributed among them.

@andyoknen andyoknen changed the title fix: rebuilding the SocialRegistry table fix: SocialRegistry and sql cache Apr 18, 2025
@andyoknen andyoknen merged commit 42a5a61 into 0.22 Apr 23, 2025
5 checks passed
@andyoknen andyoknen deleted the fix/socialregistry branch April 23, 2025 09:59
@andyoknen andyoknen linked an issue Apr 29, 2025 that may be closed by this pull request
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Getting not in memory pool stake wins. v0.22.14 and v0.22.15 - possible code regression. V22.11,v22.12 and v22.13 possible consensus issues

1 participant

Comments