Skip to content

Conversation

jkrvivian
Copy link
Contributor

Description of change

Port MystenLabs/sui@42dbb62

From Sui's commit description:
For fullnodes that do not prune the perpetual DB, especially
transactions and effects cfs, they can run into write stalls
occasionally that makes the fullnode non operational. Since fullnodes
have to accept all writes from checkpoints, throttling writes do not
seem to make much sense.
Write stalling on validators is left enabled.

Links to any relevant issues

Part of #3990

Type of change

  • Enhancement

Change checklist

Tick the boxes that are relevant to your changes, and delete any items that are not.

  • I have followed the contribution guidelines for this project
  • I have performed a self-review of my own code
  • I have commented my code, particularly in hard-to-understand areas
  • I have made corresponding changes to the documentation
  • I have added tests that prove my fix is effective or that my feature works
  • I have checked that new and existing unit tests pass locally with my changes

@jkrvivian jkrvivian self-assigned this May 8, 2025
@jkrvivian jkrvivian requested review from a team as code owners May 8, 2025 13:27
@jkrvivian jkrvivian added the node Issues related to the Core Node team label May 8, 2025
Copy link

vercel bot commented May 8, 2025

The latest updates on your projects. Learn more about Vercel for Git ↗︎

4 Skipped Deployments
Name Status Preview Comments Updated (UTC)
apps-backend ⬜️ Ignored (Inspect) Visit Preview May 20, 2025 1:04pm
apps-ui-kit ⬜️ Ignored (Inspect) Visit Preview 💬 Add feedback May 20, 2025 1:04pm
rebased-explorer ⬜️ Ignored (Inspect) Visit Preview May 20, 2025 1:04pm
wallet-dashboard ⬜️ Ignored (Inspect) Visit Preview 💬 Add feedback May 20, 2025 1:04pm

@jkrvivian jkrvivian changed the base branch from develop to core-node/upstream-changes/mainnet-1.32.2-1.33.3 May 8, 2025 13:28
objects_table_config(db_options.clone()),
),
(
"indirect_move_objects".to_string(),
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Isn't the entry for live_owned_object_markers_table_config missing maybe?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I added it now. I'm pretty sure it was missing.

@bingyanglin bingyanglin force-pushed the core-node/upstream-changes/mainnet-1.32.2-1.33.3 branch from 312b9c5 to d3c32a2 Compare May 20, 2025 02:40
@alexsporn alexsporn changed the title upstream(v1.32-v1.33): disable write stall on fullnodes perpetual db feat: disable write stall on fullnodes perpetual db May 20, 2025
@muXxer muXxer force-pushed the node/upstream-v1.32-1.33/42dbb branch from 270ec02 to 15668a7 Compare May 20, 2025 12:53
@muXxer muXxer merged commit b6a68ab into core-node/upstream-changes/mainnet-1.32.2-1.33.3 May 21, 2025
33 of 35 checks passed
@muXxer muXxer deleted the node/upstream-v1.32-1.33/42dbb branch May 21, 2025 14:23
muXxer added a commit that referenced this pull request May 21, 2025
# Description of change

Port
MystenLabs/sui@42dbb62

From Sui's commit description:
For fullnodes that do not prune the `perpetual` DB, especially
`transactions` and `effects` cfs, they can run into write stalls
occasionally that makes the fullnode non operational. Since fullnodes
have to accept all writes from checkpoints, throttling writes do not
seem to make much sense.
Write stalling on validators is left enabled.

## Links to any relevant issues

Part of #3990 

## Type of change
- Enhancement 

---------

Co-authored-by: muXxer <[email protected]>
muXxer added a commit that referenced this pull request May 26, 2025
# Description of change

Port
MystenLabs/sui@42dbb62

From Sui's commit description:
For fullnodes that do not prune the `perpetual` DB, especially
`transactions` and `effects` cfs, they can run into write stalls
occasionally that makes the fullnode non operational. Since fullnodes
have to accept all writes from checkpoints, throttling writes do not
seem to make much sense.
Write stalling on validators is left enabled.

## Links to any relevant issues

Part of #3990 

## Type of change
- Enhancement 

---------

Co-authored-by: muXxer <[email protected]>
muXxer added a commit that referenced this pull request May 26, 2025
# Description of change

Port
MystenLabs/sui@42dbb62

From Sui's commit description:
For fullnodes that do not prune the `perpetual` DB, especially
`transactions` and `effects` cfs, they can run into write stalls
occasionally that makes the fullnode non operational. Since fullnodes
have to accept all writes from checkpoints, throttling writes do not
seem to make much sense.
Write stalling on validators is left enabled.

## Links to any relevant issues

Part of #3990 

## Type of change
- Enhancement 

---------

Co-authored-by: muXxer <[email protected]>
muXxer added a commit that referenced this pull request May 27, 2025
# Description of change

Port
MystenLabs/sui@42dbb62

From Sui's commit description:
For fullnodes that do not prune the `perpetual` DB, especially
`transactions` and `effects` cfs, they can run into write stalls
occasionally that makes the fullnode non operational. Since fullnodes
have to accept all writes from checkpoints, throttling writes do not
seem to make much sense.
Write stalling on validators is left enabled.

## Links to any relevant issues

Part of #3990 

## Type of change
- Enhancement 

---------

Co-authored-by: muXxer <[email protected]>
muXxer added a commit that referenced this pull request Jun 2, 2025
# Description of change

Port
MystenLabs/sui@42dbb62

From Sui's commit description:
For fullnodes that do not prune the `perpetual` DB, especially
`transactions` and `effects` cfs, they can run into write stalls
occasionally that makes the fullnode non operational. Since fullnodes
have to accept all writes from checkpoints, throttling writes do not
seem to make much sense.
Write stalling on validators is left enabled.

## Links to any relevant issues

Part of #3990 

## Type of change
- Enhancement 

---------

Co-authored-by: muXxer <[email protected]>
muXxer added a commit that referenced this pull request Jun 2, 2025
…ainnet-1.32.2 to mainnet-1.33.3 (#6952)

# Description of change

This PR contains a batch of node-team related upstream changes. The
changes have been individually reviewed before squash-merging. None of
them modify protocol parameters, so a new protocol version is not
necessary.

This PR should be rebased on top of develop without squashing as it
already contains squashed PRs.

## Links to any relevant issues

List of included PRs:

* #6179 
* #6181 
* #6182 
* #6183 
* #6334 
* #6184 
* #6623 
* #6390
* #6414 
* #6440
* #6786 
* #6791 
* #6817 

List of included upstream commits:

*
MystenLabs/sui@3a68b7d
*
MystenLabs/sui@07f8b22
*
MystenLabs/sui@3467759
*
MystenLabs/sui@a22b71b
*
MystenLabs/sui@070a2c3
*
MystenLabs/sui@6a0adb5
*
MystenLabs/sui@a8e969a
*
MystenLabs/sui@82bf44e
*
MystenLabs/sui@fa7419b
*
MystenLabs/sui@0e78656
*
MystenLabs/sui@e1abe09
*
MystenLabs/sui@262390b
*
MystenLabs/sui@42dbb62

## Type of change

- Enhancement (a non-breaking change which adds functionality)

## How the change has been tested

- [x] Basic tests (linting, compilation, formatting, unit/integration
tests)
- [ ] Patch-specific tests (correctness, functionality coverage)

### Release Notes

- [ ] Protocol:
- [x] Nodes (Validators and Full nodes):
  * Migrate users of iota_metrics Histogram to prometheus Histogram
  * Rename metric names:
    * last_created_checkpoint_age_ms => last_created_checkpoint_age
    * last_certified_checkpoint_age_ms => last_certified_checkpoint_age
    * checkpoint_creation_latency_ms => checkpoint_creation_latency
    * checkpoint_contents_age_ms => checkpoint_contents_age
    * last_executed_checkpoint_age_ms => last_executed_checkpoint_age
    * checkpoint_summary_age_ms => checkpoint_summary_age
    * checkpoint_prepare_latency_us => checkpoint_prepare_latency
    * checkpoint_exec_latency_us => checkpoint_exec_latency
* handle_soft_bundle_certificates_count =>
validator_service_handle_soft_bundle_certificates_count
* handle_soft_bundle_certificates_size_bytes =>
validator_service_handle_soft_bundle_certificates_size_bytes
* Add metric for active time of monitored futures
(monitored_future_active_duration_ns)
* Add metric for number of rocksdb files and level 0 files
(rocksdb_total_num_files, rocksdb_num_level0_files)
- [ ] Indexer:
- [ ] JSON-RPC:
- [ ] GraphQL:
- [ ] CLI:
- [ ] Rust SDK:
- [ ] REST API:
filipdulic pushed a commit that referenced this pull request Jul 16, 2025
# Description of change

Port
MystenLabs/sui@42dbb62

From Sui's commit description:
For fullnodes that do not prune the `perpetual` DB, especially
`transactions` and `effects` cfs, they can run into write stalls
occasionally that makes the fullnode non operational. Since fullnodes
have to accept all writes from checkpoints, throttling writes do not
seem to make much sense.
Write stalling on validators is left enabled.

## Links to any relevant issues

Part of #3990 

## Type of change
- Enhancement 

---------

Co-authored-by: muXxer <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
core-protocol node Issues related to the Core Node team
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants