Comments (6)
avantgardnerio
commented on June 20, 2024
1
Thanks for the clarification!
from openraft.
github-actions
commented on June 20, 2024
👋 Thanks for opening this issue!
Get help or engage by:
/help: to print help messages./assignme: to assign this issue to you.
from openraft.
drmingdrmer
commented on June 20, 2024
0.7.3 does not implement persistent learner storage in Membership.
If you need this feature, use the main branch instead. :(
from openraft.
avantgardnerio
commented on June 20, 2024
Thanks for your response. I switched to the master branch, and I can confirm it works there!
I did want to ask about some possibly unrelated behavior: on the 0.7.x branch I saw Blank messages were sent as heartbeats, but I didn't think they were appended to the log or applied to the state machine. In master, I see lots of Blank traffic:
Node 1: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }, payload: Blank }
Node 2: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }, payload: Blank }
Node 3: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }, payload: Blank }
Node 5: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }, payload: Blank }
Node 4: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }
Node 4: sending AppendEntriesRequest to 1 AppendEntriesRequest { vote: Vote { term: 1, node_id: 4, committed: true }, prev_log_id: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }), entries: [Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }], leader_commit: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }) }
Node 4: sending AppendEntriesRequest to 2 AppendEntriesRequest { vote: Vote { term: 1, node_id: 4, committed: true }, prev_log_id: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }), entries: [Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }], leader_commit: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }) }
Node 4: sending AppendEntriesRequest to 3 AppendEntriesRequest { vote: Vote { term: 1, node_id: 4, committed: true }, prev_log_id: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }), entries: [Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }], leader_commit: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }) }
Node 4: sending AppendEntriesRequest to 5 AppendEntriesRequest { vote: Vote { term: 1, node_id: 4, committed: true }, prev_log_id: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }), entries: [Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }], leader_commit: Some(LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 69 }) }
Node 1: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }
Node 2: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }
Node 3: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }
Node 5: append_to_log log_idx=LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 } entry=Entry { log_id: LogId { leader_id: LeaderId { term: 1, node_id: 4 }, index: 70 }, payload: Blank }
Is that normal? It seems like heartbeats in the raft log would cause quite a bit of write amplification.
from openraft.
avantgardnerio
commented on June 20, 2024
Follow up confirmation, this filter used to work for heartbeats:
impl RaftNetwork<ColDbRaftConfig> for GrpcNetworkConnection {
async fn send_append_entries(
&mut self,
req: AppendEntriesRequest<ColDbRaftConfig>,
) -> Result<AppendEntriesResponse<Nid>, RPCError<Nid, UuidNode, AppendEntriesError<Nid>>> {
let is_heartbeat = req.entries.is_empty();
if !is_heartbeat {
println!("Node {}: sending AppendEntriesRequest to {} {:?}", self.owner.node_id, self.target, req);
}
from openraft.
drmingdrmer
commented on June 20, 2024
Yes, the heartbeat message is changed:
In main branch, I was trying to implement a heartbeat with a message with a blank log, instead of just a blank message, i.e., every heartbeat will commit a read log entry.
This simplifies the logic when granting a candidate's vote request. But it introduces unnecessary IO(to flush blank log to disk).
See:
from openraft.
Related Issues (20)
- Feature: Monoio Runtime HOT 3
- Add sync primitives to `AsyncRuntime` trait HOT 1
- Add `AsyncRuntime::oneshot` HOT 2
- RaftMsg::ExternalRequest: should not rely on RaftLogStorage and RaftNetworkFactory HOT 3
- Add random number generation to `Runtime` HOT 5
- Ditch `async_trait` in favor of the official `async_fn_in_trait` feature HOT 6
- Refactor: Move `client_resp_channels` from `LeaderData` to new `tx_...`-field in `RaftCore` HOT 2
- Linearizable read with `ReadIndex` HOT 1
- about single node HOT 6
- Cluster Bootstrap and Node Initialization Sequence HOT 4
- Tracking issue for examples update HOT 1
- Update example `raft-kv-memstore` to use `storage-v2` HOT 2
- Update example `raft-kv-rocksstore` to use `storage-v2` HOT 1
- Move examples into the workspace HOT 1
- Install Snapshot v1 api HOT 3
- Observe state changes in a Raft node HOT 6
- Split metrics into data metrics and server metrics HOT 5
- About automatic remove HOT 3
- Backing up the WAL HOT 6
- Raft Core Panicking HOT 10
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from openraft.