feat(nodectl): TONCore per-slot details in 'config pool ls' + REST add by Keshoid · Pull Request #95 · RSquad/ton-rust-node

Keshoid · 2026-04-15T23:57:24Z

config pool ls and GET /v1/pools now render TONCore pools as a per-slot table with state (active / uninit / not deployed / error), balance, and on-chain pool params (validator share, max nominators, min stakes, nominators count, staked amount, pool state, last election id). Slot fetches run in parallel; per-slot RPC failures are encoded into the slot DTO instead of failing the whole response. SNP shape unchanged.

Also adds POST /v1/pools/core so config pool add core goes through REST like the rest of v0.4.0 entity CRUD instead of writing the local config file. Validates validator_share (0..=10000 bp), max_nominators > 0, sibling deploy params require validator_share, rejects SNP-name collisions and already-configured slots.

Closes SMA-78

…est api

linear · 2026-04-15T23:57:28Z

SMA-78 config pool ls: TONCore pools display is incomplete

Copilot

Pull request overview

Enhances node-control pool management by exposing TONCore pools as per-slot entries (even/odd) with on-chain details in both config pool ls and GET /v1/pools, and adds a dedicated REST endpoint to create TONCore pool slots (POST /v1/pools/core) so CLI no longer edits local config directly.

Changes:

Update GET /v1/pools to return TONCore pools as slots[] with per-slot state/balance and on-chain pool parameters fetched in parallel.
Add POST /v1/pools/core REST handler with validation to add/create TONCore pool slots.
Update nodectl config pool add core to call the new REST endpoint and update config pool ls output formatting; add focused HTTP handler unit tests.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
`src/node-control/service/src/http/mod.rs`	Registers new HTTP handler test module.
`src/node-control/service/src/http/http_server_task.rs`	Wires new `/v1/pools/core` route and adds DTOs/requests to OpenAPI schema.
`src/node-control/service/src/http/config_handlers_tests.rs`	Adds unit tests covering TONCore per-slot listing behavior and `/v1/pools/core` validation/mutation paths.
`src/node-control/service/src/http/config_handlers.rs`	Implements TONCore per-slot DTOs, parallel slot fetching for `/v1/pools`, and the `/v1/pools/core` handler.
`src/node-control/commands/src/commands/nodectl/config_pool_cmd.rs`	Switches TONCore “add” to REST, and renders pool listing as separate SNP and TONCore tables with per-slot rows.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-16T00:01:18Z

+                    if let Ok(slot) = joined {
+                        slots.push(slot);


JoinSet::join_next() errors are currently ignored (if let Ok(slot) = joined { ... }). If a task panics/cancels, the corresponding slot silently disappears from the response, breaking the “one entry per configured slot” invariant. Handle Err(join_error) by pushing a slot DTO with state="error" (and slot id/address if known) or by converting the join error into an explicit handler error.

Suggested change

if let Ok(slot) = joined {

slots.push(slot);

match joined {

Ok(slot) => slots.push(slot),

Err(join_error) => return Err(join_error.into()),

That joined is the JoinHandle result — it only errors if the spawned task panicked (a Tokio join error), not if RPC failed.

Copilot · 2026-04-16T00:01:18Z

+) -> TonCorePoolSlotDto {
+    let slot_name = if slot_idx == 0 { "even" } else { "odd" };
+    let address = match &cached {
+        Some(w) => w.address().await.ok(),


Address resolution prioritizes the cached wrapper, but if w.address().await fails it does not fall back to the config address (it just returns None and the slot becomes not deployed). Consider attempting to parse/use config_address when the cached wrapper can’t provide an address, so transient cache/wrapper errors don’t misclassify a deployed slot.

Suggested change

Some(w) => w.address().await.ok(),

Some(w) => w

.address()

.await

.ok()

.or_else(|| config_address.as_deref().and_then(|a| MsgAddressInt::from_str(a).ok())),

Copilot · 2026-04-16T00:01:19Z

+        if let Ok(d) = wrapper.get_pool_data().await {
+            dto.validator_share = Some(d.pool_config.validator_reward_share);
+            dto.max_nominators = Some(d.pool_config.max_nominators_count);
+            dto.min_validator_stake = Some(d.pool_config.min_validator_stake);
+            dto.min_nominator_stake = Some(d.pool_config.nominator_stake_threshold);
+            dto.nominators_count = Some(d.nominators_count);
+            dto.stake_amount_sent = Some(d.stake_amount_sent);
+            dto.validator_amount = Some(d.validator_amount);
+            dto.pool_state = Some(d.state);
+            dto.last_election_id = Some(d.stake_at);


If get_pool_data() fails for an active slot, the DTO is returned with state="active" but with all on-chain parameter fields left empty, which is indistinguishable from “pool data unavailable”. Since the PR goal is to encode per-slot RPC failures into the slot DTO, consider setting state="error" (or adding an explicit error field) when get_pool_data() errors so clients can reliably detect partial failures.

Suggested change

if let Ok(d) = wrapper.get_pool_data().await {

dto.validator_share = Some(d.pool_config.validator_reward_share);

dto.max_nominators = Some(d.pool_config.max_nominators_count);

dto.min_validator_stake = Some(d.pool_config.min_validator_stake);

dto.min_nominator_stake = Some(d.pool_config.nominator_stake_threshold);

dto.nominators_count = Some(d.nominators_count);

dto.stake_amount_sent = Some(d.stake_amount_sent);

dto.validator_amount = Some(d.validator_amount);

dto.pool_state = Some(d.state);

dto.last_election_id = Some(d.stake_at);

match wrapper.get_pool_data().await {

Ok(d) => {

dto.validator_share = Some(d.pool_config.validator_reward_share);

dto.max_nominators = Some(d.pool_config.max_nominators_count);

dto.min_validator_stake = Some(d.pool_config.min_validator_stake);

dto.min_nominator_stake = Some(d.pool_config.nominator_stake_threshold);

dto.nominators_count = Some(d.nominators_count);

dto.stake_amount_sent = Some(d.stake_amount_sent);

dto.validator_amount = Some(d.validator_amount);

dto.pool_state = Some(d.state);

dto.last_election_id = Some(d.stake_at);

}

Err(_) => {

dto.state = "error".to_string();

}

added log on error, but logic is the same.

Copilot · 2026-04-16T00:01:19Z

+            match cfg.pools.get_mut(&name) {
+                Some(PoolConfig::TONCore { pools }) => {
+                    pools[slot_idx] = Some(slot_cfg);
+                }
+                Some(PoolConfig::SNP { .. }) => {
+                    // Pre-validated above; race-narrow window only — leave
+                    // the existing entry untouched.
+                }


Inside update_and_save, the Some(PoolConfig::SNP { .. }) branch is intentionally a no-op (for a “race-narrow window”), but the handler still returns 200 OK even though it didn’t apply the requested change. To avoid misleading success responses, capture a flag inside the closure (e.g., Arc<AtomicBool>) to confirm the mutation happened and return a 409/400 if it didn’t, or re-check the config after the update and error if the slot wasn’t set.

mrnkslv · 2026-04-16T00:12:31Z

+            name: &self.name,
+            slot: slot_name,
+            address: address.as_deref(),
+            validator_share: self.validator_share,


I would suggest adding validation to the handler for max_nominators (more than 1 and less than 40) and validator_share (no more than 10000)

This check is implemented on server-side, but I also added in cli.

mrnkslv · 2026-04-16T00:34:43Z

+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub min_nominator_stake: Option<u64>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub nominators_count: Option<u32>,


mb not no, but not all of params have the same types as in contract (for example, nominators count is u16 in contract,state is u8 and so on) . It comes from the first wrapper version, but we don't have any problems because of it. Not urgent

mrnkslv · 2026-04-16T00:36:43Z

    pub name: String,
+    /// `"SNP"` or `"Core"` — selects which of the field groups below applies.
    pub kind: String,
+    /// SNP only: on-chain balance in nanotons. Always `None` for TONCore.


but toncore has a balance too

mb add a little bit more information to avoid misunderstanding?

added a comment.

mrnkslv · 2026-04-16T00:43:31Z

+            dto.stake_amount_sent = Some(d.stake_amount_sent);
+            dto.validator_amount = Some(d.validator_amount);
+            dto.pool_state = Some(d.state);
+            dto.last_election_id = Some(d.stake_at);


looks like a bug here

Why? stake_at in the pool contract storage is the election ID — it records which election round the pool staked at.
The DTO doc comment makes the mapping explicit:
/// Last election id the pool staked at (stake_at in pool storage).

mrnkslv · 2026-04-16T00:45:38Z

+    path = "/v1/pools/core",
+    request_body = PoolAddCoreRequest,
+    responses(
+        (status = 200, description = "TONCore pool slot added", body = EntityRefResponse),


maybe 201? 200 - success, 201 usually means "successfully created"

mrnkslv · 2026-04-16T00:48:02Z

+    }
+    if let Some(mn) = req.max_nominators {
+        if mn == 0 {
+            return Err(AppError::bad_request("max_nominators must be > 0"));


and less than 40

…-ls-toncore-pools-display-is-incomplete

Keshoid added 2 commits April 16, 2026 01:56

feat(service,commands): impl config ls for ton core pools through r…

2b2ccea

…est api

feat(service,commands): add TONCore pool slot via REST

815d1ee

Copilot AI review requested due to automatic review settings April 15, 2026 23:57

Copilot started reviewing on behalf of Keshoid April 15, 2026 23:57 View session

Keshoid requested review from ITBear and mrnkslv April 15, 2026 23:58

Copilot AI reviewed Apr 16, 2026

View reviewed changes

mrnkslv reviewed Apr 16, 2026

View reviewed changes

Keshoid added 2 commits April 16, 2026 09:35

fix: after review

5a3dcbe

Merge branch 'release/nodectl/v0.4.0' into feature/sma-78-config-pool…

5fecb23

…-ls-toncore-pools-display-is-incomplete

ITBear approved these changes Apr 16, 2026

View reviewed changes

mrnkslv self-requested a review April 16, 2026 08:32

mrnkslv approved these changes Apr 16, 2026

View reviewed changes

Keshoid merged commit b7597c0 into release/nodectl/v0.4.0 Apr 16, 2026
5 of 7 checks passed

Keshoid deleted the feature/sma-78-config-pool-ls-toncore-pools-display-is-incomplete branch April 16, 2026 09:01

-                    if let Ok(slot) = joined {
-                        slots.push(slot);
+                    match joined {
+                        Ok(slot) => slots.push(slot),
+                        Err(join_error) => return Err(join_error.into()),

Conversation

Keshoid commented Apr 15, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

linear Bot commented Apr 15, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Apr 16, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 16, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 16, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 16, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Keshoid commented Apr 15, 2026 •

edited

Loading