mirror of
https://github.com/vitorpamplona/amethyst.git
synced 2026-07-30 19:46:17 +00:00
fix(coordinator): retryable batched-REQ dedup and race-free EOSE aggregator
Reviewer davotoula (PR #3483) flagged two commons/relayClient issues on FeedMetadataCoordinator that both bite the Android app once WoT is wired there: 5. loadKind3Batched / loadMetadataBatched marked pubkeys as sent BEFORE any relay EOSE'd. On flaky-network cold-starts where every index relay timed out, the pubkeys stayed permanently marked and WoT was silently empty for the whole session — the next call short-circuited. Fix: pubkeys enter `queuedKind3Pubkeys` / `queuedPubkeys` only after ≥1 EOSE; on zero-EOSE timeout they roll out of the new `inFlightBatched*` sets so a subsequent call retries. 6. `val eoseReceived = mutableSetOf<NormalizedRelayUrl>()` was mutated from per-relay `onEose` callbacks the client dispatches on `Dispatchers.IO`. Concurrent `add()`/`size` on an unsynchronised HashSet could drop entries or throw CME, forcing the batch to wait the full timeout instead of firing early. Fix: `BatchEoseGate` funnels EOSE notifications through a `Channel` so a single consumer coroutine is the sole reader/writer of the `seen` set — KMP-safe, no `synchronized {}` or JVM-only atomics. Tests exercise: - zero-EOSE timeout → retry re-fires - ≥1 EOSE → next call short-circuits - full-EOSE from 20 relays hammered from Dispatchers.IO in parallel - clear() releases in-flight dedup - same semantics on loadMetadataBatched Plan: commons/plans/2026-07-06-fix-wot-outbox-model-and-review-fixes-plan.md
This commit is contained in:
+98
-18
@@ -39,6 +39,7 @@ import com.vitorpamplona.quartz.nip18Reposts.RepostEvent
|
||||
import com.vitorpamplona.quartz.nip25Reactions.ReactionEvent
|
||||
import kotlinx.coroutines.CompletableDeferred
|
||||
import kotlinx.coroutines.CoroutineScope
|
||||
import kotlinx.coroutines.channels.Channel
|
||||
import kotlinx.coroutines.launch
|
||||
import kotlinx.coroutines.withTimeoutOrNull
|
||||
|
||||
@@ -76,6 +77,14 @@ class FeedMetadataCoordinator(
|
||||
private val queuedBoostedIds = mutableSetOf<HexKey>()
|
||||
private val queuedKind3Pubkeys = mutableSetOf<HexKey>()
|
||||
|
||||
// Batched paths only — pubkeys currently in-flight in a batched REQ.
|
||||
// Prevents rapid re-fire of the same batch. Distinct from queuedPubkeys
|
||||
// and queuedKind3Pubkeys (which record "asked and at least one relay
|
||||
// returned EOSE") so a batch that times out with zero events can be
|
||||
// retried on the next call — see PR #3483 review finding 5.
|
||||
private val inFlightBatchedMetadata = mutableSetOf<HexKey>()
|
||||
private val inFlightBatchedKind3 = mutableSetOf<HexKey>()
|
||||
|
||||
/**
|
||||
* Start processing the subscription queue.
|
||||
* Call once when coordinator is created.
|
||||
@@ -253,14 +262,24 @@ class FeedMetadataCoordinator(
|
||||
/**
|
||||
* Fast-path: batched metadata subscription for visible-viewport authors.
|
||||
* Bypasses rate limiter. Single filter with all authors. Closes after EOSE.
|
||||
*
|
||||
* Pubkeys are moved into [queuedPubkeys] (dedup) only after at least one
|
||||
* relay EOSE'd. On timeout with zero EOSE (index relays all unreachable)
|
||||
* they roll out of [inFlightBatchedMetadata] so a subsequent call can
|
||||
* retry — see PR #3483 review finding 5.
|
||||
*/
|
||||
fun loadMetadataBatched(
|
||||
pubkeys: List<HexKey>,
|
||||
timeoutMs: Long = 5_000L,
|
||||
) {
|
||||
val newPubkeys = pubkeys.filter { it !in queuedPubkeys }.distinct()
|
||||
val newPubkeys =
|
||||
pubkeys
|
||||
.asSequence()
|
||||
.filter { it !in queuedPubkeys && it !in inFlightBatchedMetadata }
|
||||
.distinct()
|
||||
.toList()
|
||||
if (newPubkeys.isEmpty()) return
|
||||
queuedPubkeys.addAll(newPubkeys)
|
||||
inFlightBatchedMetadata.addAll(newPubkeys)
|
||||
|
||||
scope.launch {
|
||||
val filter =
|
||||
@@ -271,8 +290,7 @@ class FeedMetadataCoordinator(
|
||||
)
|
||||
val filterMap = indexRelays.associateWith { listOf(filter) }
|
||||
val subId = newSubId()
|
||||
val eoseReceived = mutableSetOf<NormalizedRelayUrl>()
|
||||
val allEose = CompletableDeferred<Unit>()
|
||||
val gate = BatchEoseGate(scope, target = indexRelays.size)
|
||||
|
||||
val listener =
|
||||
object : SubscriptionListener {
|
||||
@@ -289,16 +307,18 @@ class FeedMetadataCoordinator(
|
||||
relay: NormalizedRelayUrl,
|
||||
forFilters: List<Filter>?,
|
||||
) {
|
||||
eoseReceived.add(relay)
|
||||
if (eoseReceived.size >= indexRelays.size) {
|
||||
allEose.complete(Unit)
|
||||
}
|
||||
gate.notifyEose(relay)
|
||||
}
|
||||
}
|
||||
|
||||
client.subscribe(subId, filterMap, listener)
|
||||
withTimeoutOrNull(timeoutMs) { allEose.await() }
|
||||
val eosedRelays = gate.awaitAll(timeoutMs)
|
||||
client.unsubscribe(subId)
|
||||
|
||||
if (eosedRelays > 0) {
|
||||
queuedPubkeys.addAll(newPubkeys)
|
||||
}
|
||||
inFlightBatchedMetadata.removeAll(newPubkeys.toSet())
|
||||
}
|
||||
}
|
||||
|
||||
@@ -311,18 +331,29 @@ class FeedMetadataCoordinator(
|
||||
* so relays with per-filter author caps (nostr-rs-relay defaults to
|
||||
* ~100) don't silently truncate the batch. Aggregates EOSE across
|
||||
* chunks and calls [onEose] once (or after [timeoutMs]).
|
||||
*
|
||||
* Pubkeys are moved into [queuedKind3Pubkeys] (dedup) only after at
|
||||
* least one relay EOSE'd. On timeout with zero EOSE (index relays all
|
||||
* unreachable — common on flaky mobile networks) they roll out of
|
||||
* [inFlightBatchedKind3] so the next `loadKind3Batched` call retries
|
||||
* — see PR #3483 review finding 5.
|
||||
*/
|
||||
fun loadKind3Batched(
|
||||
pubkeys: Collection<HexKey>,
|
||||
timeoutMs: Long = 5_000L,
|
||||
onEose: () -> Unit = {},
|
||||
) {
|
||||
val newPubkeys = pubkeys.filter { it !in queuedKind3Pubkeys }.distinct()
|
||||
val newPubkeys =
|
||||
pubkeys
|
||||
.asSequence()
|
||||
.filter { it !in queuedKind3Pubkeys && it !in inFlightBatchedKind3 }
|
||||
.distinct()
|
||||
.toList()
|
||||
if (newPubkeys.isEmpty()) {
|
||||
onEose()
|
||||
return
|
||||
}
|
||||
queuedKind3Pubkeys.addAll(newPubkeys)
|
||||
inFlightBatchedKind3.addAll(newPubkeys)
|
||||
|
||||
scope.launch {
|
||||
val filters =
|
||||
@@ -335,8 +366,7 @@ class FeedMetadataCoordinator(
|
||||
}
|
||||
val filterMap = indexRelays.associateWith { filters }
|
||||
val subId = newSubId()
|
||||
val eoseReceived = mutableSetOf<NormalizedRelayUrl>()
|
||||
val allEose = CompletableDeferred<Unit>()
|
||||
val gate = BatchEoseGate(scope, target = indexRelays.size)
|
||||
|
||||
val listener =
|
||||
object : SubscriptionListener {
|
||||
@@ -353,16 +383,19 @@ class FeedMetadataCoordinator(
|
||||
relay: NormalizedRelayUrl,
|
||||
forFilters: List<Filter>?,
|
||||
) {
|
||||
eoseReceived.add(relay)
|
||||
if (eoseReceived.size >= indexRelays.size) {
|
||||
allEose.complete(Unit)
|
||||
}
|
||||
gate.notifyEose(relay)
|
||||
}
|
||||
}
|
||||
|
||||
client.subscribe(subId, filterMap, listener)
|
||||
withTimeoutOrNull(timeoutMs) { allEose.await() }
|
||||
val eosedRelays = gate.awaitAll(timeoutMs)
|
||||
client.unsubscribe(subId)
|
||||
|
||||
if (eosedRelays > 0) {
|
||||
queuedKind3Pubkeys.addAll(newPubkeys)
|
||||
}
|
||||
inFlightBatchedKind3.removeAll(newPubkeys.toSet())
|
||||
|
||||
onEose()
|
||||
}
|
||||
}
|
||||
@@ -375,5 +408,52 @@ class FeedMetadataCoordinator(
|
||||
queuedPubkeys.clear()
|
||||
queuedNoteIds.clear()
|
||||
queuedKind3Pubkeys.clear()
|
||||
inFlightBatchedMetadata.clear()
|
||||
inFlightBatchedKind3.clear()
|
||||
}
|
||||
|
||||
/**
|
||||
* Aggregates EOSE notifications from per-relay `onEose` callbacks
|
||||
* (which the client may dispatch on `Dispatchers.IO`) via a
|
||||
* [Channel]. The consumer coroutine is the sole reader/writer of the
|
||||
* `seen` set, eliminating the race the previous `mutableSetOf` +
|
||||
* shared-state check had — see PR #3483 review finding 6.
|
||||
*
|
||||
* [awaitAll] blocks up to [timeoutMs] and returns the number of
|
||||
* relays that EOSE'd (may be less than [target] on timeout). The
|
||||
* count feeds the retry decision in the batched loaders.
|
||||
*/
|
||||
private class BatchEoseGate(
|
||||
private val scope: CoroutineScope,
|
||||
private val target: Int,
|
||||
) {
|
||||
private val incoming = Channel<NormalizedRelayUrl>(Channel.UNLIMITED)
|
||||
private val done = CompletableDeferred<Unit>()
|
||||
|
||||
@Volatile private var lastCount = 0
|
||||
|
||||
fun notifyEose(relay: NormalizedRelayUrl) {
|
||||
incoming.trySend(relay)
|
||||
}
|
||||
|
||||
suspend fun awaitAll(timeoutMs: Long): Int {
|
||||
if (target <= 0) return 0
|
||||
val consumer =
|
||||
scope.launch {
|
||||
val seen = mutableSetOf<NormalizedRelayUrl>()
|
||||
for (relay in incoming) {
|
||||
if (seen.add(relay)) {
|
||||
lastCount = seen.size
|
||||
if (seen.size >= target && !done.isCompleted) {
|
||||
done.complete(Unit)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
withTimeoutOrNull(timeoutMs) { done.await() }
|
||||
incoming.close()
|
||||
consumer.join()
|
||||
return lastCount
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
+297
@@ -0,0 +1,297 @@
|
||||
/*
|
||||
* Copyright (c) 2025 Vitor Pamplona
|
||||
*
|
||||
* Permission is hereby granted, free of charge, to any person obtaining a copy of
|
||||
* this software and associated documentation files (the "Software"), to deal in
|
||||
* the Software without restriction, including without limitation the rights to use,
|
||||
* copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the
|
||||
* Software, and to permit persons to whom the Software is furnished to do so,
|
||||
* subject to the following conditions:
|
||||
*
|
||||
* The above copyright notice and this permission notice shall be included in all
|
||||
* copies or substantial portions of the Software.
|
||||
*
|
||||
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
||||
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS
|
||||
* FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR
|
||||
* COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
|
||||
* AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
||||
* WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
||||
*/
|
||||
package com.vitorpamplona.amethyst.commons.relayClient.assemblers
|
||||
|
||||
import com.vitorpamplona.quartz.nip01Core.core.HexKey
|
||||
import com.vitorpamplona.quartz.nip01Core.relay.client.EmptyNostrClient
|
||||
import com.vitorpamplona.quartz.nip01Core.relay.client.INostrClient
|
||||
import com.vitorpamplona.quartz.nip01Core.relay.client.reqs.SubscriptionListener
|
||||
import com.vitorpamplona.quartz.nip01Core.relay.filters.Filter
|
||||
import com.vitorpamplona.quartz.nip01Core.relay.normalizer.NormalizedRelayUrl
|
||||
import kotlinx.coroutines.CoroutineScope
|
||||
import kotlinx.coroutines.Dispatchers
|
||||
import kotlinx.coroutines.SupervisorJob
|
||||
import kotlinx.coroutines.cancel
|
||||
import kotlinx.coroutines.delay
|
||||
import kotlinx.coroutines.launch
|
||||
import kotlinx.coroutines.runBlocking
|
||||
import org.junit.After
|
||||
import org.junit.Assert.assertEquals
|
||||
import org.junit.Assert.assertTrue
|
||||
import org.junit.Before
|
||||
import org.junit.Test
|
||||
|
||||
/**
|
||||
* Regression tests for PR #3483 review findings on FeedMetadataCoordinator:
|
||||
*
|
||||
* - Finding 5: `queuedKind3Pubkeys` was marked-on-send, so if every index
|
||||
* relay timed out the pubkeys were permanently marked and subsequent
|
||||
* calls short-circuited — WoT stayed empty for the whole session.
|
||||
* Fix: pubkeys land in `queuedKind3Pubkeys` only after ≥1 EOSE; on
|
||||
* zero-EOSE timeout they roll out of `inFlightBatchedKind3` for retry.
|
||||
*
|
||||
* - Finding 6: `eoseReceived: MutableSet` was mutated from per-relay
|
||||
* `onEose` callbacks running on `Dispatchers.IO` with no sync. Fix:
|
||||
* `BatchEoseGate` funnels EOSE notifications through a `Channel` so a
|
||||
* single consumer coroutine is the sole reader/writer of the `seen`
|
||||
* set.
|
||||
*/
|
||||
class FeedMetadataCoordinatorTest {
|
||||
private lateinit var scope: CoroutineScope
|
||||
private val relay1 = NormalizedRelayUrl("wss://relay1.test/")
|
||||
private val relay2 = NormalizedRelayUrl("wss://relay2.test/")
|
||||
private val relay3 = NormalizedRelayUrl("wss://relay3.test/")
|
||||
private val indexRelays = setOf(relay1, relay2, relay3)
|
||||
|
||||
@Before
|
||||
fun setup() {
|
||||
scope = CoroutineScope(SupervisorJob() + Dispatchers.Default)
|
||||
}
|
||||
|
||||
@After
|
||||
fun teardown() {
|
||||
scope.cancel()
|
||||
}
|
||||
|
||||
private fun pubkey(seed: Int): HexKey = seed.toString(16).padStart(64, '0')
|
||||
|
||||
/**
|
||||
* Fake client that captures subscribe/unsubscribe and lets the test
|
||||
* drive EOSE notifications on any dispatcher we choose.
|
||||
*/
|
||||
private class ControllableClient(
|
||||
private val delegate: INostrClient = EmptyNostrClient(),
|
||||
) : INostrClient by delegate {
|
||||
val subscriptions = mutableMapOf<String, SubscriptionListener?>()
|
||||
val subscribeCalls = mutableListOf<Map<NormalizedRelayUrl, List<Filter>>>()
|
||||
var unsubscribeCallCount = 0
|
||||
private set
|
||||
|
||||
override fun subscribe(
|
||||
subId: String,
|
||||
filters: Map<NormalizedRelayUrl, List<Filter>>,
|
||||
listener: SubscriptionListener?,
|
||||
) {
|
||||
subscriptions[subId] = listener
|
||||
subscribeCalls.add(filters)
|
||||
}
|
||||
|
||||
override fun unsubscribe(subId: String) {
|
||||
subscriptions.remove(subId)
|
||||
unsubscribeCallCount++
|
||||
}
|
||||
|
||||
fun fireEose(relay: NormalizedRelayUrl) {
|
||||
subscriptions.values.filterNotNull().forEach { it.onEose(relay, forFilters = null) }
|
||||
}
|
||||
}
|
||||
|
||||
@Test
|
||||
fun `loadKind3Batched retries after zero-EOSE timeout`() =
|
||||
runBlocking {
|
||||
val client = ControllableClient()
|
||||
val coordinator =
|
||||
FeedMetadataCoordinator(
|
||||
client = client,
|
||||
scope = scope,
|
||||
indexRelays = indexRelays,
|
||||
)
|
||||
|
||||
val pubkeys = listOf(pubkey(1), pubkey(2), pubkey(3))
|
||||
|
||||
// Call 1 — no relay EOSEs; must time out.
|
||||
coordinator.loadKind3Batched(pubkeys, timeoutMs = 200)
|
||||
delay(350) // exceed the timeout
|
||||
|
||||
// Call 2 — the same pubkeys must be re-subscribed since call 1
|
||||
// never got a successful EOSE. The old code would silently
|
||||
// short-circuit here.
|
||||
coordinator.loadKind3Batched(pubkeys, timeoutMs = 200)
|
||||
delay(50) // let the launcher run
|
||||
|
||||
assertEquals(
|
||||
"Zero-EOSE timeout must not permanently dedup pubkeys",
|
||||
2,
|
||||
client.subscribeCalls.size,
|
||||
)
|
||||
assertEquals(
|
||||
"Second call must re-request the same author set",
|
||||
pubkeys.size,
|
||||
client.subscribeCalls[1]
|
||||
.values
|
||||
.first()
|
||||
.first()
|
||||
.authors!!
|
||||
.size,
|
||||
)
|
||||
}
|
||||
|
||||
@Test
|
||||
fun `loadKind3Batched short-circuits after successful EOSE`() =
|
||||
runBlocking {
|
||||
val client = ControllableClient()
|
||||
val coordinator =
|
||||
FeedMetadataCoordinator(
|
||||
client = client,
|
||||
scope = scope,
|
||||
indexRelays = indexRelays,
|
||||
)
|
||||
|
||||
val pubkeys = listOf(pubkey(1), pubkey(2))
|
||||
|
||||
coordinator.loadKind3Batched(pubkeys, timeoutMs = 1_000)
|
||||
// Give the launcher time to register the listener before we fire.
|
||||
delay(50)
|
||||
indexRelays.forEach(client::fireEose)
|
||||
delay(200) // let the coordinator finish + promote to queued
|
||||
|
||||
coordinator.loadKind3Batched(pubkeys, timeoutMs = 200)
|
||||
delay(50)
|
||||
|
||||
assertEquals(
|
||||
"Successful call must dedup subsequent identical calls",
|
||||
1,
|
||||
client.subscribeCalls.size,
|
||||
)
|
||||
}
|
||||
|
||||
@Test
|
||||
fun `loadKind3Batched promotes even when only some relays EOSE`() =
|
||||
runBlocking {
|
||||
val client = ControllableClient()
|
||||
val coordinator =
|
||||
FeedMetadataCoordinator(
|
||||
client = client,
|
||||
scope = scope,
|
||||
indexRelays = indexRelays,
|
||||
)
|
||||
|
||||
val pubkeys = listOf(pubkey(1))
|
||||
|
||||
coordinator.loadKind3Batched(pubkeys, timeoutMs = 300)
|
||||
delay(30)
|
||||
// Only 1 of 3 EOSEs — timeout still fires but we made progress.
|
||||
client.fireEose(relay1)
|
||||
delay(400)
|
||||
|
||||
coordinator.loadKind3Batched(pubkeys, timeoutMs = 200)
|
||||
delay(50)
|
||||
|
||||
assertEquals(
|
||||
"≥1 EOSE = progress = promote to queued (avoid re-asking)",
|
||||
1,
|
||||
client.subscribeCalls.size,
|
||||
)
|
||||
}
|
||||
|
||||
/**
|
||||
* Regression for finding 6 — pumps EOSE from many dispatchers in
|
||||
* parallel. The old MutableSet-based code could drop entries or throw
|
||||
* ConcurrentModificationException on the internal HashSet iterator.
|
||||
* BatchEoseGate must aggregate every distinct relay exactly once.
|
||||
*/
|
||||
@Test
|
||||
fun `EOSE aggregator is safe under concurrent per-relay callbacks`() =
|
||||
runBlocking {
|
||||
val bigIndexSet =
|
||||
(0..19).map { NormalizedRelayUrl("wss://relay$it.test/") }.toSet()
|
||||
val client = ControllableClient()
|
||||
val coordinator =
|
||||
FeedMetadataCoordinator(
|
||||
client = client,
|
||||
scope = scope,
|
||||
indexRelays = bigIndexSet,
|
||||
)
|
||||
|
||||
coordinator.loadKind3Batched(listOf(pubkey(1)), timeoutMs = 2_000)
|
||||
delay(50) // wait for subscription
|
||||
|
||||
// Fire EOSEs concurrently from many dispatchers.
|
||||
val jobs =
|
||||
bigIndexSet.map { relay ->
|
||||
scope.launch(Dispatchers.IO) {
|
||||
client.fireEose(relay)
|
||||
}
|
||||
}
|
||||
jobs.forEach { it.join() }
|
||||
|
||||
// The 2nd call must short-circuit — every relay EOSE'd, so
|
||||
// pubkey(1) is now in queuedKind3Pubkeys.
|
||||
delay(100)
|
||||
coordinator.loadKind3Batched(listOf(pubkey(1)), timeoutMs = 200)
|
||||
delay(50)
|
||||
|
||||
assertEquals(
|
||||
"Under concurrent EOSE from all relays, aggregator must reach target",
|
||||
1,
|
||||
client.subscribeCalls.size,
|
||||
)
|
||||
}
|
||||
|
||||
@Test
|
||||
fun `loadMetadataBatched follows the same retry semantics`() =
|
||||
runBlocking {
|
||||
val client = ControllableClient()
|
||||
val coordinator =
|
||||
FeedMetadataCoordinator(
|
||||
client = client,
|
||||
scope = scope,
|
||||
indexRelays = indexRelays,
|
||||
)
|
||||
|
||||
val pubkeys = listOf(pubkey(1), pubkey(2))
|
||||
|
||||
// Call 1 — zero EOSE, timeout.
|
||||
coordinator.loadMetadataBatched(pubkeys, timeoutMs = 200)
|
||||
delay(350)
|
||||
// Call 2 — must re-subscribe.
|
||||
coordinator.loadMetadataBatched(pubkeys, timeoutMs = 200)
|
||||
delay(50)
|
||||
|
||||
assertTrue(
|
||||
"Metadata batch also retries on zero-EOSE timeout",
|
||||
client.subscribeCalls.size >= 2,
|
||||
)
|
||||
}
|
||||
|
||||
@Test
|
||||
fun `clear releases in-flight dedup so a fresh call always fires`() =
|
||||
runBlocking {
|
||||
val client = ControllableClient()
|
||||
val coordinator =
|
||||
FeedMetadataCoordinator(
|
||||
client = client,
|
||||
scope = scope,
|
||||
indexRelays = indexRelays,
|
||||
)
|
||||
|
||||
coordinator.loadKind3Batched(listOf(pubkey(1)), timeoutMs = 200)
|
||||
delay(50)
|
||||
// clear() must drop the in-flight tracker even mid-request.
|
||||
coordinator.clear()
|
||||
delay(300) // let call 1 finish + roll back
|
||||
|
||||
coordinator.loadKind3Batched(listOf(pubkey(1)), timeoutMs = 200)
|
||||
delay(50)
|
||||
|
||||
assertTrue(client.subscribeCalls.size >= 2)
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user