Skip to content

fix: add retry logic for channel monitor retrieval during migration #424

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
jvsena42 wants to merge 2 commits into
base: master
Choose a base branch
from
Open

fix: add retry logic for channel monitor retrieval during migration #424

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
0267e23
fix: add retry logic and loggin for channel retrieve
jvsena42 Feb 2, 2026
110a344
chore: add error cause to log
jvsena42 Feb 3, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
56 changes: 47 additions & 9 deletions Bitkit/Services/MigrationsService.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1915,6 +1915,28 @@ extension MigrationsService {
Logger.info("RN remote backup restore completed", context: "Migration")
}

private func retrieveChannelMonitorWithRetry(
channelId: String,
maxAttempts: Int = 3,
baseDelayMs: UInt64 = 1000
) async -> Data? {
for attempt in 1 ... maxAttempts {
do {
return try await RNBackupClient.shared.retrieveChannelMonitor(channelId: channelId)
} catch {
if attempt < maxAttempts {
let delayMs = baseDelayMs * UInt64(attempt)
Logger.debug(
"Retrying channel monitor retrieval for \(channelId) (attempt \(attempt + 1)/\(maxAttempts)) after \(delayMs)ms: \(error)",
context: "Migration"
)
try? await Task.sleep(nanoseconds: delayMs * 1_000_000)
}
}
}
return nil
}

private func fetchRNRemoteLdkData() async {
do {
let files = try await RNBackupClient.shared.listFiles(fileGroup: "ldk")
Expand All @@ -1924,28 +1946,44 @@ extension MigrationsService {
return
}

let monitors = await withTaskGroup(of: Data?.self) { group in
var results: [Data] = []
let expectedCount = files.channel_monitors.count
let monitorResults = await withTaskGroup(of: (String, Data?).self) { group in
var results: [(String, Data?)] = []
for monitorFile in files.channel_monitors {
let channelId = monitorFile.replacingOccurrences(of: ".bin", with: "")
group.addTask {
let channelId = monitorFile.replacingOccurrences(of: ".bin", with: "")
return try? await RNBackupClient.shared.retrieveChannelMonitor(channelId: channelId)
await (channelId, self.retrieveChannelMonitorWithRetry(channelId: channelId))
}
}
for await monitor in group {
if let monitor {
results.append(monitor)
}
for await result in group {
results.append(result)
}
return results
}

let failedMonitors = monitorResults.filter { $0.1 == nil }.map(\.0)
if !failedMonitors.isEmpty {
Logger.error(
"Failed to retrieve \(failedMonitors.count)/\(expectedCount) channel monitors after retries: \(failedMonitors.joined(separator: ", "))",
context: "Migration"
)
}

let monitors = monitorResults.compactMap(\.1)

if monitors.count < expectedCount {
Logger.warn(
"Channel monitor count mismatch: expected \(expectedCount), got \(monitors.count). Some channels may not be recoverable.",
context: "Migration"
)
}

if !monitors.isEmpty {
pendingChannelMigration = PendingChannelMigration(
channelManager: managerData,
channelMonitors: monitors
)
Logger.info("Prepared \(monitors.count) channel monitors for migration", context: "Migration")
Logger.info("Prepared \(monitors.count)/\(expectedCount) channel monitors for migration", context: "Migration")
}
} catch {
Logger.error("Failed to fetch remote LDK data: \(error)", context: "Migration")
Expand Down
Loading