stablyai/orca · error · Error

Remote filesystem unavailable. Reconnect the SSH target befo

Error message

Remote filesystem unavailable. Reconnect the SSH target before retrying.

What it means

Thrown by 'hooks:writeIssueCommand' for an SSH repo (repo.connectionId set) when getSshFilesystemProvider(repo.connectionId) returns nothing. The remote filesystem provider is the bridge that writes .orca/issue-command over SSH; its absence means the SSH connection is gone, so writing is refused rather than silently dropping the command.

Source

Thrown at src/main/ipc/worktrees.ts:3446

              : ('none' as const)
        }
      }
      return readIssueCommand(repo.path)
    }
  )

  ipcMain.handle(
    'hooks:writeIssueCommand',
    async (_event, args: { repoId: string; content: string; hostId?: ExecutionHostId }) => {
      const repo = getRepoForWorktreeRemoval(store, args.repoId, args.hostId)
      if (!repo || isFolderRepo(repo)) {
        return
      }
      if (repo.connectionId) {
        const issueCommandPath = joinWorktreeRelativePath(repo.path, '.orca/issue-command')
        const fsProvider = getSshFilesystemProvider(repo.connectionId)
        if (!fsProvider) {
          throw new Error(
            'Remote filesystem unavailable. Reconnect the SSH target before retrying.'
          )
        }
        const trimmed = args.content.trim()
        if (!trimmed) {
          await fsProvider.deletePath(issueCommandPath, false).catch((error: unknown) => {
            if (!isENOENT(error)) {
              throw error
            }
          })
          return
        }
        await fsProvider.createDir(joinWorktreeRelativePath(repo.path, '.orca'))
        const gitignorePath = joinWorktreeRelativePath(repo.path, '.gitignore')
        try {
          const result = await fsProvider.readFile(gitignorePath)
          if (!result.isBinary && !/^\.orca\/?$/m.test(result.content)) {
            const separator = result.content.endsWith('\n') ? '' : '\n'

View on GitHub (pinned to 1136503c6a)

Solutions

1. Reconnect the SSH target, then retry the write.
2. Guard the UI: if getSshFilesystemProvider(repo.connectionId) is falsy, disable Save and show a reconnect prompt.
3. For local repos this path is not taken; ensure repo.connectionId is not stale.

Example fix

// before
await ipc.invoke('hooks:writeIssueCommand', { repoId, content, hostId })
// after
if (repo.connectionId && !getSshFilesystemProvider(repo.connectionId)) {
  await promptReconnect(repo.connectionId)
}
await ipc.invoke('hooks:writeIssueCommand', { repoId, content, hostId })

if (repo.connectionId && !getSshFilesystemProvider(repo.connectionId)) { await promptReconnect(repo.connectionId); return }

function hasRemoteFilesystem(connectionId: string | undefined): boolean {
  return !connectionId || Boolean(getSshFilesystemProvider(connectionId))
}

try { await ipc.invoke('hooks:writeIssueCommand', { repoId, content, hostId }) }
catch (e) {
  if (/Remote filesystem unavailable/.test(String((e as Error).message))) await reconnectSsh(repo.connectionId)
  else throw e
}

When it happens

Trigger: The SSH target dropped or was disconnected after the repo was opened, then the user edits the issue-command. Also when the SSH provider module is torn down during quit while a write is in flight.

Common situations: Network interruption to the SSH host, SSH agent/keys removed, host rebooted, or the connection was closed from another pane. Reconnecting usually restores the provider.

Related errors

• SSH target "${input.target.connectionId}" is not connected.
• SSH target did not connect: ${state.status}
• Remote connection dropped. Click Reconnect on the SSH target
• SSH Git provider is not available. Reconnect to this target
• SSH Git provider is not available. Reconnect to this target

AI-assisted analysis of stablyai/orca@1136503c6a (2026-08-12). Data as JSON: /api/errors/a38394e7951a6b78. Report an issue: GitHub.