Fix TOCTOU race in snapshot stub ETS operations

[Copilot]

The snapshot stub's start/0 and stop/0 functions have a race window between ets:info/1 check and ets:new/2/ets:delete/1 operations. Concurrent calls can cause badarg exceptions.

Changes

• Wrap ETS operations in try/catch: Both ets:new/2 and ets:delete/1 now catch badarg and return ok, treating the race as a success case (table already created/deleted)

• Add concurrent test coverage: New test module validates idempotent behavior under concurrent access (10-20 processes calling start/stop simultaneously)

start() ->
    case ets:info(?TABLE) of
        undefined ->
            try
                _ = ets:new(?TABLE, [set, named_table, public]),
                ok
            catch
                error:badarg -> ok  % Another process created it
            end;
        _ -> ok
    end.

The operations remain idempotent while handling the edge case gracefully.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• esm.ubuntu.com
  • Triggering command: /usr/lib/apt/methods/https (dns block)
• repo.hex.pm
  • Triggering command: /home/REDACTED/.asdf/installs/erlang/27.1.2/erts-15.1.2/bin/inet_gethost 4 (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Harden ETS start/stop in snapshot stub against concurrent TOCTOU races</issue_title>
<issue_description>## Context

In the test stub module event_sourcing_core_store_snapshot_stub.erl, the current start/0 and stop/0 implementations use ets:info/1 before ets:new/2 and ets:delete/1, which creates a theoretical TOCTOU (Time-Of-Check-Time-Of-Use) race condition if multiple processes attempt to start or stop the backend concurrently.

Current Behavior

Between the check (ets:info/1) and the action (ets:new/2 or ets:delete/1), another process could create or delete the named table, potentially causing a badarg exception.

Proposed Improvement

Wrap the ets:new/2 and ets:delete/1 calls in try/catch blocks to handle badarg gracefully, making the operations truly idempotent under concurrent access.

Priority

This is a theoretical issue in the current usage, as the backend is not started/stopped from multiple processes. However, it would be good practice to ensure proper concurrent idempotence for robustness.

References

• PR: Refactor/event store #26
• Discussion: Refactor/event store #26 (comment)
• Requested by: @ccamel</issue_description>

Comments on the Issue (you are @copilot in this section)

@ccamel Not a real issue in tests, but worth fixing.

Note

Custom agent used: erlang_otp_developer
Erlang/OTP Developer with expertise in building fault-tolerant, distributed systems using Erlang and OTP principles in an event-sourced architecture.

• Fixes Harden ETS start/stop in snapshot stub against concurrent TOCTOU races #28

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}