Cairn
NousResearch/hermes-agent

SQLite read errors on WSL2/ZFS filesystem edge cases

Cluster 100871. Data as of

Nothing open on GitHub. Fix merged in #108082 on 11 Sep 2026. Check it covers your case.

The short answer

0open issues
0open PRs
4items grouped
Known?
Yes. 4 related items grouped here, 1 issue already closed.
Fix event
A fix was merged to main in PR #108082. Check it covers your case. No release data is tracked.
Tracker
The maintainers' tracker marks this resolved, deciding PR #108082. That is separate from the fix event.
Still open
Nothing open on GitHub.

These issues and fixes concern SQLite I/O errors and read connection instability encountered in multi-process or WAL-mode scenarios on WSL2 (ext4-on-vhdx) and OpenZFS filesystems. The work spans bug reports, related issue references, and proposed fixes for session and state database read paths.

Open issues0

No open issues.

Open PRs0

No open pull requests.

Fixes on record

Merged to main. A stamp counts the issues a fix is linked to.

  • #108082 state.db no longer wedges on a healthy WAL: OpenZFS/close() false DeletedWalGenerationError, unconfirmed-WAL read pool, transient read IOERR, stale lock banner (#107401 #105567 #86515 #100871 #108031, salvage #107411 #105578 #105711 #106958)

    merged to main

    files: agent/turn_explainers.py, contributors/emails/[redacted:email], contributors/emails/[email protected], gateway/run_notifications.py, hermes_state.py, hermes_state_dbfile.py, hermes_state_errors.py, hermes_state_readpool.py, hermes_state_wal.py, tests/gateway/test_session_db_warning_recheck.py, tests/hermes_state/test_deleted_wal_generation_guard.py, tests/hermes_state/test_read_path_transient_ioerr.py, tests/hermes_state/test_wal_active_confirmed.py, tests/run_agent/test_turn_completion_explainer.py, tests/test_hermes_state_wal_fallback.py

2 PRs closed without merging

From the close comments: done elsewhere 2

  • #100882 fix(sessions): keep WSL recovery alive through transient read IOERR · done elsewhere → d956e05, #108082
  • #100883 fix(state): pooled read connections survive transient EIO (WSL2/vhdx, ZFS) · done elsewhere → d956e05, #108082
Closed issues · 1
  • Closed#100871[Bug] Repeated SQLITE_IOERR in SessionDB.get_session read path on WSL2 (ext4-on-vhdx), multi-process, starting at 0.21.0 — WAL stable but DELETE fallback collapses read throughput ~30000x
How this cluster was grouped
  • Link evidence (structural links throughout): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.
  • Pairwise link density 0.50: the share of item pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. 1.00 means every pair is linked.

Pick up a Lab task

Check whether a fix covers the open issues here, with your own agent. Bring back evidence for maintainers.

Open the Lab

At a glance

Merged PRs
#108082
Closed issues
1
Attempts
2 PRs closed without merging
Fix link checked
11 Sep 2026

For your agent

The same facts, typed.

cluster/100871.json
cluster/100871.md