Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

question(cloud): supported cutover to a fresh Cloud database without losing sync cursor and deletion safety

Open
#1,719 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
25/100
Issue type
Feature
Clarity
Clearly specified
Activity status
Active
Tech stack
go, sqlite

Research direction

Read the 3.2.1 autosync paths in internal/cloud/autosync/manager.go (defaults and pull cursor), chunk tracking in internal/sync/sync.go, and remote mutation/deletion-floor handling in internal/store/store.go. First determine whether the protocol has a server identity or epoch and how pending writes and other clients behave across a rebuilt database. Done means maintainers define a supported, sequence-safe cutover procedure or clarify that none exists; the issue reports no end-to-end cutover test.

Written by the indexing model from the issue text.

Description

status:needs-review type:question
Question

What is the supported Engram 3.2.1 procedure for moving an existing SQLite client to a fresh official Engram Cloud database while preserving synchronization continuity and deletion safety?

The fresh Cloud is populated through the official current-state remirror path, but its mutation sequence range can be lower than the existing client's persisted global cloud pull cursor. How should a migration handle that cursor, previously imported chunks, and deletion floors without resetting store state manually, fabricating acknowledgments, resurrecting deleted entities, or losing newer writes?

Does Engram support a server-instance identity/epoch or an official migration operation for this case? If not, what sequence-preserving procedure do maintainers recommend before changing the client endpoint?

Why does this need issue tracking?

This requires a durable sync-protocol decision or a supported migration contract, not only deployment troubleshooting. A successful current-state export and import into an empty client does not demonstrate that an existing autosync client can consume a rebuilt server's journal safely. Guidance should account for deletion floors, pending local mutations, concurrent writes, and other existing clients.

Affected Area

Sync (multi-instance)

Additional Context

Scope: official Engram CLI and Cloud 3.2.1. This is a maintainer guidance request based on source inspection and isolated current-state import checks; an end-to-end autosync cutover was not executed, and no production cursor or deletion floor was reset.

Relevant release source:

Related, but not equivalent: #1603 and #570 concern routing/named remotes; #595 concerns full current-state remirror. This question is specifically about preserving existing-client continuity when the backing Cloud database and its sequence space are recreated. It does not propose a routing implementation or manual SQL reset.

Dominant language
Go
Stars
7.1k
Forks
729
Avg merge
10h 43m
Merged PRs (30d)
332

Getting set up

Open in Codespaces

Starts the project's dev container in your browser, under your own GitHub account.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from Gentleman-Programming/engram

All issues in Gentleman-Programming/engram

Similar issues

More Go issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.