JSM administrative recovery checklist
Export Wezy Voting Data Before Uninstalling: JSONL Checklist
Prepare an authorized Wezy JSM recovery export before uninstalling. Save every JSONL batch and manifest, verify local files, and distinguish export from restore.
In short
Use administrative recovery, not the analytical CSV, when preserving authorized Wezy-held records before removal. Save every batch and its manifest evidence, check the files locally, and never treat an export as an automatic reinstallation or restore plan.
In this guide
Before uninstalling Wezy, decide whether you need an authorized recovery export and complete it while the required access is available. An analytical CSV is a reporting view, not a substitute for administrative recovery. Neither kind of export is a documented automatic restore into a new installation.
This runbook helps an administrator produce an evidence-backed handover. It does not authorize bulk collection outside your role, prescribe a legal retention period or guarantee recovery after removal.
Choose the export that matches the purpose
| Purpose | Documented route | Boundary |
|---|---|---|
| Review current demand or a covered analytical period | Dashboard analytical CSV | Filtered and permission-scoped reporting |
| Recover available Wezy administrative records | Administration · Data recovery | Authorized project administration and its allowed scope |
| Back up the entire Jira site | Not provided by this Wezy workflow | Use the appropriate Atlassian administration process |
| Restore votes automatically into another installation | Not established by these exports | Do not promise an import or restoration capability |
The official recovery documentation says recovery can be available without an active commercial license, subject to authorization. That is not a reason to remove the app first or assume every former administrator will retain access afterward.
Assign ownership and prepare a restricted destination
Identify the person authorized to run recovery, the person responsible for checking completeness, and who may receive the files. Choose an approved restricted storage location before starting. Recovery files can contain Atlassian identifiers, vote states, organization attributions and settings.
Document the project, installation context, account role, app version, start time and intended purpose. Keep passwords, API tokens and unnecessary private ticket content out of the handover note. If a broader export is needed, obtain the appropriate authorization rather than widening permissions as a troubleshooting shortcut.
This preparation matters because Wezy cannot remotely revoke a file already downloaded to a computer.
Receive and save the complete sequence
In the dashboard, expand Administration · Data recovery. The documented sequence is:
- Select Receive first batch.
- Save the result with Save JSONL batch.
- Select Receive next batch and save that batch.
- Continue until the interface indicates completion.
- Keep all batches and the accompanying manifest metadata.
Do not overwrite earlier files with a later batch. Use a local naming convention that preserves the sequence and links files to the same recovery operation, without assuming that a filename itself is proof of the native batch number.
The manifest can report inaccessible requests that were omitted. A completed operation means completion of the authorized export process; it does not mean all Jira requests across the site were included.
Verify four different kinds of completeness
Sequence completeness
Compare the saved files with the batches actually received and the completion/manifest information. A single valid JSONL file does not establish that you saved every batch. If the operation remains incomplete or the next-batch result was not saved, stop before uninstalling.
File readability
Open or validate each file locally using approved tooling. Check that the expected batch files exist, are not unexpectedly empty and contain parseable JSONL. A parser proves syntax, not the meaning or completeness of the recovered records.
Do not upload a full recovery export to a public JSON validator or an external support conversation just to check its format.
File integrity after handover
Calculate a SHA-256 digest for each saved file and store it in your handover inventory. Recalculate it after transfer to detect changed bytes. A matching hash proves consistency with the hashed file; it does not prove that the export contained every eligible record in the first place.
Scope completeness
Read the reported omissions and permissions context. Document what was excluded and why the recipient should not treat the export as a site-wide census. If an authorized reviewer expected an accessible request that is absent, investigate before treating the handover as complete.
Use a local evidence inventory
The following columns are a proposed team worksheet, not claimed Wezy manifest fields:
Recovery operation reference and observation time
Project and exporting role
Saved filename
Batch label or sequence evidence shown by the interface
Locally observed JSONL record count
SHA-256 digest
Manifest/completion evidence location
Omitted scope and unresolved questions
Checker and handover destination
For example, three saved files with valid JSON can still be incomplete if one is a duplicate of the first batch and the second batch was never saved. That is why file count, syntax and hashes must be checked alongside the operation’s sequence evidence. This is a fictional failure example, not a reported Wezy incident.
Check saved JSONL files without uploading their contents
Download the local JSONL inventory helper and run it with the specific saved files:
python jsm-jsonl-inventory.py batch-a.jsonl batch-b.jsonl
It reads files locally, checks each nonblank line as UTF-8 JSON, counts records and calculates SHA-256 digests. Empty files, malformed JSON and invalid encoding stop the inventory with an error. The output contains filenames, sizes, counts and hashes—not the recovered record contents. Repeated digests are flagged for review, not automatically deleted or declared invalid.
The helper does not understand Wezy’s native manifest, verify the record schema or certify that all batches were received. Its complete_recovery result is deliberately unavailable. Compare the native sequence, completion state and omissions separately, and keep the generated inventory in the same restricted handover location as the files.
Decide what removal should actually accomplish
Removing a request from voting closes its current voting journey without erasing the request’s Wezy data. Withdrawing changes one customer’s support. Authorized erasure is a separate irreversible process for the permitted Wezy-held scope; it does not delete the Jira request. Those operations should not be substituted for one another to achieve a superficially clean dashboard.
If the actual problem is suspended writes, inspect Data and privacy → Voting availability and follow the displayed guidance. Refreshing settings rereads the state; it does not itself lift a suspension. The maintenance guide specifically advises against uninstalling to resolve this situation.
Close the handover before uninstalling
Record that required batches are saved, local checks are complete, omissions are understood, the recipient can access the approved destination and unresolved restoration requirements have been addressed. Mark each item passed, failed or not tested; do not silently count a missing check as success.
Only then carry out the intended removal through Atlassian app management and separately verify the subscription status. Reinstalling does not automatically restore the former installation. Hosted lifecycle and retention conditions are not a substitute for a verified pre-removal recovery plan.
Wezy Customer Voting’s current listing links its documentation and support. Use Wezy support for unresolved scope or lifecycle questions with a sanitized description first, not an unsolicited full export. The value of this checklist is a defensible record of what was preserved—not an unsupported promise that removal is reversible.
Published by the team behind Wezy Customer Voting. Product capabilities are linked to their official documentation; worked examples are illustrative.
Let customers vote on your JSM portal.
Install Wezy from Atlassian Marketplace, open one request for voting and see the first vote in Jira.