Slack Import Shows 0 Users and 0 Channels

Community Support
1

Description

When I attempt to follow the instructions on importing Slack data, I see the import includes about 4K messages but 0 users and 0 channels. I cannot proceed with an import whether I upload via the web UI or import from a server path. How do I import Slack data when Rocket.Chat indicates there are no users or channels, despite being present in the export data?

image

Server Setup Information

  • Version of Rocket.Chat Server: 5.4.2
  • Operating System: Debian 11
  • Deployment Method: manual installation for Debian based distros
  • Number of Running Instances: 1
  • DB Replicaset Oplog: Enabled
  • NodeJS Version: 14.21.1
  • MongoDB Version: 5.0.14
  • Proxy: Apache
  • Firewalls involved: none enabled

Any additional Information

I attempted to identify what the system was doing during the import, but Administration > View Logs shows essentially no information (only that I am accessing the import URL) and doesn’t indicate any issues. No import errors are displayed in the web UI.

2

Hi! Apologies for delayed response. Please, review the steps to follow again and make sure your channels are not private and can be visible from “Admin > Workspace > Rooms”

Additionally, check what channels and info you’re importing when ticking the boxes:

image
image698×386 3.95 KB

Best!

3

The issue ended up being that the export I received included a folder “between” the archive and the actual files. It seems the import process understood that messages were present by searching for them recursively, but didn’t understand that this folder existed.

The following archive layout could not (and possibly still cannot) be imported:

  • archive.zip
    • archive/
      • channel1/
      • channel2/
      • channeln/
      • users.json
      • integration_logs.json
      • channels.json

The following archive layout could be imported:

  • archive.zip
    • channel1/
    • channel2/
    • channeln/
    • users.json
    • integration_logs.json
    • channels.json

There may be an opportunity to improve the import process or documentation to either allow imports of archives structured like this or make it clear archives should not have leading root directories in them. I would imagine exports structured this way are not uncommon.

4

Hi @qskwood thanks for sharing solution and the feedback, we’ll take it into consideration.

Additionally, if you notice in the documentation, it’s already considered:

If facing difficulties during the upload phase of the import process, try copying the archive directly to the uploads folder using other means such as scp. Then select Server File Path as the File Type.

But we’ll improve it to make it clear about the Directory between files.