FileConnect Enterprise: Upgrading from 3.0 to 3.1

  • Updated

📌 Who this is for: existing FileConnect Enterprise 3.0 customers moving to 3.1. New installations don't need this guide — see the FileConnect Enterprise v3.1 Setup guide instead.

 

This guide walks through upgrading an existing FCE 3.0 deployment to FCE 3.1. It assumes a single-node Elasticsearch running on the same machine as FCE. If you run a multi-node or external Elasticsearch, contact RecordPoint before you begin.

Existing 3.0 customers can keep running on the RecordPoint-supplied virtual machine (OVA) — you do not need to rebuild on a new host to move to 3.1. (From 3.1 onward RecordPoint no longer supplies or updates the OVA, but your existing VM continues to work.)
 

Before you begin

You will need:

  • The latest FCE 3.1 installer package (downloaded from the FCE Connector Configuration page in the RecordPoint Data Governance platform).
  • Root or sudo access on the machine hosting FCE.
  • The Elasticsearch and RabbitMQ usernames and passwords used by your FCE 3.0 deployment.
  • An access token for 3.1 — contact RecordPoint to obtain one exclusive to your account.
  • Your 3.0 installation updated to the latest available 3.0 version before upgrading.

Step 1 — Shut down FCE and back up

  1. Disable any repeating scan tasks and FCE policy schedules, and wait until all running operations have finished.
  2. Stop FCE services — either power down the virtual machine, or run:
 
   sudo /opt/recordpoint-connector-IndexBuilder/Deployment/fcectl.sh --stop --all
  1. Take a backup:
    • If FCE runs on a virtual machine: create a VM snapshot.
    • Otherwise: back up the Elasticsearch datastore.

Step 2 — Prepare the installer

  1. Power the machine back on (if you powered it down).
  2. Download the FCE 3.1 installer package from the FCE Connector Configuration page, copy it to the FCE machine, and unzip it.
  3. If your 3.0 system used the VirtualBox OVA, remove the guest additions:
 
   sudo dnf remove kmod-vbox-guest-additions

Step 3 — Back up and upgrade Diskover

  1. Stop services and back up the existing Diskover directory:
 
   sudo /opt/recordpoint-connector-IndexBuilder/Deployment/fcectl.sh --stop --all
   sudo systemctl stop diskoverd
   cd /opt
   sudo mv diskover/ diskoverBackup/
  1. Run the Diskover installer using the access token supplied by RecordPoint:
 
   sudo ./install-Diskover.sh --access-token "<name>:<token>"

⚠️ Review the output carefully. Do not continue if any critical failures occur.

  1. Re-apply your package exclusions so routine system updates don't upgrade managed components — see the FileConnect Enterprise v3.1 Setup guide, "Prevent unintended package upgrades."

Step 4 — Refresh and reconfigure

  1. Refresh services:
 
   sudo nginx -s reload
   sudo systemctl daemon-reload
   sudo systemctl start elasticsearch
  1. Reconfigure Diskover using the configuration process in the FileConnect Enterprise v3.1 Setup guide. Because this is an upgrade, reuse your existing deployment details:
    • On the Elasticsearch configuration page, enter your existing ES hostname, username, and password, and set HTTPS to ON. Copy the same settings to Diskover-Web.
    • For the licenses, copy the values from your backup:
 
     sudo gedit /opt/diskoverBackup/diskover.lic
 
 Paste the contents into the Diskover License field. Repeat for Diskover-Web:
 
     sudo gedit /var/www/diskover-web/src/diskover/diskover-web.lic
  • On the Message Queue page, enter your existing RabbitMQ password.
  • For the remaining pages, test (where available) and save.

Step 5 — Update services

  1. Restart your shell:
 
   exec $SHELL
  1. Edit the Diskover service definition:
 
   sudo gedit /etc/systemd/system/diskoverd.service

Update the ExecStart line to use the new Python environment:

 
   ExecStart=/opt/python-venv-diskover/bin/python3.11 diskoverd.py -n worker-%H
  1. Reload and restart services:
 
   sudo systemctl daemon-reload
   sudo systemctl start diskoverd celery
   sudo systemctl restart diskoverd diskover-admin
   sudo nginx -s reload
   sudo systemctl daemon-reload

Step 6 — Run the FCE installer

Run the FCE installer as described in the FileConnect Enterprise v3.1 Setup guide. This upgrades the FCE service from 3.0 to 3.1:

 
sudo ./install-FCE.sh --verbose

Confirm the upgrade by opening the FCE UI at http://<fcehost>:8080 — the Settings page displays the FCE version.

Troubleshooting

Kibana won't start. Open the Kibana config:

 
sudo gedit /etc/kibana/kibana.yml

Look for duplicated settings and comment out the first (top-down) instance of each of the following, then ensure the username/password lines are commented out:

 
server.host: "0.0.0.0"
elasticsearch.ssl.certificateAuthorities: [ "/etc/kibana/certs/http_ca.crt" ]
elasticsearch.username:
elasticsearch.password:

Then restart:

 
sudo systemctl restart kibana

Diskover Web UI doesn't display. Check the NGINX logs first:

 
sudo cat /var/log/nginx/error.log

A common cause is a broken certificate symlink. Check it:

 
readlink -f /etc/pki/ca-trust/source/anchors/http_ca.crt

If that returns nothing, the symlink is broken. Replace it with the real certificate and refresh services:

 
sudo rm -f /etc/pki/ca-trust/source/anchors/http_ca.crt
sudo cp /etc/elasticsearch/certs/http_ca.crt /etc/pki/ca-trust/source/anchors/http_ca.crt
sudo chmod 644 /etc/pki/ca-trust/source/anchors/http_ca.crt
sudo update-ca-trust
sudo nginx -s reload
sudo systemctl daemon-reload
sudo /opt/recordpoint-connector-IndexBuilder/Deployment/fcectl.sh --restart --all

Then reload the Diskover-Web UI at http://localhost:8000.

FAQ

I'm running 3.0 on the RecordPoint-supplied OVA. Can I upgrade this system? 

Yes. You can keep the virtual machine that came with 3.0 and upgrade it in place using the steps above. Note that from 3.1 onward RecordPoint no longer supplies or updates the OVA, so the 3.0 OVA image itself will not receive further updates.

I'm running 3.0 without the OVA (a bare-metal / host install). How do I upgrade? 

No special steps — follow the upgrade above.

I'm on the 3.0 OVA and want to move to a different host as part of the upgrade. Can I? 

Yes. At a high level: stand up a new Elasticsearch 8 system, migrate your data from the 3.0 Elasticsearch into it, then run an advanced deployment of FCE pointed at the new Elasticsearch. Contact RecordPoint for the FCE Advanced Deployment guide.

Was this article helpful?

0 out of 0 found this helpful

Have more questions? Submit a request