000038539 - NWSETUP-TUI mistakenly run on the RSA NetWitness Platform Admin Server

Document created by RSA Customer Support Employee on Mar 10, 2020
Version 1Show Document
  • View in full screen mode

Article Content

Article Number000038539
Applies ToRSA Product Set: NetWitness Platform
RSA Product/Service Type: NetWitness Platform/Nw Admin Server
RSA Version/Condition: 11.3.x
Platform: CentOS
O/S Version: EL7
IssueUpgrade fails for any node.

Running upgrade using CLI on the NetWitness Admin Server results in a similar error as below:

[2020-02-28T23:11:21+00:00] <98045> (ERROR) Command line upgrade can only be executed on the NW Server (please re-run from the NW Server).

Missing /var/netwitness/common/repo/ folder - no repo files

Running orchestration-cli update-admin-node on the NetWitness Admin server fails with errors.
Causenwsetup-tui was run by mistake on the NetWitness Admin node

Usually happens when user has the habit of doing ssh from the head unit connecting to the other hosts. Sometimes thinking they are already ssh'd to the target host, and run nwsetup-tui to initiate the node configuration. 
ResolutionOn the NetWitness Admin server (nw-node-zero):

Re-create the repo folders, and copy/unzip the update files.

Edit the /etc/netwitness/platform/nw-node-type to show 'node-zero'

Ensure in mongo db that the nw-node-zero host UUID has the parameter -  "node-zero" : true

Run orchestration-cli-client --update-admin-node until there are no errors and it completes successfully.
WorkaroundTo re-create the repo folder and files:

Download the corresponding NetWitness update packages, for example, netwitness-, netwitness- from RSA Link.
-up to and depending on your current version
SCP/Copy the zip files to a temp directory on your NetWitness Admin with enough filesystem disk space
Manually create the corresponding repository directories:

mkdir -p /var/netwitness/common/repo/
mkdir -p /var/netwitness/common/repo/

mkdir -p /var/netwitness/common/repo/
mkdir -p /var/netwitness/common/repo/

Unzip the NetWitness update packages to their corresponding repository directories.
Using vi, edit the /etc/netwitness/platform/nw-node-type file to show 'node-zero' instead of 'node-x'
Ensure in mongo db that the nw-node-zero host UUID has the parameter -  "node-zero" : true


On node-zero ssh:
# cat /etc/salt/minion   ---note the UUID of node-zero
# mongo admin -u deploy_admin 
Enter password:
> show dbs
> use orchestration-server
> show collections
>  db.host.find({_id:"ba847be4-afca-4df4-beca-e6df7ac3a228"}).pretty()
        "_id" : "ba847be4-afca-4df4-beca-e6df7ac3a228",
        "hostname" : "",
        "displayName" : "nwadmin1",
        "version" : {
                "major" : 11,
                "minor" : 3,
                "servicePack" : 1,
                "patch" : 1,
                "snapshot" : false,
                "rawVersion" : ""
        "thirdParty" : false,
        "installedServices" : [
        "meta" : {
                "node-zero" : true
        "_class" : "com.rsa.asoc.orchestration.host.HostEntity"
> exit

Note: If value "node-zero" : false" is seen, contact RSA Support for assistance.

Run the command 'orchestration-cli-client --update-admin-node' until there are no errors and it completes successfully.