VMware Blockchain client node restore fails

Products

• VMware
• VMware Blockchain

Issue/Introduction

Symptoms:

• You are using the instructions at Restore Client Node from the Full Backup Data on vSphere to restore a VMware Blockchain (VMBC) client node.
• While executing step 7d (pgbackrest --stanza=daml-indexdb --log-level-console=info restore), you see messages similar to the following:

WARN: unknown group 'root' in backup manifest mapped to current group
ERROR: [088]: unable to set ownership for '/var/lib/postgresql/data/recovery.conf': [1] Operation not permitted

• If you attempt to complete the rest of the restore process on the client node, the daml_index_db container will fail to start and the client node will not be operational.

Environment

VMware Blockchain 1.6
VMware Blockchain 1.7

Cause

This issue is caused by a file system permission issue in the daml_index_db container on the VMBC client node.

Resolution

This is a known issue affecting VMware Blockchain. There is currently no resolution.

Workaround:
Before configuring your backup, per the instructions at Back-Up Client Node on vSphere , perform the following steps:

1. Log in to the client node as the vmbc user.
2. Issue the following command to open a shell to the daml_index_db container:

sudo docker exec -it daml_index_db bash

Note: Your prompt should look similar to root@16aa15d0f24f:~

3. Issue the following command to change the user and group ownership on the /var/lib/postgresql folder and subfolders:

chown -R postgres:postgres /var/lib/postgresql

4. Type exit to leave the daml_index_db container shell.
5. Repeat these steps on any other client nodes.

Note: If you re-deploy any client nodes, or add additional client nodes, you will need to perform these same steps on those client nodes prior to configuring backups.

Additional Information

Back-Up Client Node on vSphere
Restore Client Node from the Full Backup Data on vSphere

Impact/Risks:
VMBC client node restore operations will fail.