Removing unassigned nodes
To be completely removed from the infrastructure, the node must not be a part of the storage cluster.
Prerequisites
- The node is removed from the storage cluster, as described in Releasing nodes from the storage cluster.
To remove a node from the infrastructure
Admin panel
- Select the unassigned node on the Infrastructure > Nodes screen, and then click Remove.
-
For security purposes, clean up the node certificates and identity:
-
For an online node, run the following commands:
# /usr/libexec/vstorage-ui-agent/bin/generate_id.sh --rewrite-ids # rm -rf /var/lib/pgsql/* # rm -rf /usr/libexec/vstorage-ui-backend/backups/* # rm -rf /usr/libexec/vstorage-ui-backend/etc/backend.cfg
-
For an offline node, run the following commands once it becomes available again:
# rm -rf /usr/libexec/vstorage-ui-backend/ca # rm -rf /etc/nginx/ssl # /usr/libexec/vstorage-ui-agent/bin/generate_id.sh --rewrite-ids # rm -rf /var/lib/pgsql/* # rm -rf /usr/libexec/vstorage-ui-backend/backups/* # rm -rf /usr/libexec/vstorage-ui-backend/etc/backend.cfg
After such a cleanup, the only way to add the node back to the cluster is by re-installing Virtuozzo Hybrid Infrastructure on it from scratch.
-
Command-line interface
-
Unregister the node from Virtuozzo Hybrid Infrastructure by using the
vinfra node forget
command. For example, to unregister the nodenode005
, run:# vinfra node forget node005
-
For security purposes, clean up the node certificates and identity:
-
For an online node, run the following commands:
# /usr/libexec/vstorage-ui-agent/bin/generate_id.sh --rewrite-ids # rm -rf /var/lib/pgsql/* # rm -rf /usr/libexec/vstorage-ui-backend/backups/* # rm -rf /usr/libexec/vstorage-ui-backend/etc/backend.cfg
-
For an offline node, run the following commands when it becomes online:
# rm -rf /usr/libexec/vstorage-ui-backend/ca # rm -rf /etc/nginx/ssl # /usr/libexec/vstorage-ui-agent/bin/generate_id.sh --rewrite-ids # rm -rf /var/lib/pgsql/* # rm -rf /usr/libexec/vstorage-ui-backend/backups/* # rm -rf /usr/libexec/vstorage-ui-backend/etc/backend.cfg
After such a cleanup, the only way to add the node back to the cluster is by re-installing Virtuozzo Hybrid Infrastructure on it from scratch.
-