docker
/
docs
mirror of https://github.com/docker/docs.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Address backup/restore comments and add minor flow details 

Signed-off-by: Alex Mavrogiannis <alex.mavrogiannis@docker.com>
This commit is contained in:
Alex Mavrogiannis 2017-02-16 15:57:45 -08:00 committed by Joao Fernandes
parent dad2695eef
commit 5e2e92f7a9
3 changed files with 36 additions and 28 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

56
datacenter/ucp/2.1/guides/admin/backups-and-disaster-recovery.md
View File

@ -15,20 +15,20 @@ The next step is creating a backup policy and disaster recovery plan.
As part of your backup policy you should regularly create backups of UCP. As part of your backup policy you should regularly create backups of UCP.
To create a UCP backup, you may use the `{{ page.docker_image }} backup` command To create a UCP backup, you can run the `{{ page.docker_image }} backup` command
against a single UCP manager, according to the instructions in the next section. on a single UCP manager. This command creates a tar archive with the
This command creates a tar archive with the contents of all the [volumes used by contents of all the [volumes used by UCP](../architecture.md) to persist data
UCP](../architecture.md) to persist data and streams it to stdout. and streams it to stdout.
You only need to run the backup command on a single UCP manager node. Since UCP You only need to run the backup command on a single UCP manager node. Since UCP
stores the same data on all manager nodes, you do not need to capture periodic stores the same data on all manager nodes, you only need to take periodic
backups from more than one manager node. backups of a single manager node.
To create a consistent backup, the backup command temporarily stops the UCP To create a consistent backup, the backup command temporarily stops the UCP
containers running on the node where the backup is being performed. User containers running on the node where the backup is being performed. User
resources, such as services, containers and stacks are not affected by this resources, such as services, containers and stacks are not affected by this
operation and will continue operating as expected. Any long-lasting `exec`, operation and will continue operating as expected. Any long-lasting `exec`,
`logs`, `events` or `attach` operations against the affected manager node will `logs`, `events` or `attach` operations on the affected manager node will
be disconnected. be disconnected.
Additionally, if UCP is not configured for high availability, you will be Additionally, if UCP is not configured for high availability, you will be
@ -50,10 +50,12 @@ verify its contents:
# Create a backup, encrypt it, and store it on /tmp/backup.tar # Create a backup, encrypt it, and store it on /tmp/backup.tar
$ docker run --rm -i --name ucp \ $ docker run --rm -i --name ucp \
-v /var/run/docker.sock:/var/run/docker.sock \ -v /var/run/docker.sock:/var/run/docker.sock \
{{ page.docker_image }} backup --interactive /tmp/backup.tar {{ page.docker_image }} backup --interactive > /tmp/backup.tar
# Ensure the backup is a valid tar and list its contents # Ensure the backup is a valid tar and list its contents
$ tar --list /tmp/backup.tar # In a valid backup file, over 100 files should appear in the list
# and the `./ucp-node-certs/key.pem` file should be present
$ tar --list -f /tmp/backup.tar
``` ```
A backup file may optionally be encrypted using a passphrase, as in the A backup file may optionally be encrypted using a passphrase, as in the
@ -75,12 +77,12 @@ $ gpg --decrypt /tmp/backup.tar | tar --list
The restore command can be used to create a new UCP cluster from a backup file. The restore command can be used to create a new UCP cluster from a backup file.
After the restore operation is complete, the following data will be recovered After the restore operation is complete, the following data will be recovered
from the backup file: from the backup file:
* Users, Teams and Permissions. * Users, teams and permissions.
* All UCP Configuration options available under `Admin Settings`, such as the * All UCP configuration options available under `Admin Settings`, such as the
DDC Subscription license, scheduling options, Content Trust and authentication DDC subscription license, scheduling options, Content Trust and authentication
backends. backends.
There restore operation can be performed in any of three environments: There are two ways to restore a UCP cluster:
* On a manager node of an existing swarm, which is not part of a UCP * On a manager node of an existing swarm, which is not part of a UCP
installation. In this case, a UCP cluster will be restored from the backup. installation. In this case, a UCP cluster will be restored from the backup.
* On a docker engine that is not participating in a swarm. In this case, a new * On a docker engine that is not participating in a swarm. In this case, a new
@ -90,12 +92,12 @@ In order to restore an existing UCP installation from a backup, you will need to
first uninstall UCP from the cluster by using the `uninstall-ucp` command first uninstall UCP from the cluster by using the `uninstall-ucp` command
The example below shows how to restore a UCP cluster from an existing backup The example below shows how to restore a UCP cluster from an existing backup
file: file, presumed to be located at `/tmp/backup.tar`:
```bash ```bash
$ docker run --rm -i --name ucp \ $ docker run --rm -i --name ucp \
-v /var/run/docker.sock:/var/run/docker.sock \ -v /var/run/docker.sock:/var/run/docker.sock \
{{ page.docker_image }} restore < backup.tar {{ page.docker_image }} restore < /tmp/backup.tar
``` ```
If the backup file is encrypted with a passphrase, you will need to provide the If the backup file is encrypted with a passphrase, you will need to provide the
@ -118,7 +120,7 @@ $ docker run --rm -i --name ucp \
{{ page.docker_image }} restore -i {{ page.docker_image }} restore -i
``` ```
## Disaster Recovery ## Disaster recovery
In the event where half or more manager nodes are lost and cannot be recovered In the event where half or more manager nodes are lost and cannot be recovered
to a healthy state, the system is considered to have lost quorum and can only be to a healthy state, the system is considered to have lost quorum and can only be
@ -129,16 +131,22 @@ no loss of running services or configuration data. To properly protect against
manager failures, the system should be configured for [high availability](configure/set-up-high-availability.md). manager failures, the system should be configured for [high availability](configure/set-up-high-availability.md).
1. On one of the remaining manager nodes, perform `docker swarm init 1. On one of the remaining manager nodes, perform `docker swarm init
--force-new-cluster`. This will instantiate a new single-manager swarm by --force-new-cluster`. You may need to specify also need to specify an
recovering as much state as possible from the existing manager. This is a `--advertise-addr` parameter which is equivalent to the `--host-address`
disruptive operation and any existing tasks will be either terminated or parameter of the `docker/ucp install` operation. This will instantiate a new
suspended. single-manager swarm by recovering as much state as possible from the
existing manager. This is a disruptive operation and existing tasks may be
either terminated or suspended.
2. Obtain a backup of one of the remaining manager nodes if one is not already 2. Obtain a backup of one of the remaining manager nodes if one is not already
available. available.
3. Perform a restore operation on the recovered swarm manager node. 3. If UCP is still installed on the cluster, uninstall UCP using the
4. For all other nodes of the cluster, perform a `docker swarm leave --force` `uninstall-ucp` command.
and then a `docker swarm join` operation with the cluster's new join-token. 4. Perform a restore operation on the recovered swarm manager node.
5. Wait for all nodes of the swarm to become healthy UCP nodes. 5. Log in to UCP and browse to the nodes page, or use the CLI `docker node ls`
command.
6. If any nodes are listed as `down`, you'll have to manually [remove these
nodes](../configure/scale-your-cluster.md) from the cluster and then re-join
them using a `docker swarm join` operation with the cluster's new join-token.
## Where to go next ## Where to go next

2
datacenter/ucp/2.1/guides/architecture.md
View File

@ -95,7 +95,7 @@ Docker UCP uses these named volumes to persist data in all nodes where it runs:
|:----------------------------|:-----------------------------------------------------------------------------------------| |:----------------------------|:-----------------------------------------------------------------------------------------|
| ucp-auth-api-certs | Certificate and keys for the authentication and authorization service | | ucp-auth-api-certs | Certificate and keys for the authentication and authorization service |
| ucp-auth-store-certs | Certificate and keys for the authentication and authorization store | | ucp-auth-store-certs | Certificate and keys for the authentication and authorization store |
| ucp-auth-store-data | Data of the authentication and authorization store, replicated across managers | ucp-auth-store-data | Data of the authentication and authorization store, replicated across managers |
| ucp-auth-worker-certs | Certificate and keys for authentication worker | | ucp-auth-worker-certs | Certificate and keys for authentication worker |
| ucp-auth-worker-data | Data of the authentication worker | | ucp-auth-worker-data | Data of the authentication worker |
| ucp-client-root-ca | Root key material for the UCP root CA that issues client certificates | | ucp-client-root-ca | Root key material for the UCP root CA that issues client certificates |

2
datacenter/ucp/2.1/guides/get-support.md
View File

@ -26,7 +26,7 @@ support dump:
2. On the top-right menu, **click your username**, and choose **Support Dump**. 2. On the top-right menu, **click your username**, and choose **Support Dump**.
An archive will be downloaded by your browser after a brief time interval. An archive will be downloaded by your browser after a brief time interval.
If the User Interface is not accessible, you may perform the following number of If the user interface is not accessible, you may perform the following number of
steps instead to obtain a single-node version of the support dump: steps instead to obtain a single-node version of the support dump:
1. Obtain direct CLI access to the docker daemon on a UCP manager node. 1. Obtain direct CLI access to the docker daemon on a UCP manager node.