Skip to content

Commit 47d101d

Browse files
andy-stark-redisandy-stark-redis
authored
Merge pull request #1089 from redis/DOC-4753-rdi-upgrade-guide
DOC-4753 added RDI upgrade guide
2 parents cfe66d4 + 4352d97 commit 47d101d

File tree

2 files changed

+161
-4
lines changed

2 files changed

+161
-4
lines changed

‎content/integrate/redis-data-integration/installation/_index.md

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -1,23 +1,23 @@
11
---
2-
Title: Installation
2+
Title: Install and upgrade
33
aliases: /integrate/redis-data-integration/ingest/installation/
44
alwaysopen: false
55
categories:
66
- docs
77
- integrate
88
- rs
99
- rdi
10-
description: Learn how to install RDI
10+
description: Learn how to install and upgrade RDI
1111
group: di
1212
hideListLinks: false
13-
linkTitle: Installation
13+
linkTitle: Install/upgrade
1414
summary: Redis Data Integration keeps Redis in sync with the primary database in near
1515
real time.
1616
type: integration
1717
weight: 2
1818
---
1919

20-
The guides in this section explain the options you have for installing RDI.
20+
The guides in this section explain the options you have for installing and upgrading RDI.
2121
Before you use RDI, you must also configure your source database to use the CDC connector. See the
2222
[Prepare source databases]({{< relref "/integrate/redis-data-integration/data-pipelines/prepare-dbs" >}})
2323
section to learn how to do this.

‎content/integrate/redis-data-integration/installation/upgrade.md

Lines changed: 157 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,157 @@
1+
---
2+
Title: Upgrading RDI
3+
alwaysopen: false
4+
categories:
5+
- docs
6+
- integrate
7+
- rs
8+
- rdi
9+
description: Learn how to upgrade an existing RDI installation
10+
group: di
11+
hideListLinks: false
12+
linkTitle: Upgrade
13+
summary: Redis Data Integration keeps Redis in sync with the primary database in near
14+
real time.
15+
type: integration
16+
weight: 3
17+
---
18+
19+
## Upgrading a VM installation
20+
21+
Follow the steps below to upgrade an existing
22+
[VM installation]({{< relref "/integrate/redis-data-integration/installation/install-vm" >}})
23+
of RDI:
24+
25+
1. Download the RDI installer from the
26+
[Redis download center](https://cloud.redis.io/#/rlec-downloads)
27+
(under the *Modules, Tools & Integration* dropdown) and extract it to your
28+
preferred installation folder.
29+
30+
1. Go to the installation folder:
31+
32+
```bash
33+
cd rdi_install/$RDI_VERSION
34+
```
35+
36+
1. Run the `upgrade` script as a privileged user:
37+
38+
```bash
39+
sudo ./upgrade.sh
40+
```
41+
42+
### Recovering from failure during a VM upgrade
43+
44+
If the previous version is v1.4.4 or later, go to the `rdi_install/<PREVIOUS_VERSION>`
45+
directory and run `sudo ./upgrade.sh`, as described in the section
46+
[Upgrading a VM installation](#upgrading-a-vm-installation) above.
47+
48+
If the version you are replacing is earlier than v1.4.4, follow these steps:
49+
50+
1. Run `redis-di --version` to check the current version.
51+
52+
If the version is the new one, copy the previous version
53+
of the RDI CLI to `/usr/local/bin` with the following command:
54+
55+
```bash
56+
sudo cp rdi_install/<PREVIOUS_VERSION>/deps/rdi-cli/<OS>/redis-di usr/local/bin
57+
```
58+
59+
1. Check that the CLI version is correct by running `redis-di --version`.
60+
61+
Then, go to the `rdi_install/<PREVIOUS_VERSION>` directory and run the
62+
following command;
63+
64+
```bash
65+
sudo redis-di upgrade --rdi-host <RDI_REDIS_HOST> --rdi-port <RDI_REDIS_PORT>
66+
```
67+
68+
{{< note >}}If the `collector-source` or the `processor` are not in the `Running` state during
69+
the upgrade, you must run `redis-di deploy` and check again that they are both in the
70+
`Running` state.
71+
{{< /note >}}
72+
73+
### Upgrading a VM installation with High availability
74+
75+
If there is an active pipeline, the upgrade process will involve upgrading RDI on the active
76+
VM first which will cause downtime for the collector-source (see
77+
[Upgrade a VM installation](#upgrade-a-vm-installation) above). Afterwards, the passive
78+
VM will be upgraded. Switching over won't eliminate the downtime because switching between
79+
VMs also requires a about a minute of downtime.
80+
81+
## Upgrading a Kubernetes installation
82+
83+
Follow the steps below to upgrade an existing
84+
[Kubernetes]({{< relref "/integrate/redis-data-integration/installation/install-k8s" >}})
85+
installation of RDI:
86+
87+
1. Download the new versions of the images, if you are using a private registry:
88+
89+
```bash
90+
docker pull redis/rdi-processor:tagname
91+
docker pull redis/rdi-operator:tagname
92+
docker pull redis/rdi-api:tagname
93+
docker pull redis/rdi-monitor:tagname
94+
docker pull redis/rdi-collector-initializer
95+
docker pull redis/rdi-collector-api
96+
```
97+
98+
1. Download the RDI helm chart tar file from the
99+
[Redis download center](https://cloud.redis.io/#/rlec-downloads).
100+
101+
1. Run the `helm upgrade` command:
102+
103+
```bash
104+
helm upgrade [RELEASE_NAME] [CHART]
105+
```
106+
107+
Note that you don't need to
108+
[deploy]({{< relref "/integrate/redis-data-integration/data-pipelines/deploy" >}})
109+
again after this step.
110+
111+
### Verifying the upgrade
112+
113+
Check the upgrade with the following command:
114+
115+
```bash
116+
sudo k3s kubectl get all -n <namespace>
117+
```
118+
119+
You should find that all the pods are running (they will have `1/1` in the `READY` column of the
120+
command's output).
121+
Check for any pods that don't have `1/1` in the `READY` column (which is the second
122+
column). For example, the pod below has `0/1` in the second column, which indicates the
123+
deployment hasn't worked:
124+
125+
```bash
126+
<pod_name> 0/1 CrashLoopBackOff 1881 (91s ago) 6d17h
127+
```
128+
129+
You can also check that the latest version is running using the following command on one of
130+
the pods:
131+
132+
```bash
133+
sudo k3s kubectl describe <pod_name> -n <namespace>
134+
```
135+
136+
Search for the image tag `Image: docker.io/redis/<pod_name>:<version/image_tag>`
137+
in the command's output to verify the version.
138+
139+
If you find that the upgrade hasn't worked for any reason, then run the `helm upgrade`
140+
command again (as described in the section
141+
[Upgrading a Kubernetes installation](#upgrading-a-kubernetes-installation) above),
142+
but this time with the previous version you were upgrading from. This will restore your
143+
previous working state.
144+
145+
## What happens during the upgrade?
146+
147+
The upgrade process replaces the current RDI components with the new versions:
148+
149+
- Firstly, the control plane components are replaced. At this point, the pipeline
150+
is still active but monitoring will be disconnected.
151+
152+
- Secondly, the pipeline data path components are replaced with the new versions.
153+
If a pipeline is active while upgrading, the `collector-source` will be restarted
154+
as a result of restarting the `collector-initializer`. The pipeline will pause for
155+
about two minutes but it will catch up very quickly after restarting.
156+
The pipeline data and state are both stored in Redis, so data will never normally
157+
be lost during the downtime while upgrading.

0 commit comments

Comments
 (0)