This repository has been archived by the owner on Jan 10, 2019. It is now read-only.
-
Notifications
You must be signed in to change notification settings - Fork 68
Add docs for multi-dc support #130
Merged
+189
−2
Merged
Changes from all commits
Commits
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,178 @@ | ||
| --- | ||
| title: Multi Datacenter | ||
| --- | ||
|
|
||
| # Multi Datacenter | ||
|
|
||
| Beginning with Cassandra on Mesos version 0.2.0, Cassandra on Mesos can be configured so that multiple datacenters are aware of each other. | ||
|
|
||
| See [Deploying Cassandra across Multiple Data Centers](http://www.datastax.com/dev/blog/deploying-cassandra-across-multiple-data-centers) for motivations regarding running Cassandra in multiple datacenters. | ||
|
|
||
| ## How it works | ||
|
|
||
| Cassandra on Mesos can be configured to provide datacenter and rack information to Cassandra upon starting each server. | ||
|
|
||
| Cassandra on Mesos can also be configured to query other Cassandra instances to provide seed sets across multiple clusters so that each of the clusters can coordinate and replicate data. | ||
|
|
||
|
|
||
| ## Requirements | ||
|
|
||
| 1. Every cluster must be able to reach the IP addresses of every other cluster. | ||
| 2. All clusters must have the same cluster name. | ||
| 3. Datacenters must be defined before starting the framework. New datacenters cannot be added after a cluster has initialized. | ||
| 4. Node sizes and counts must be the same across all clusters. **Note:** Cassandra on Mesos does not enforce or validate this, so be certain that this is configured correctly. | ||
|
|
||
| ## Configuration | ||
|
|
||
| ### DC and Rack | ||
|
|
||
| Configuration is provided via environment variables. | ||
|
|
||
| ``` | ||
| # Value to be used as the default DC name | ||
| CASSANDRA_DEFAULT_DC=DC1 | ||
| # Value to be used as the default RACK name | ||
| CASSANDRA_DEFAULT_RACK=RAC1 | ||
| ``` | ||
|
|
||
| ### Other instances of Cassandra on Mesos | ||
|
|
||
| Cassandra on Mesos is informed of other Cassandra on Mesos instances running in other Mesos clusters by | ||
| environment variables in the format `CASSANDRA_EXTERNAL_DC_<dcName>=<cassandraOnMesosUrl>`. | ||
|
|
||
| ## Example on Mesos | ||
|
|
||
| In this example, three datacenters dc0, dc1, dc2 are running three different Mesos clusters. | ||
|
|
||
| The following configuration would be used for each instance of Cassandra on Mesos: | ||
|
|
||
| ##### dc0 | ||
| ``` | ||
| CASSANDRA_CLUSTER_NAME=multi-dc | ||
| CASSANDRA_DEFAULT_DC=dc0 | ||
| CASSANDRA_EXTERNAL_DC_dc1=http://multi-dc.cassandra.marathon.mesos1:10001 | ||
| CASSANDRA_EXTERNAL_DC_dc2=http://multi-dc.cassandra.marathon.mesos2:10002 | ||
| ``` | ||
|
|
||
| ##### dc1 | ||
| ``` | ||
| CASSANDRA_CLUSTER_NAME=multi-dc | ||
| CASSANDRA_DEFAULT_DC=dc1 | ||
| CASSANDRA_EXTERNAL_DC_dc0=http://multi-dc.cassandra.marathon.mesos0:10000 | ||
| CASSANDRA_EXTERNAL_DC_dc2=http://multi-dc.cassandra.marathon.mesos2:10002 | ||
| ``` | ||
|
|
||
| ##### dc2 | ||
| ``` | ||
| CASSANDRA_CLUSTER_NAME=multi-dc | ||
| CASSANDRA_DEFAULT_DC=dc2 | ||
| CASSANDRA_EXTERNAL_DC_dc0=http://multi-dc.cassandra.marathon.mesos0:10000 | ||
| CASSANDRA_EXTERNAL_DC_dc1=http://multi-dc.cassandra.marathon.mesos1:10001 | ||
| ``` | ||
|
|
||
| 1. Ensure there are TCP routes allowing for traffic flow between each of the Mesos clusters. | ||
| 2. Start each cluster with the necessary configuration. | ||
| 3. Wait for each cluster to allocate its seed nodes. | ||
| 4. After the seeds have been allocated, restart all of the nodes that have already been started. Send a `POST` to the http API at `/cluster/rolling-restart/start` and the framework will restart each of the nodes so that the new seed set will take effect. | ||
|
|
||
| Your clusters should now be able to replicate data between each other. | ||
|
|
||
| ## Example on DCOS | ||
|
|
||
| In this example, three datacenters dcos0, dcos1, dcos2 are running three different DCOS clusters. | ||
|
|
||
| To install each instance of Cassandra on DCOS using the DCOS CLI: | ||
|
|
||
| 1. On DCOS Cluster 0, install with `dcos package install cassandra --options=dcos0.json` | ||
|
|
||
| ```json | ||
| { | ||
| "cassandra": { | ||
| "cluster-name": "multi-dc", | ||
| "resources": { | ||
| "cpus": 1.0, | ||
| "mem": 2048, | ||
| "disk": 6144 | ||
| }, | ||
| "dc": { | ||
| "default-dc": "dcos0", | ||
| "default-rack": "rack0", | ||
| "external-dcs": [ | ||
| { | ||
| "name": "dcos1", | ||
| "url": "http://dcos1/service/cassandra.multi-dc" | ||
| }, | ||
| { | ||
| "name": "dcos2", | ||
| "url": "http://dcos2/service/cassandra.multi-dc" | ||
| } | ||
| ] | ||
| } | ||
| } | ||
| } | ||
| ``` | ||
|
|
||
| 2. On DCOS Cluster 1, install with `dcos package install cassandra --options=dcos1.json` | ||
|
|
||
| ```json | ||
| { | ||
| "cassandra": { | ||
| "cluster-name": "multi-dc", | ||
| "resources": { | ||
| "cpus": 1.0, | ||
| "mem": 2048, | ||
| "disk": 6144 | ||
| }, | ||
| "dc": { | ||
| "default-dc": "dcos1", | ||
| "default-rack": "rack1", | ||
| "external-dcs": [ | ||
| { | ||
| "name": "dcos0", | ||
| "url": "http://dcos0/service/cassandra.multi-dc" | ||
| }, | ||
| { | ||
| "name": "dcos2", | ||
| "url": "http://dcos2/service/cassandra.multi-dc" | ||
| } | ||
| ] | ||
| } | ||
| } | ||
| } | ||
| ``` | ||
|
|
||
| 3. On DCOS Cluster 2 install with `dcos package install cassandra --options=dcos2.json` | ||
|
|
||
| ```json | ||
| { | ||
| "cassandra": { | ||
| "cluster-name": "multi-dc", | ||
| "resources": { | ||
| "cpus": 1.0, | ||
| "mem": 2048, | ||
| "disk": 6144 | ||
| }, | ||
| "dc": { | ||
| "default-dc": "dcos2", | ||
| "default-rack": "rack2", | ||
| "external-dcs": [ | ||
| { | ||
| "name": "dcos0", | ||
| "url": "http://dcos0/service/cassandra.multi-dc" | ||
| }, | ||
| { | ||
| "name": "dcos1", | ||
| "url": "http://dcos1/service/cassandra.multi-dc" | ||
| } | ||
| ] | ||
| } | ||
| } | ||
| } | ||
| ``` | ||
|
|
||
| 4. Wait for each cluster to allocate its seed nodes. | ||
| 5. After the seeds have been allocated, restart all of the nodes that have already been started. Send a `POST` to the http API at `/cluster/rolling-restart/start` and the framework will restart each of the nodes so that the new seed set will take effect. | ||
|
|
||
| Your clusters should now be able to replicate data between each other. | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It'd be really useful to add a high level explanation of why multiple datacenter support is useful/interesting. (You could also link to: http://www.datastax.com/dev/blog/deploying-cassandra-across-multiple-data-centers)
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd reword to: "As of Cassandra 0.2.0, you can configure Cassandra on Mesos so that multiple datacenters are aware of each other."