Object Storage Configuration

A centralised volume on a bucket is a mandatory requirement to configure a Redundant with User Mail Replica scenario.

Before creating and configuring a volume, you should already have an S3-like service at your disposal, either within your infrastructure or purchased from a third-party. Indeed, the commands in this page will only connect to the bucket, then create a centralised volume and configure it for the use with Carbonio.

In this section are showed as examples the commands to set up MinIO and S3 volumes. In case you prefer to create a volume on other types of bucket (for example, Alibaba, EMC, and other), you can find the precise command and all options in the corresponding page:

Alibaba Ceph Cloudian Custom S3 EMC FileBlob OpenIO S3 Scality S3 Swift

After you create the volumes, remember to set them as primary and centralised, by using the commands in Section Define Centralised Volume.

Note

All commands below must be executed as the zextras user. Remember to replace all the example values with values suitable with your infrastructure.

Connect Bucket to Carbonio

Minio

If you use MinIO, the command below will create a bucket.

Hint

Do not forget to create a bucket in MinIO before executing the command. The remoteBucketName is the name given in MinIO.

zextras$ carbonio core doCreateBucket Type remoteBucketName username\accesskey password\secretKey url localBucketName prefix somePrefix

For example:

zextras$ carbonio core doCreateBucket MINIO carbonio-ha admin MyBestPassword http://162.19.69.216:9000 ha-primary

This command outputs a number of information about the bucket. Write down the Bucket ID, because you will need it when creating a centralised primary volume: it will be called ID_FROM_PREVIOUS_COMMAND in Section Define Centralised Volume.

S3

If you use an S3 Object Storage, this command creates a bucket.

zextras$ carbonio core doCreateBucket Type remoteBucketName accesskey secretKey region us-east-1 ha-primary

For example:

zextras$ carbonio core doCreateBucket S3 bucket_name carbonio-ha accesskey secretkey region us-east-1 ha-primary

This command outputs a number of information about the bucket. Write down the Bucket ID, because you will need it when creating a centralised primary volume: it will be called ID_FROM_PREVIOUS_COMMAND in Section Define Centralised Volume.

Define Centralised Volume

In order to set up the volume as a centralised one, execute the following two commands, replacing TYPE with the type of the bucket, which is MINIO or S3 in our example, or the type you used.

Remember also to use the bucket ID from the output of the previous commands (called ID_FROM_PREVIOUS_COMMAND in the command below)

zextras$ carbonio powerstore doCreateVolume  TYPE centralized-ha primary ID_FROM_PREVIOUS_COMMAND centralized true

Finally, set the volume as current, running on every Node with the Mailstore & Provisioning Component the command

zextras$ carbonio powerstore doUpdateVolume TYPE centralized-ha current_volume true