Difference between revisions of "Trove/disk-space-coefficient"
< Trove
(→Justification) |
(→Public API) |
||
| Line 30: | Line 30: | ||
=== Public API === | === Public API === | ||
| + | |||
| + | ==== New routes ==== | ||
No changes | No changes | ||
| + | |||
| + | ==== Modification to existing ==== | ||
| + | No changes | ||
| + | |||
| + | ==== Behaviour ==== | ||
| + | API will raize HTTP Bad Request (code 400) if mentioned storage size is not enough to pull and/or unpack backup. | ||
=== Internal API === | === Internal API === | ||
Revision as of 08:35, 21 July 2014
Contents
Disk space coefficient
Description
- This feature is being proposed as part of backup size validation process.
Justification/Benefits
Justification
- For certain reasons not all native backup tools are able to do data streaming (like innobackupex, mysqldump). So when restore process gets initialized there's possible case when proposed by user disk space(volume, flavor, ephemeral disk) has not enough space to pull and unpack backup from its storage.
Benefits
- By adding specific per-datastore coefficient deployers are able to define how much times bigger given instance storage should be to be able to accomplish restoring.
Impacts
- From user perspective. API service could raise certain exceptions if given instance storage is not enough to pull and unpack given backup.
Configuration
- Configuration parameters are API service specific
| Option group | Option name | Option type | Default |
|---|---|---|---|
| per_datastore | storage_space_coefficient | Integer | 1 |
Database
No changes
Public API
New routes
No changes
Modification to existing
No changes
Behaviour
API will raize HTTP Bad Request (code 400) if mentioned storage size is not enough to pull and/or unpack backup.
Internal API
From trove-api to trove-taskamanger
No changes
From trove-taskamanger to trove-guestagent
No changes
