Skip to main content

 

Druva Documentation

Backups fail after inSync Server upgrade to version 5.9.8 or later

This article applies to:

  • OS: All inSync supported OS versions
  • Product edition: inSync On-Premise

Problem description

Backups stop after upgrading the inSync Server to the version 5.9.8 or later. 

Cause

Druva inSync On-Premise versions 5.9.8 and later do not support disk space savings storge (BDB Storage) on Oracle Berkeley DB.

Traceback

The following error is written to the inSync Client logs.

[2018-10-17 13:57:40,200] [WARNING] Backup failed. Error: Invalid store type for this operation. (#10000001d)

Resolution

  1. Check if BDB storage is installed on the server as follows:
    • Log in to the inSync Server and navigate to the database location.
    • Check whether the db folder name is _drstdb_ or_bynamodb_, where _drstdb_ indicates a BDB storage and _bynamodb_ indicates a Bynamo storage.
  2. If required, create a new Bynamo storage and move all the users to the new storage.
  3. Run a test backup from the Bynamo storage to check if it completes successfully.

Additional information

inSync versions lower than 5.9.8 supported the following storages:

  • BDB - The disk space saving storage:  BDB is a legacy storage system that posed several performance issues with the BDB storage engine.  No improvements have been introduced in the BDB storage engine. inSync support for BDB was withdrawn from version 5.9.8.
  • Bynamo storage - The performance-based storage: This storage was introduced from inSync Server version 5.3 with the following advantages over the BDB storage engine.
    • Hot Backup feature that snapshots the database. This feature is useful in recovering data during DB corruption.
    • Signature file verification to avoid data and database corruption when the storage component is not connected to the server.
    • Allows adding multiple data volumes to the storage (not supported in BDB).
    • Substantially small database size. (4 to 5 percent of total data).
    • Scalable database with lesser hardware requirements.
    • Data pooling (disk spanning), flexible storage mapping (ability to move users across storages), load balancing, increased parallel connections to 500 (compared to 100 in BDB)