Link Search Menu Expand Document

Enhancements, Fixes, and Known Issues 03/16/2023

The following enhancements, fixes, and known issues are found in this release, dated 03/16/2023:

New features & enhancements

  • The number of retries in ScaleZ operations was increased to improve background jobs stability.
  • Postgres backup and restore operations through Elastio CLI were extended to support RDS databases.
  • Phased progress reporting was introduced into jobs: now AWS operations consisting of several child jobs would be reported in elastio CLI according to the number of completed jobs, i.e. [3/5].
  • EC2 backup would now be aborted if the instance specified in the command does not exist in the AWS account and region used. This would lead to such jobs to not be counted as failed in compliance reports, as assets that do not exist could not be possibly backed up.
  • Validation was introduced to make sure that empty recovery points don’t get created.
  • Elastio postgres restore functionality would now create a database automatically when restoring a plain SQL database.
  • Source asset data was introduced into jobs in Elastio CLI. Now when performing a restore, iscan or filesystem check asset ID will be visible in the terminal.
  • Password parameter was made optional in postgres backup.
  • Implemented direct AWS EBS snapshot iscan in Elastio CLI.
  • Postgres recovery test command was introduced to Elastio CLI. It accepts the test located on an S3 bucket in AWS specified in .yaml or .json formats. The way it works is as follows: a temporary EBS volume is created, Postgres database is restored to it, then recovery test is performed on it, after which the temporary setup is cleaned up.
  • AWS Backup recovery point import was extended to give possibility to import recovery points from other regions.
  • Recovery points retention was introduced to Elastio tenant. Now you can setup rules to keep the recovery points according to a set of rules, not all of the created ones.
  • Spot instances types were optimized for best performance with minimal cost.

Bug Fixes

  • EBS volumes created by Elastio at the time of restore would in some cases lack the “elastio:nobackup=true” tag that occasionally lead to them being backed up at the time when data was restored to the volume. This was fixed.
  • Error messages for mounting recovery points was improved to include the details on different fail types.
  • Recovery points imported from AWS Backup would in some cases have the incorrect creation time. This was fixed.
  • Recovery points imported from AWS Backup are now always deduplicated.
  • AWS Backup recovery point import and Elastio recovery point iscan operations integration with EventBus was fixed to include default behavior and avoid redundancy in parameters.
  • EC2 restore can now be performed on a recovery point imported from AWS Backup.
  • Fixed tags cleanup logic by removing the “elastio:resource” tag last.
  • Device is now kept connected if mount failed providing a proper error message, it can be dismounted like any other device.
  • Improved job results reporting for the filesystem check jobs. Previously it could fail but the results were reported as successful, now the reported status matches the actual one.
  • Previously if recovery point contained more than 1 asset snapshot the EventBridge map for such recovery point would come out incorrect. This was fixed.
  • In some cases after an ìscan failure the mounted recovery point would not get dismounted - this was fixed.
  • Whe aborting a job the child job status would not always become Aborted - this behavior was improved to check for the status of the child job being Aborted before aborting the parent job.
  • A panic would occur in some cases when detecting a filesystem type in filesystem check jobs. This was fixed.
  • In some cases a restored EBS volume would not get into protection statistics due to “elastio:volume” tag presence. This was fix.
  • RDS postgres recovery points would sometimes be missing in the elastio rp listcommand output. This was fixed, now all postgres recovery points show up on the list.
  • EC2 \ EBS backup would sometimes panic with “BUG: Returned block doesn’t have first block token”. This was fixed.
  • EventBridge permissions were extended to include cross account operations.
  • A panic would sometimes occur in recovery point import from AWS Backup is incorrect arguments were passed. This was fixed.
  • Inconsistent job update would cause the backup to fail during EC2 backup with VSS-consistent snapshots enabled. This was fixed.
  • Link Source and Repair Source jobs would previously get stuck on resync stage. This problem was eliminated, the jobs now successfully complete.