Pure Release Notes
Release Notes for Archiware Pure version 1.0
Version 1.1.2 released. Changes since last release:
- Fixed: Fixed an error where ESXi hosts would sometimes not be correctly listed in VMware inventory if a datacenter was divided in clusters.
- Fixed: Special precautions were added to prevent database being reported locked when executing parallel backups on more than one storage pool.
Version 1.1.1 released. Changes since last release:
- Fixed: Fixed an error where backups would fail when trying to get properties from a vSAN datastore.
Version 1.1.0 released. Changes since last release:
- Improved: Single file restore has been greatly improved and now supports restoring files from a wide range of systems. In addition to MBR and GPT partition maps, BSD disklabels are now also supported. Files can be restored from the following file systems: ext2, ext3, ext4, NTFS, FAT, HPFS, btrfs, ReiserFS, SquashFS, MINIX, JFS, XFS, UFS, Xenix, System V, HFS, HFS+ and exFAT.
- New: Archiware Pure now has an Event Log section which gives you an overview of all the backup, verify and restore jobs. Clicking on the '..' icon immediately to the left of a VM name will bring up the detailed message log for that particular operation.
- New: A new 'Overview' section has appeared in the Backup Administration side panel (found on the right hand side of the screen), showing you a visual representation of the last 7 backups for the selected VM.
- New: 'Delete from repository' option (accessed from the quick actions menu immediately to the left of a VM name) allows removing a VM and all its data from Pure backup repository. Keep in mind that if the VM is still present within connected VMware infrastructure, Pure will auto-discover if again and assign it to the default backup pool. The data that was occupied by the deleted VM will be freed from the repository on next backup window run.
- New: Running backup, verify and restore jobs can now be cancelled by clicking on the 'x' icon which will appear next to the job progress bar.
- Improved: Restore method has seen significant performance improvements. Disk allocation is now preserved in restored VMs, meaning that the thin disks will be restored as thin.
Version 1.0.4 released. Changes since last release:
- Fixed: Removed a bug that allowed the wrong resource pool to be selected during restore process (marked as '/'), which caused the restore to fail.
Version 1.0.3 released. Changes since last release:
- Fixed: Removed a bug preventing Pure from collecting the correct list of available resource pools on hosts in a HA cluster. VMs can now be properly restored onto a HA cluster assigned hosts.
- Improved: Further improvements to job scheduling algorithm assure that vCenter VM is backed up separately from the VMs that it manages. Earlier, if a vCenter VM got stunned during backup snapshot creation or removal, all other backups would fail as Pure was not able to communicate with the vCenter. Both during backup and verify, vCenter VM now waits for all other VMs to finish first and no other VMs are being processed until vCenter backup is complete.
Version 1.0.2 released. Changes since last release:
- Fixed: Fixed an error causing VM restore to fail when restoring to a HA Cluster having only the default compute resource without any child resources defined.
- Improved: If Pure wakes up (boots) after a Backup window has already started, the VMs belonging to that BW will still be sent for execution. Previously, the whole BW was skipped.
- Improved: Scheduling mechanism has been improved so that each VM can only be backed up and verified once during a single Backup window.
Version 1.0.1 released. Changes since last release:
- Fixed: Fixed an error in the web server that would cause Pure to erroneously think that a VMware infrastructure server is offline under certain circumstances.
- Improved: Log messages have been cleaned up by decreasing log level of some recurring notifications.
- Improved: Email report scheduling has been improved to assure that the report is rescheduled each time one of the corresponding settings is changed. Removing the recipients setting will not correctly unschedule the report.
- Fixed: If the very first backup of a VM fails, corrupt (incomplete) backup data is now immediately deleted as originally intended.
- Fixed: Resolved the issue where in certain occasions email subject of the daily report was not correctly set.
- Fixed: Resolved the issue where VM could not be restored if it had a Cdrom device pointing to a file on a datastore unless that disk image was ejected from the Cdrom device. Due to the inherent VMware limitations, VMs with a file based Cdroms will be restored with a generic Cdrom pointing to a host's Cdrom device.
Version 1.0 released.
For a list of features, please click here.