Ask the Community
Groups
Known Issues for PHDVB 6.5 - Connect IT Community | Kaseya
<main> <article class="userContent"> <h2 data-id="summary"><strong>SUMMARY</strong></h2> <p>A list of known issues for PHDVB 6.5</p> <h2 data-id="issue"><strong>ISSUE</strong></h2> <p></p> <h2 data-id="phd-virtual-backup-v6-5-known-issues">PHD Virtual Backup v6.5 Known Issues</h2> <p> </p> <h2 data-id="general">General</h2> <ul><li>Archiving the same backup results in an overwritten archive history record. (DCK-3065)</li></ul><ul><li> <p>After the PHD VBA is hard reset either manually or due to another reason, you may encounter the following error, Startup: Failed to mount Dedupe Store: No section: 'PHDVBStore' Please reconfigure this PHDVBA VM. To resolve this issue, soft reset the appliance (Restart Guest Reboot). (DCK-2946)</p> </li></ul><ul><li>If you remove an attached disk backup store from a configured v6 PHD VBA then add a legacy attached disk (pre-LVM), you will need to first configure the PHD VBA's backup storage by setting the Storage Type to None, then using the Configure dialog to add the legacy disk.</li> <li>If you are using a legacy (pre-LVM) attached disk data store with a v6.0 PHD VBA and you then remove that disk from the PHD VBA, the PHD Console may lose the ability to communicate with the PHD VBA. To work around this issue, add a new disk to the PHD VBA to re-establish communication again. (DCK-2692)</li> <li>When editing network settings using the PHD VBA's text-based network configuration interface (CTRL-N) from the VM's hypervisor console, both DNS options are required. If you do not have an alternate DNS, enter the primary DNS address again to save your settings. (DCK-934)</li> <li>When the PHD Console computer's regional format is set to allow time formats that use a dot (for example Italian time formats allow HH.mm), creating jobs with a specified start time may not work correctly (a 500 error is displayed and incorrect times may be reported in the PHD Console). To workaround this issue, you can change the computer time format to use a colon (HH:mm). Note that you may also need to change the region to one that allows the selection of a time format with colon (for example, English (US). (DCK-723)</li> <li>Non-ASCII characters are not currently supported. Backing up and restoring VMs that include non-ASCII characters in the VM name, disk name, or text fields (Annotations Description, etc.) can cause the VBA to hang or crash. (DCK-508)</li> <li>Datastore names that contain square brackets [ ] can prevent backups and restores from functioning correctly. (DCK-378) VMware only</li> <li>Jobs that use quiesce may fail due to a hypervisor limitation. This is related to the issue described in the following VMware Communities article: <a rel="nofollow" href="/home/leaving?allowTrusted=1&target=http%3A%2F%2Fcommunities.vmware.com%2Fdocs%2FDOC-11987">http://communities.vmware.com/docs/DOC-11987</a>. To work around this issue you can create smaller, separate jobs for only those VMs that require quiesce. Do not use the quiesce function in larger jobs that include many VMs that do not need to be quiesced before backing up. In addition, If using PHD Guest Tools, if the quiesce process is eventually completing, but PHD Virtual Backup is timing out, contact PHD Virtual support to increase the timeout value. VMware only</li> <li>Currently, vNetwork Distributed Switches are not supported for backup, restore, or replication.VMs that use these switches can still be backed up, but network information will not be available for restore. Refer to the Restore section, below, for specific issues and workarounds. VMware only</li> <li>When restoring a thin-disk VM that had a snapshot at time of backup, the restored VM disks are created as thick on vSphere 4.x. This is a known issue with VMware version 4.x, When a virtual machine with thin disk has a snapshot, the redo log is incorrectly created as thick disk. The bug was fixed in vSphere 5. For version 4.x, you can first commit or remove the snapshot before taking the backup. (DCK-2168) VMware only</li> </ul><h3 data-id="phd-console">PHD Console</h3> <ul><li>The Backup Archive Process tab initially loads duplicate records. Clicking refresh removes all duplicates. (DCK-3058)</li> <li>Sorting columns in the Backup Archive Process tab results in an unhandled exception in some cases after refreshing the table. (DCK-3044)</li> </ul><ul><li>Migrating jobs with schedules set to run in the past result in a 500 - Internal Server error in the PHD Console. (DCK-3043)</li> <li>Pre-version 6.2.0 VBAs are listed in the Migration Source dropdown menus. Migrating settings and jobs to version 6.5 is supported only from version 6.2 and later. (DCK-3049)</li> <li>In some cases, the Instant Recovery Write Space details displayed on the Dashboard may not be accurate (free Write Space may always display 100% for example). (DCK-2895)</li> <li>If the following error is encountered after adding an additional Write Space disk and saving the configuration: Failed to save changes: Value must be between the minimum and maximum values. Parameter name: Value clicking Save again will resolve the issue. (DCK-2919).</li> <li>Migrating job history fail when all settings are enabled. To workaround this issue, job history can be migrated individually or you can export then import the job history. (DCK-2739)</li> <li>In some cases, changing configuration details (Email, Retention, Support, etc.) has resulted in mount errors with CIFS-based backup data stores. An example error string, Appliance returned an error: Error with local storage configuration: Failed to mount CIFS datastore: mount error(13)... Restarting the PHD VBA will force the data store to re-mount. To prevent additional errors, avoid making configuration changes while jobs are in progress. (DCK-2744)</li> <li>When selecting multiple backups in the Backup Catalog, the Recover button is not active. To restore multiple VM backups, you can select a single backup to open the wizard, then select any additional backups. (DCK-1578)</li> <li>PHD Console does not refresh after a PHD VBA's IP address has changed. Close then open the PHD Console to reestablish the connection to the PHD VBA using the new IP address. (DCK-909)</li> <li>If the PHD Console does not open from vSphere Client, make sure you are logged in to vCenter with an account that has vCenter permissions. Accounts with permissions at a Datacenter level are not sufficient. (DCK-593) VMware only</li> <li>iSCSI mounted backups may fail to delete in the PHD Virtual Backup Console even though they are not connected to an initiator. If a machine is connected to a PHD backup via its iSCSI initiator and that machine is restarted, you may not be able to delete the iSCSI mounts from the PHD Management Console File Recovery catalog. Restart the VBA and the mounts will be automatically deleted from the File Recovery catalog. (DCK-557)</li> <li>PHD Virtual Backup Console does not detect iSCSI Initiator when using Windows 2003, 64-bit. Make sure the iSCSI Initiator is installed then use the Windows Start menu to run the program and manually mount any iSCSI targets that you've created using the iSCSI Initiator. (DCK-450)</li> <li>The System Alert Viewer displays backup times based on the time zone of the machine on which the PHD Console is installed. If the PHD VBA uses a different time zone, the backup time shown in the Backup Catalog will not match what is displayed in the Alert Viewer. Verify the time zone of the PHD Console before looking for backups included in the alert reports. Any backups with an alert will have an alert icon displayed next to the backup name.</li> </ul><h3 data-id="backup">Backup</h3> <ul><li>Running Full/Incremental backups of VMs that contain special characters in their names always results in a Full backup with job log warning, Unable to correct incremental chain, forcing full. int() argument must be a string or a number, not 'NoneType.' (DCK-2862)</li> <li>In some instances, when a Full/Incremental backup job is canceled or fails, incomplete data may be left behind in the backup data store. This can be manually removed by enabling access to the Raw Data Access share on the backup data store. A fix for this issue will be available in an upcoming hot fix or release. (DCK-2849)</li> <li>If two hosts have the same UUID in an environment, and one is not licensed, errors may be encountered when attempting to backup VMs on the licensed host. (DCK-2225) VMware only</li> <li>Physical RDM disks are not actually excluded in the Backup wizard resulting in warnings in the backup report. These warnings will not affect performance. (DCK-1124) VMware only</li> <li>When backing up a container (Cluster, folder, etc.) although the Backup Wizard includes the PHD VBA size in the calculated Data Size, the PHD VBA is not included in the actual backup job. (DCK-452) ) VMware only</li> </ul><h3 data-id="cloudhook-cloud-backup-storage">CloudHook - Cloud Backup Storage</h3> <ul><li>Cloud: OpenStack Swift version 2 with Keystone authentication is not currently supported.</li> <li>Migrating or importing settings to a PHD VBA that previously had a cloud bds configured will result in an unresponsive PHD VBA. Use a newly deployed VBA as the migration or import target or contact support to resolve the issue with the existing VBA. (DCK-2944)</li> <li>Cloud Storage Paths that include upper case characters may result in a mount error. Use lower case characters when creating storage paths/buckets. (DCK-2909)</li> <li>If a PHD VBA loses connection to a cloud storage location, to re-mount storage, the PHD VBA must be restarted. (DCK-2936)</li> </ul><h3 data-id="recovery">Recovery</h3> <ul><li>When using the VBA's backup data store as Write Space, if a float division error is encountered, Write Space threshold emails will not be sent if either the warning or stop threshold is reached. To workaround this issue, use a separate attached disk as the Write Space location. (DCK-2907)</li> <li>Attempting to create an Instant Recovery job using a PHD VBA that does not have Write Space configured will result in a 400 error. Configure Write Space for the PHD VBA and create the job again. (DCK-2928)</li> <li>Rollback recovery using a Full/Incremental backup ends successfully with warning, 'Backup ... is incomplete, trying next available backup. This error can be ignored. (DCK-2845)</li> <li>Recovering a VM that was backed up without a NIC may result in a restore job error (Error recreating networks). This will not affect the performance of the recovered VM. To avoid this error the VM can be restored without a NIC selected using the Edit window at the Summary step of the wizard. (DCK-2758)</li> <li>Some virtual machine attributes are not included with restored virtual machines. For a complete list, refer to the User Guide.</li> <li>If a selected network becomes unavailable before a restore job is run, VMs may be restored without a network interface card. Make sure any networks selected for use with each restored VM are valid. (DCK-1946)</li> <li>When the Recovery wizard is opened, all VM backups are expanded - if you had selected a VM before opening the wizard, you may need to scroll to find the selected VM. (DCK-1916)</li> <li>Archived backups do not display an archive icon in the Recovery wizard Select VMs list. This is a cosmetic issue that does not affect functionality. (DCK-1710)</li> <li>Windows Server 2012 VMs with a network adapter set to E1000E results in no network adapter included with the restored or replicated VM. Change the network adapter type to VMXNET 3 or manually add the network adapter after the restore or the initial replication process.</li> </ul><h3 data-id="replication">Replication</h3> <ul><li>Scheduled replication jobs run immediately after creation when system date and VBA date are in different timezones. (DCK-3011)</li> <li>You may encounter a permission denied error when using the Discover option to mount a CIFS backup data store that uses special characters as a replication data store. Manually entering the data store details will work around this issue. (DCK-2695)</li> <li>When adding a PHD VBA's own CIFS backup storage as a replication data store on the same PHD VBA, if that CIFS share uses special characters, you will encounter permission denied errors. In addition, the data store cannot be removed or edited once added in the replication configuration tab. To resolve this issue, the backup data store information must be removed first then added again. (DCK-2695)</li> <li>If a selected network becomes unavailable before an initial replication job is run, replicas may be created without a network interface card. Make sure any networks selected for use with each replica are valid. (DCK-1946)</li> <li>After refreshing the list of VMs available for replication, if a specific backup is not displayed, you can force an update by either 1) remove, then add the backup storage location as a replication store, or 2) change the contents of the backup catalog by running another backup or deleting a backup. After trying either workaround, refresh the list again and the missing backup should be displayed. (DCK-1871)</li> <li>When a source virtual disk size has changed, the next replication job will report an error. To replicate the updated disk you will need to remove the replica then run the job again. ( DCK-1544)</li> </ul><h3 data-id="phd-exporter">PHD Exporter</h3> <ul><li>When a new backup data store is added during the Create new job wizard it cannot be used unless the wizard is closed then re-opened. (DCK-2099)</li> <li>Backed up paravirtualized VMs are not paravirtualized after export. Contact PHD Support for help editing the OVF file to enable paravirtualization. (DCK-2633)</li> <li>When using any paths that include special characters, OVF creation may fail. Paths to backup storage that contain special characters are not currently supported. (DCK-2034, DCK-2028)</li> <li>If you cannot edit a previously added backup store (Save fails with an inaccessible error), remove the backup store then add a new store with the updated backup location information. ( DCK-1940)</li> </ul> </article> </main>