Chatee ahora con Soporte
Chat con el soporte

QoreStor 7.4.1 - Release Notes

Deprecated features

The following is a list of operating systems that are no longer supported starting with QoreStor 7.1.2.

  • CentOS 7.x, 8.x
  • RHEL 7.3 - 7.8, 8.2- 8.7
  • Oracle Linux 7.3 - 7.8, 8.2- 8.7

NOTE: As of version 7.4.1, QoreStor deprecated support for CentOS, RHEL, and Oracle Linux versions older than 7.9, and RHEL and Oracle Linux version 8.0-8.7. If you currently use a deprecated 7.x version, for best performance and product support, Quest, upgrade to version 7.9. If you use a deprecated 8.x version of RHEL or Oracle Linux, Quest recommends upgrading to version 8.10. If you use an 8.x version of CentOS, Quest recommends migrating to a supported 8.x platform.

Upgrade notes

This topic contains important information for users who are upgrading from a previous QoreStor release.

  • If the existing installation of QoreStor uses the default passwords for the UI admin login, qsservice user, or qsadmin user, then you must update these passwords before upgrading to QoreStor 7.4.1
  • Users with UI monitor role cannot be used for protocol authentication for backups. For example, user cannot have monitor and RDA as the roles. Upon upgrade to 7.2.0 release, such users will lose monitor role. New users can be created with monitor role as required.
  • Admin user is required to change password upon first time logging in to QoreStor UI.
  • Auto upgrade is not available from the QoreStor Menu.
  • AWS Lambda and IAM permissions are required for an Archive Tier. After you complete an upgrade, to update the IAM roles and permissions, see the Quest QoreStor Release Notes.
  • Existing QoreStor installations can not be upgraded to use Object Direct storage. To use QoreStor in Object Direct mode, you must perform a fresh installation.
  • During the upgrade, the QoreStor installation script will verify that the Repository and Metadata filesystems are mounted with the noatime mount options. If not, the installation script will prompt the user to allow the script to make the necessary changes in /etc/fstab and prompt you to reboot the system and re-run the QoreStor installation process.

    NOTE: Once upgraded to QoreStor 7.4.0, the "dirsync" mount option can safely be removed from /etc/fstab. While QoreStor will still function normally with the dirsync option enabled, removing the option will increase performance, slightly.
  • The metadata volume storage requirement for Object Direct configuration has been extended to 4TB, 18TB, and 44TB for cloud-optimized, standard, and enterprise modes. Although it is not mandatory to extend the metadata volume on existing QoreStor instances before upgrade, Quest recommends you extend the volume when it is reaches 90% full.
  • QoreStor 7.4.1 contains VTL support for RHEL/Oracle Linux/AlmaLinux/Rocky Linux versions 7.9, 8.8-8.10, and 9.2-9.4. To enable VTL support, see the Quest QoreStor Installation Guide.
  • QoreStor 7.4.0 consumes approximately 1.6 GB of disk space under the /opt directory. For this, the QoreStor 7.4.1 installer checks for a minimum of 3 GB of available disk space in the root file system before proceeding with an installation or upgrade.

    When using QoreStor with VTL, if want to upgrade the operating system, upgrade QoreStor before you upgrade the operating system.

Upgrading to Rapid CIFS Plug-in 4.0.4.4002 with Secure Connect

Secure Connect is comprised of a set of client and server components that create a resilient and secure communication channel for use by WAN-connected clients. The server components are installed and enabled by default during the installation of the QoreStor server. Starting with the Rapid CIFS plug-in version 4.0.4.4002, the plug-in installer includes the option to install and enable Secure Connect.

IMPORTANT: Refer to the Known Issues section of this document for Rapid CIFS issues known to be present at the time of release.

If you have WAN-connected Rapid CIFS clients and choose to install Secure Connect, you can do so by following the steps below:

  1. On the media server, map a network share to your CIFS-enabled container.
  2. Download the plug-in installer from the website, as detailed previously.
  3. Open a command prompt with the “Run as Administrator” option selected. To do this using the Windows Start menu, click Start → All Programs → Accessories. Right-click Command Prompt and selectRun as Administrator.

    This gives all the required privileges to install/copy the driver files to the Windows drivers folder.

  4. Run QuestRapidCIFS.4.0.4.4002.exe.
  5. When prompted, select Install a Secure Connection to install Secure Connect.

    Secure Connect ensures backups to QoreStor are completed in spite of slow or unreliable WAN connections. When selecting this option, you will be prompted for additional configuration information required to make Secure Connect function correctly.

  6. At the Configure Secure Connect page, complete the dialog using the IP or fully qualified domain name (FQDN) of the QoreStor server and appropriate credentials.

    The user account used must have the CIFS role assigned.

    NOTE: The Configure Secure Connect operation is not supported for CIFS containers on DR Series appliances. DR Series appliances do not contain the required Secure Connect server component . This operation is available only for QoreStor instances. For CIFS containers on DR Series appliances, connect using the Rapid CIFS plug-in and the DR Series host name.

    NOTE: The value entered in the Display Name field must not match the hostname of the QoreStor server.

Upgrading to Rapid NFS Plug-in 4.0.3320.1

Secure Connect is comprised of a set of client and server components that create a resilient and secure communication channel for use by WAN-connected clients. The server components are installed and enabled by default during the installation of the QoreStor server. Starting with the Rapid NFS plug-in version 4.0.3320.1, the plug-in installation script includes the option to install and enable Secure Connect.

IMPORTANT: Refer to the Known Issues section of this document for Rapid NFS issues known to be present at the time of release.

If you have WAN-connected Rapid NFS clients and choose to install Secure Connect, you can do so by following the steps below:

  1. Download QuestRapidNFS-4.0.3320.1-x86_64.bin.gz from the website.
  2. Unzip the package using the following gunzip command:
    gunzip QuestRapidNFS-4.0.3320.1-x86_64.bin.gz
  3. Change the permission of the binary package to allow it to run:
    chmod +x QuestRapidNFS-4.0.3320.1-x86_64.bin
  4. Install the Rapid NFS package. Before installing, remove the stale NFS entry.
    QuestRapidNFS-4.0.3320.1-x86_64.bin -install
  5. Load the file system in user space (FUSE) module, if not already loaded:
    modprobe fuse 
    
  6. Create a directory on the client. For example:
    mkdir /mnt/backup
  7. Mount Rapid NFS as a file system type using the mount command. For example:
     mount -t rdnfs 10.0.0.1:/containers/backup /mnt/backup 
    

    To enable Secure Connect, mount Rapid NFS using the sc flag, as shown below:

    mount -t rdnfs 10.0.0.1:/containers/backup /mnt/backup sc,user=backup_user,password=<backup_user password>

    If you are using a DMA that supports a marker, set the marker by using -o in the mount command. For example, if you wanted to set the CommVault marker (cv):

    mount -t rdnfs 10.0.0.1:/containers/backup /mnt/backup -o marker=cv

    NOTE: If you want to do a mount on AIX, you must set the nfs_use_reserved_ports and portcheck parameters first. The parameters cannot be set to 0. For example: root@aixhost1 / # nfso -po portcheck=1 root@aixhost1 / # nfso -po nfs_use_reserved_ports=1

Resolved issues

The following is a list of issues addressed in this release. 7.4.1

Table 1: Resolved issues in 7.4.1.169 release

Issue ID Resolved Issue  
QS-16390 Cleaner stats for Cloud Tier (stats --cloud_tier) shows overall cleaner status.  
QS-16403 UMEM_CACHE: Get a way around lock contention while memory is allocated and freed.  
QS-16411 Tune count of files going to the cloud in parallel by using environment variable.  
QS-16487 Disable continuous replication support for object container.  
QS-16503 Backup failing due to plug-in initialization.  
QS-16541 Anomaly Detection service failed to start due to low space during installation.  
QS-16542 Add retries in cloud read for S3 when error received is EAGAIN from the cloud provider.  
QS-16550 Prune fast cache space by crawling across multiple containers.  
QS-16564 Fine tune memory limits of cleaner base on total available system memory.  
QS-16571 Reset the cleaner's pending bytes to zero when the cloud and local cleaner (both) reach to the DONE state at least once.  

Open issues

Known issues

The issues below are known to exist in this release.

Table 2: Known issues in this release
ID Issue

QS-437

QS-438

When using multiple NetVault Servers to access the same QoreStor Server, use storage group, container, and user names unique to each NetVault server to avoid accidental deletions.

QS-1446

If the QoreStor partition is not set to automount, and the QoreStor server is rebooted, the system appears to be operational but shows 0 physical capacity.

QS-1588 When running Optimized Duplication jobs in Backup Exec, the container quota is not enforced as it is for backup jobs.
QS-1666

When NetVault is installed on a Linux server, the RDA Client version is not displayed correctly.

Workaround:

To view the build number of the plugin, run the command

strings /usr/netvault/dynlib/librofsoca.so | grep "(EAR-4.1.0) Build:"

QS-1670 For Linux clients, the full version number of RDA plug-ins is not displayed.

QS-1823

In some situations, the deduplication savings for a storage group shows as 100% before a backup has occurred.  Once a backup has occurred, the deduplication savings will be displayed accurately.

QS-2242

When installing QoreStor, the filesystem selected for the QoreStor repository should not contain data from other application or users. QoreStor must have exclusive access to the repository filesystem.

QS-2604

If unregistered clients access QoreStor CIFS shares by mounting the share, the /var/log/messages will contain numerous host name/mismatch errors.

Workaround:

Unregistered clients need to be updated in DNS server manually.

QS-2694

When the system time on the QoreStor server is synchronized with the domain time, this may cause the system time to adjust backwards. If this occurs, system statistics (like read/write throughput statistics) will appear to be hung for an amount of time equal to the time adjustment.

QS-3779

When using Secure Connect with NetVault in a multiple-machine configuration (NetVault server, NetVault client, QoreStor server), Secure Connect can respond to errors only for connections between the NetVault server and QoreStor or the NetVault client and QoreStor. Secure Connect does not address connections between the NetVault server and NetVault client.

QS-4169

When using the Linux exports http_proxy or https_proxy, no_proxy is also required for the watcher service to function correctly. Refer to "Accessing QoreStor" in the QoreStor User Guide.

QS-5072

When authenticated to a domain, connecting to CIFS shares with a localhost account is not successful.

QS-5105 Changes to a Cloud Tier Policy for inclusions or exclusions will not take effect until the QoreStor service has been restarted.
QS-5438 After performing a DR to QoreStor migration, network settings are reset to default.
QS-5627 The current error output of "rda --limit" without any argument incorrect suggests limits are enforced via Mbps (megabits per second) when the limit is actually in MBps (megabytes per second).
QS-5629

When configuring a custom graph, if you select a period for which there is no data, the graph will be blank and display "No Data Available"

QS-5826 When performing a DR to QoreStor migration, bonding modes and MTU settings are not
QS-6291

When using the latest version of the Rapid CIFS plug-in with Veeam Backup and Recovery, the configurations below are required:

  • For the Backup Repository, change the Gateway Server option to The following server and enter the IP or hostname of the local machine.
  • For VMware Backup, the Gateway server should be changed from the Automatic selection to the Use the selected backup proxy servers only.
QS-6653

OST MSI plugin reports a new version is already installed when installing an updated driver

Workaround:

Uninstall the OST driver before installing the newer driver

QS-6782 When installing Rapid CIFS on Windows 2008 R2, the installation deploys driver version 4.0.3207.1 , causing the version displayed by rcifsctl.exe driver -q to be different from the version in the Windows Programs and Features dialog. The functionality of Secure Connect is not changed.
QS-6808

When configuring Backup Exec with CIFS shares using the display name, the steps below must be followed:

  1. In the command prompt, change directpries to the path C:\Program Files\Quest\Rapid CIFS.
  2. Run rdcifsctl.exe sc -d for disable a secure connection.
  3. Configure storage in Backup Exec by specifying \\displayname\container as network share. During this configuration all connections from Backup Exec to QoreStor are secured.
  4. Run rdcifsctl.exe sc -e to enable a secure connection.
QS-6860 During upgrade, the SMB config file is overwritten. Any manual changes to this file will be lost on upgrade.
QS-8945

When QoreStor is installed on Linux versions 7.3 - 7.7, configuring an archive tier may result in a memory leak. This is due to an issue with the curl package installed on those versions of Linux.

Workaround

Either upgrade the Linux version on the QoreStor server to 7.8, or update the curl package to curl 7.29.0-57.el7. DR Series systems migrated to QoreStor which are on Oracle Linux 7.3 should update the curl package to 7.29.0-57.el7

QS-9056

On DR Series Systems migrated to QoreStor, the maximum inode limit for an object container is set to 1 million. In some cases, this may result in a "maximum allowable File(s) and Directories limit" error.

Workaround:

To increase that, the environmental variable MDS_OT_INODES_ALLOWED in /etc/oca/customer.env has to be set to a higher value. Please contact Quest Support for assistance making this change.

QS-9021

During object direct installation, if the message "Failed to validate object direct connection string! "may be received if the certificate has been changed or is expired on the QoreStor server.

Workaround

Purge the certificate with the command below

cloud_tier --purge-ssl-certificate

QS-9177 When configuring the Rapid NFS client with Secure Connect, You may mount the plug-in with either the IP address or the fully qualified domain name (FQDN). If you use the FQDN, you must also add an entry in the /etct/hosts file on the QoreStor system and verify that the hostname is resolvable.
QS-10018

When writing to an NFS container on a system with 24GB or less of memory, some customers may receive "out of memory" errors.

Workaround:

On the QoreStor system, run the command sysctl -a | grep tcp_mem.

This will return your tcp memory consumption, like the example value below. The return indicates your minimum, average, and maximum values.

net.ipv4.tcp_mem = 285198 380264 570396

Using the command below, set the tcp_mem maximum value to the average value (380264, in the example above). Note that your values will be different. The value in the center is the average.

sysctl -w "net.ipv4.tcp_mem=285198 380264 380264"

QS-10042

When attempting to delete a bucket containing a large amount of data, QoreStor will appear to freeze. This is due to an AWS S3 limitation that requires the objects within a bucket to be deleted before the bucket can be deleted. During the bucket delete operation, the objects are enumerated and deleted one at a time.

Workaround:

Before deleting the bucket, delete the objects within the bucket using the AWS S3 console.

QS-10117 System stops while upgrading from the user interface terminal.
QS-10772

Replication update fails with replication between QoreStor 7.4.1 and the DR appliance.

Workaround:

For DR-to-QoreStor replication, update replication by way of the QoreStor user interface or command line interface.

QS-11329 During replication, crash may occur on a target if the replication source has archive tier configured.
QS-11588

When there is not enough space under "/opt" to install or upgrade QoreStor, an error message occurs.

Workaround:

Free space under the "/opt" partition.

QS-11613`

During replication, QoreStor is unable to locate a specific blockmap entry.

Workaround:

If a replication source already has archive tier replication, do not configure archive tier replication on the replication target.

QS-11818

New backups jobs fail with Backup Exec 20.x after stubbing the existing data.

Workaround:

Delete the Changer.cfg meta file in the CIFS container, which triggers the start of a new backup job.

QS-11819 Received the following error: RDA SRVR RDA_OPEN_FILE : failed to add open image to hash map for LSU 9 scid 1152921504606847079 (Bad file descriptor).
QS-11936 Netvault Backup Jobs may stall if cloud upload on QoreStor is not happening because of cloud provider experiencing high request rate.
QS-11972

Adding replication between DR 4.0.4 and QoreStor 7.1 fails.

Workaround:

Run the replication commands on QoreStor.

QS-12017

QoreStor cannot add replication in a situation when there are two Secure Connect server processes and one is in unknown state after upgrade.

Workaround:

Stop the old Secure Connect server in the unknown process.

QS-12113 Alert generated for OFSCK scan is still present after QoreStor becomes operational.
QS-12117

During replication, the error "Failed to set ns txlog csn on scid" occurs.

Workaround:

You can ignore these messages. Replication eventually succeeds.

QS-12228 Storage group encryption once disabled cannot be enabled.
QS-12240 Replication add from migrated QS between RDA containers may fail.
QS-12260 VTL tape export is successful even without cloud replication configured.
QS-14868

CWF : "authenticate --leave" fails to disjoin system from domain in case domain controller is not reachable/ trust is broken.

Workaround:

It is advised to retry “authenticate --leave” operations after “service ocards restart” or a system reboot.

QS-15305

Object role for users is not effective on Object containers created in 7.2.0 version.

Workaround: For Object containers created in 7.2.0 version, users are per container and they should be added for each container from the container -> Edit page in QoreStor UI.

QS-15387

While running filesystem scan Object container show Inode count as zero Object container files are scanned but the count is not updated.

Workaround: The total system level count is shown which includes Object container inode count.

QS-15554

Changing QoreStor hostname renders loss of ownership of files/folders under CIFS containers.

Workaround: Preserve system SID by backing up the SID before changing the hostname and restore it afterwards. To backup the SID, execute the following command and capture its output:

/opt/qorestor/samba/usr/bin/net getlocalsid

To restore the SID, execute the following command:

/opt/qorestor/samba/usr/bin/net setlocalsid <SID captured from getlocalsid>

QS-15756

Domain join or leave is either hung or getting error on 9.x OS systems Workaround: OS service realmd might not be responding.

Workaround: Restart the service using shell command 'systemctl restart realmd.service'

QS-15789

BackupExec with Cloud Lock throws errors during restore job.

Workaround: As part of cloud tiering policy, exclude files of pattern (regular expression) ".*.(cfg)"

QS-16500

Starting from version 7.4.1, QoreStor does not support continuous replication for object container.

NOTE: Object container replication is not recommended for large backups.

QS-16589

Unable to join the QoreStor to the Windows 2016 DC.

Workaround: comment the mentioned line of code → "includedir /etc/krb5.conf.d/" in "/etc/krb5.conf" .

Documentos relacionados

The document was helpful.

Seleccionar calificación

I easily found the information I needed.

Seleccionar calificación