DELLTechnologies PowerStoreOS Container Based Storage System Owner’s Manual

DELLTechnologies PowerStoreOS Container Based Storage System

FAQ

• Q: What are the benefits of certificate-based authentication for PowerStore CLI commands?
  • A: Certificate-based authentication allows users to run CLI commands with enhanced security and operator role options.
• Q: How can I maintain connectivity to Dell monitoring services after SRS 3.x reaches EOL?
  • A: Upgrade to PowerStoreOS 3.6.1 or migrate to a Secure Connect Gateway before December 31, 2024, to ensure uninterrupted connectivity.

Information

This document describes the new features, changed features, and resolved issues in this release of PowerStoreOS.

• Current Release Version: 3.6.1.0 (Build: 2261541)
• Release Type: Service (SR)

Revision history

Table 1: Document revision history

2261541. A01| March 2024| Corrected the Workaround/Solution for MDT-562457
         A02| April 2024| Added MDT-582560 to Known issues
         A03| May 2024| Added MDT-622297 and MDT-631862 to Known issues

Product Description

The ground-breaking Dell PowerStore storage system achieves new levels of operational simplicity and agility, using a container-based architecture, advanced storage technologies, and intelligent automation to unlock the power of your data.
Based on a scale-out architecture and hardware-accelerated advanced data reduction, PowerStore clusters are designed to deliver enhanced resource utilization and performance that keeps pace with application and system growth.
PowerStore T model clusters provide organizations with all the benefits of an enterprise unified storage platform for block, file and vVol data, while enabling flexible growth with the intelligent scale-up and scale-out capability of appliance clusters.

Features include:

• All NVMe Platform:
  • Active/active architecture
  • Container-based PowerStoreOS
  • Block, File, and vVols
  • NVMe Flash and Storage Class Memory (SCM) media support
  • Front-End Connectivity:
  • FC: 32 Gb NVMe/FC, 32/16/8 Gb FC
  • Ethernet: 25/10 GbE, NVMe/TCP: 25/10 GbE iSCSI and File
  • Compact 2U starting form factor
• Enterprise Data Services and Data Reduction:
  • Inline deduplication and compression
  • Native async replication
  • Snapshots and space efficient thin clones
  • Advanced drive failure protection and sparing technology
• Simple and intelligent Management and Serviceability:
  • Embedded management
  • Built in AI for simple, autonomous storage administration and proactive health analytics
  • CloudIQ
  • VM Visibility
  • New Anytime Upgrade Program
  • Integration with automation framework
• Flexible and Granular Scalability:
  • Scale up in single drive increments, up to 4.7 PBe per appliance
  • Scale out to four appliances, up to 18.8 PBe per cluster
  • Scale down
• VMware: VMware vVols 2.0/VASA 3.0 support

Changed features

The following features have changed in this release:

Table 2: Changed features in PowerStoreOS 3.6.1.0

ability to run PowerStore CLI commands using certificate-based authentication.| Customers can now run PowerStore CLI commands using certificate-based authentication instead of user-based authentication.

NOTE: Certificate-based authentication only provides operator role options.

Support Connectivity| Maintaining connectivity to Dell monitoring and support services after Secure Remote Services (SRS) 3.x reaches End of Life (EOL).| PowerStore clusters with a Support Connectivity setting of Connect Directly will lose connectivity to SRS 3.x after December 31, 2024.

NOTE: To maintain connectivity to Dell monitoring and support services, you must upgrade to PowerStoreOS 3.6.1 or migrate to a Secure Connect Gateway before December 31, 2024.

For more information, see Dell Knowledge Base Article 000222594.

Resolved issues

The following high severity issues are resolved in this release:

Table 3: Resolved issues in PowerStoreOS 3.6.1.0

and an iSCSI host attempts to use the SHA-256 algorithm for authentication, the host fails to log in to the cluster because PowerStore clusters only support the MD5 algorithm for CHAP authentication.
MDT-597799| Connectivity – Hosts| If CHAP is enabled on a PowerStore cluster and the CHAP secret uses the maximum number of characters (64 characters), CHAP authentication fails because the CHAP secret cannot be decrypted correctly.
MDT-607945| Data Protection| If metro replication is configured on a PowerStore cluster and there is a metadata issue that requires recovery, the metro replication session may break when the recovery occurs.
MDT-577741| Data Protection| The Data Path (DP) container on a node may restart because of a software issue that occurs when metro volumes are configured on the affected appliance.
MDT-571128| Data Protection| If a multi-appliance cluster that was created using PowerStoreOS 3.2 or earlier is configured for iSCSI replication, the ALUA access values access_1 and access_2 may be missing for replication tunnel endpoints (TEs) on appliances other than the first appliance in the cluster. The missing ALUA access values can cause Recovery Point Objective (RPO) failures after the cluster is upgraded to PowerStoreOS 3.5 or later.
MDT-522348| Data Protection| If an error occurs while updating the remote system transport type from iSCSI to TCP, the TCP data connection can be created with invalid credentials. As a result, the remote system data connection between the cluster may be lost, which prevents replication sessions from synching.
---|---|---
MDT-573040| Hardware| A failed NVMe drive may cause a node to run out of memory, and the node may reboot if the drive is not replaced.
MDT-442199| Hardware| If Fibre Channel (FC) cables are accidentally connected to the wrong FC ports and then disconnected, the FC ports may remain in a down_in_use state and the generated alert prevents a Non-Disruptive Upgrade (NDU) from being performed.
MDT-589731| Install or Upgrade| If a node is replaced on a PowerStore 500T appliance running PowerStoreOS 3.2 or earlier, performing an NDU to PowerStoreOS 3.5 or

3.6 may fail if the node was not rebooted before the NDU was performed.

MDT-587696| Install or Upgrade| Performing an NDU from PowerStore 2.1. to PowerStore 3.5 or 3.6 on a Unified or NAS-enabled appliance may fail when the memory that is allocated to a Python process surpasses its limit and the Python process is terminated.
MDT-574240| Install or Upgrade| During an NDU, the post upgrade commands may fail with the following error when a TCP replication session with stale internal snapshots is resumed:

Upgrade succeeded. Failed to perform post upgrade commands: java.util.NoSuchElementException

MDT-566742| Install or Upgrade| If a drive firmware upgrade fails, retrying the drive firmware upgrade may fail at the prestage step.
MDT-523154| Install or Upgrade| If the primary M.2 device on a node encounters a hardware issue that causes an NDU to fail, replacing the primary M.2 device may fail to complete successfully and block NDU recovery.
MDT-498068| Install or Upgrade| If a PowerStore cluster was previously upgraded from PowerStoreOS 2.x to PowerStoreOS 3.x, upgrading the cluster to another version of PowerStoreOS 3.x may fail if CHAP was enabled after the initial upgrade to PowerStoreOS 3.x.
MDT-557390| Monitoring| Performance metrics may appear for deleted volumes in PowerStore Manager.
MDT-607714| Notifications and Alerts| The false alert uninitialized power supply unit has failed may be raised and automatically cleared for a power supply on an NVMe expansion enclosure because of the high sensitivity of the hardware component checks.
MDT-593377| Notifications and Alerts| An out of space alert may fail to clear when a PowerStore cluster exits an out of space condition if the cluster entered and exited the out of space condition several times within a short time period.
MDT-583881| Notifications and Alerts| An appliance may incorrectly raise an alert stating that the speed state on an expansion enclosure controller port has changed after performing an NDU to PowerStoreOS 3.6.
MDT-520780| Notifications and Alerts| A PowerStore appliance may fail to raise an alert when it enters resiliency mode after a node is physically removed from an appliance; powered off for more than one minute; or rebooted into service mode.
MDT-581269| PowerStore CLI| An issue with the PowerStore CLI installation wizard may prevent the PowerStore CLI from being installed on some non-English Windows operating systems.
MDT-593311| PowerStore Manager| If the datastore for a virtual machine is on a volume that is deleted and placed in the Recycle Bin, an error occurs when the Virtual Machines page is displayed in PowerStore Manager.
MDT-569169| PowerStore Manager| In rare cases, PowerStore Manager may become unavailable when an appliance is removed from a cluster.
---|---|---
MDT-567995| PowerStore Manager| If a Network Time Protocol (NTP) issue results in metrics data with future timestamps, PowerStore Manager may run out of memory and stop working.
MDT-559513| PowerStore REST API| The Dell Storage Resource Manager (SRM) may be unable to collect metrics for some PowerStore appliances because of an issue that causes bulk metrics to run immediately without waiting for all appliances to complete metrics rollup.
MDT-576006| Security| A memory leak may occur on a PowerStore cluster if a failure occurs when the Key Management Service (KMS) attempts to retrieve audit logs.
MDT-539298| Security| Simultaneous login requests from LDAP users and local users may result in authentication failures. For more information, see Dell Knowledge Base Article 000218822.
MDT-592236| Storage – Block| If a PowerStore cluster is used in a large VDI environment, the cluster may experience both high latency and high CPU utilization after performing an NDU to PowerStoreOS 3.6.
MDT-576464| Storage – Block| A PowerStore appliance may stop functioning and fail to recover automatically if the appliance is handling multiple NVMe reservation commands simultaneously.
MDT-573680| Storage – Block| A PowerStore appliance may experience unexpected node reboots following high levels of vMotion activity.
MDT-572843| Storage – Block| If a PowerStore cluster contains data that is highly deduplicated, large read requests may end up reading the same physical Logical Block Address (LBA) hundreds of times, which can cause high read latency.
MDT-569252| Storage – Block| A false superblock checksum mismatch may cause the following alert to be raised for a storage resource:

A storage resource requires recovery

MDT-515943| Storage – Block| Data Path (DP) may deduplicate data to an invalid target, which can lead to uncorrectable errors.
MDT-592978| Storage – File| Files may become hidden when they are saved to an SMB share from certain applications running on macOS Ventura (version 13).
MDT-592778| Storage – File| If the STIG feature is enabled on a PowerStore cluster running PowerStoreOS 3.5 and an NDU to PowerStoreOS 3.6 is performed, some of the STIG functionality may be disabled during the NDU.
MDT-588962| Storage – File| File system usage alert thresholds are not initialized with default values when a PowerStore cluster is upgraded from PowerStoreOS 2.1 or earlier to PowerStoreOS 3.0 or later. File system usage alert thresholds are set to zero by default, which disables the alerts and prevents customers from being alerted when a file system is running out of space.
MDT-583399| Storage – File| A file system may go offline during heavy workloads such as data migration when the PowerStore environment contains files or a directory that requires more than 1 KB of data to store the Access Control List (ACL).
MDT-609714| Support Connectivity| A failure may occur while enabling Support Connectivity on a PowerStore cluster.
MDT-575019| Support Connectivity| A node may unexpectedly reboot multiple times if the eVE service failed while starting up because the node was manually rebooted or the DNS settings were changed multiple times in succession.
MDT-597359| Virtualization| The PowerStore VASA provider may go offline frequently after performing an NDU to PowerStoreOS 3.5 or 3.6 if the vCenter Server sends multiple VASA certificates to the PowerStore cluster. For more information, see Dell Knowledge Base Article 000221239: PowerStore: VASA Provider Going Offline Frequently After Update to 3.5 or 3.6.
---|---|---
MDT-576805| Virtualization| If a vVol-based VM has read-write vVol snapshots, snapshots of the replicated vVol-based VM may be lost after a failover operation.

Known issues

The following high severity issues remain unresolved in this release:

Table 4: Known issues in PowerStoreOS 3.6.1.0

Issue ID| Functional area| Description| Workaround/Solution
---|---|---|---
MDT-563875| CloudIQ| For PowerStore clusters with large-scale configurations, the collected configuration data may trigger an out-of-memory condition (OOM) in the service container. This issue may cause the configuration deltas capture script to fail and prevent configuration changes from being displayed in CloudIQ for one hour.| None. There are no additional impacts to the cluster service and the CloudIQ issue resolves itself after approximately one hour.
MDT-560828| Connectivity – Hosts| If a host is mapped to multiple vSphere hosts, duplicated items may be added to the host Watchlist with the same hostname and performance metrics, which makes them indistinguishable to the user. However, the metrics for the host are correct and can be treated as one item.| None. This issue does not affect host functionality.
MDT-431035| Connectivity – Hosts| When users configure a metro host as a uniform host, they must configure the host on both metro clusters. If one host is left partially configured or is already configured on one metro cluster, ESXi may report inconsistent path states.| Complete the configuration of the host on both metro clusters. For more information, see Dell Knowledge Base Article 000202793: PowerStore: Metro volumes are inconsistently presented to ESXi when uniform hosts are partially configured.
MDT-367147| Connectivity – Hosts| When Windows clustered hosts are attached to a multi-appliance PowerStore cluster and LUNs are migrated from one appliance to another, the LUNs do not migrate back to the original appliance unless the hosts are rebooted prior to the migration back.| Avoid migrating LUNs back and forth between appliances.
MDT-273055| Connectivity – Hosts| A user can view internal host and volume mappings using PowerStore Manager, but mapping or unmapping operations cannot be performed from PowerStore Manager.| Mapping and unmapping operations can be performed using the PowerStore REST API interface.
MDT-272385| Connectivity –

Networks

| Large amounts of external IPv6 traffic exposed to Ethernet ports can cause iSCSI performance degradation.| To avoid the problem, use separate VLANs for Storage networks and networks that contain a lot of non- storage related IPv6 traffic. Also, route management VLANs and other VLANs with large amounts of IPv6 traffic to dedicated storage ports.
---|---|---|---
MDT-211358| Connectivity – Networks| PowerStore Discovery IP by Zeroconf technology stopped advertising.| Perform one of the following actions to work around this issue:

·         Wait a few minutes for the state of the system to change and try again.

·         Reset a drive, mezzanine card, or I/O module if available and wait for few minutes and try again, and then reboot the primary node.

MDT-558759| Data Collection| I/O spikes may be observed when performance metrics are collected.| None. This issue automatically clears over time.
MDT-582560| Data Protection| When the Host Connectivity Options for hosts that use metro volumes are set to Co-located with both systems, an unexpected node reboot may occur if hosts write to the same LBA simultaneously.| Change the Host Connectivity Options from Co-located with both systems to Host is co- located with this system or Host is co-located with the remote system. For more information, see Dell Knowledge Base Article 000223948.
MDT-569362| Data Protection| The replication of vVol-based virtual machines may pause, and the following error may occur when the replication of a large number of vVol-based virtual machine (VM) clones is configured to use the same storage container where the cloned VM is provisioned:

Request to remote system with management address failed with error: Family members count 1001 exceeds maximum amount 1000 of objects.

| Contact your service provider for assistance.
MDT-563952| Data Protection| When replicating to a PowerStore 500T model cluster from a different type of PowerStore model cluster, the number of snapshots on a destination volume may exceed the snapshot limit on the PowerStore 500T model cluster.| None
MDT-557273| Data Protection| If the serial number of a PowerProtect DD system changes, modifying the PowerProtect DD remote system in PowerStore may fail because of a serial number mismatch error.| Contact your service provider for assistance
MDT-555345| Data Protection| A backup operation from a PowerStore cluster to a PowerProtect DD system may fail with an error message telling the user to check connectivity, instead of checking if a storage unit exceeded its hard quota.| If connectivity issues are not present, check the capacity usage of the storage unit in the affected PowerProtect DD system to determine if it has exceeded its hard quota.
---|---|---|---
MDT-525692| Data Protection| An issue in the Refresh Using Related Volume dialog box may cause the Refresh button to become active when the Create a backup snapshot on the volume being refreshed check box is cleared, even if a volume is not selected from the Select volume to refresh from drop-down menu. Clicking Refresh fails when a volume is not selected from the Select volume to refresh from drop- down menu.| If clicking Refresh fails, ensure that a volume is selected from the Select volume to refresh from drop-down menu, and click Refresh again.
MDT-520673| Data Protection| The status of Instant Access sessions may change to Failed after a node reboot occurs.| A PowerStore cluster does not support high availability (HA) between nodes for Instant Access sessions. For more information, see Dell Knowledge Base Article 000208509: PowerStore: Instant Access session will show failed state after node reboot.
MDT-518764| Data Protection| When a remote system connection to a PowerProtect DD system uses replication-tagged ports, a node panic may occur if the replication-tagged ports are moved to different physical ports that do not have network connectivity to the PowerStore cluster. This situation can cause the status of remote system Data Connection to change to Status not available. The Verify and Update command may take longer to run when the remote system Data Connection is in this state.| Before moving the replication tagged port to a new port, ensure that the new port has network connectivity to the PowerStore cluster.

If the status of the remote system Data Connection changes to Status not available, perform the following steps:

1.        Select the check box of the remote system and click Modify.

2.        Change the Data Transfer Address to the management IP address of the PowerProtect DD system and click Modify.

3.        Select the check box of the remote system and click Modify.

4.        Change the Data Transfer Address of the remote system back to the original IP address and click Modify.

5.        Select the check box of the remote system, select More Actions > Verify and Update, and click Verify and Update.

MDT-517565| Data Protection| Attempting to back up more than 125 volumes simultaneously to a PowerProtect DD system may result in failed backup sessions because of Control Path (CP) limitations.| When backing up more than 125 volumes to a PowerProtect DD system, stagger the backups to prevent more than 125 volumes from being backed up simultaneously.
MDT-511829| Data Protection| Instant Access is not supported on ESXi

hosts with VMFS datastores. Attempting to mount an Instant Access session resource as a VMFS datastore on an ESXi host may result in an I/O error.

| Perform a retrieve operation on the remote snapshot, then create and mount a thin clone of the snapshot on the ESXi host. For more information, see Dell Knowledge Base Article 000213256: PowerStore: Mount of an Instant Access session resource fails with I/O Error.
---|---|---|---
MDT-496905| Data Protection| When a user pauses and resumes a metro volume, the status of the metro volume may change to Fractured.| None. The status of the metro volume returns to Operating Normally (active/active) after the system performs self-healing.
MDT-438935| Data Protection| When multiple remote systems are connected to a PowerStore cluster, reconfiguring the NAS Mobility Network (NMN) after performing a Management Network Reconfiguration (MNR) might fail with the following Data Path (DP) network error if one of the remote systems is not responding:

Modify cluster IP of dpNetwork: {dpNetworkID} failed

| Resolve the issue with the remote system that is not responding or continue to use the other remote systems that are responding.
MDT-416804| Data Protection| An end metro volume operation may take longer to complete when it is issued on the remote end during self-heal sync of a metro session. The session eventually does complete successfully.| Pause the sessions and then perform the end metro volume operation.
MDT-414572| Data Protection| The status of replication sessions may be incorrect on the destination PowerStore cluster if the source PowerStore cluster is down.| See the Remote System related information and alerts on the destination PowerStore cluster for up-to-date details and suggested recovery steps.
MDT-312619| Data Protection| When replication IP addresses change, they are not automatically detected. As a result, the status of a replication session changes to System Paused.| Run the Remote System Verify operation to update the connectivity matrix. The next synchronization cycle will update the replication IP addresses.
MDT-309462| Data Protection| If the user pauses a replication session on the destination cluster during an NDU of the source cluster, a post-NDU state may occur where the source replication session is in System Paused state and the destination replication session is in Paused state.| Pause the replication session on the source cluster, where the session is in System Paused state, and then resume the replication session.
MDT-299182| Data Protection| In a large configuration, pausing or deleting a replication session may take longer when the session is queued for synchronization.| None
MDT-293141| Data Protection| If a volume is restored from a PowerProtect DD system to a PowerStore cluster where the volume was deleted or never existed, the Logical Used and Provisioned capacities that are shown in PowerStore Manager may display the same values.| The correct logical volume usage is displayed when the volume is mapped to a host. For more information, see Dell Knowledge Base Article 000208504: PowerStore: After retrieving PowerStore backups from Data Domain, the Logical Used and Provisioned capacities show the same values.
MDT-242915| Data Protection| When a replication destination cluster loses connectivity to the source cluster, the system issues specific alerts, but the state of the replication session may still display Operating Normally.

Administrators should check the alerts that are issued by the remote system and resolve the connectivity issues to recover replication sessions.

| None
---|---|---|---
MDT-133408| Data Protection| When a volume group is created with member volumes and is protected by a protection policy that includes a replication rule, the volume group members may be delayed in being replicated to the remote system until the next recovery point objective (RPO) cycle. The default replication rule RPO value in PowerStore Manager is one hour, but it can range from five minutes to 24 hours.| If the remote system connection is healthy, the member volumes should automatically be replicated to the destination during the next RPO cycle. To update the member volumes on the destination cluster before the next RPO cycle, go to Protection > Replication, select the volume group session, and perform a Synchronize operation.
MDT-333618| Discovery| In certain network environment configurations, the system is unable to discover unconfigured appliances through the network when using static IP addresses for appliance discovery.| The initial discovery of PowerStore appliances works with either the native Zeroconf-capable browser such as Bonjour or through the PowerStore Discovery Tool.
MDT-631862| Documentation| The Dell PowerStore 3.x Security Configuration Guide references that PowerStore drives are FIPS 140-2 compliant, when the drive may not be FIPS 140-2 compliant.| See the PowerStore Simple Support Matrix for federal compliance of the PowerStore drives.
MDT-579035| Documentation| Revision A05 of the PowerStore Protecting Your Data guide does not list openSUSE Leap 15.5 as a supported operating system for Metro protection.| See the latest version of PowerStore Simple Support Matrix for a complete list of operating systems that support Metro protection.
MDT-439758| Hardware| A healthy DIMM module in an NVMe expansion enclosure may be marked as missing or failed.| For more information and steps to avoid this issue, see Dell Knowledge Base Article 000204824: PowerStore: A healthy DIMM module in an NVMe expansion enclosure may be marked as missing/failed.
MDT-362070| Hardware| If one node comes up 120 seconds before the other node after a dual node reboot, the first node may reboot again to start auto- recovery.| None
MDT-286236| Hardware| In rare cases, a node that has been powered off does not remain powered off.| Power off the node again.
MDT-117061| Hardware| When creating a cluster, the hardware status is incorrectly indicated as not configured for clustering.| Wait a few minutes and try creating the cluster again.
MDT-558763| Import| When adding a remote system in

PowerStore Manager for importing external storage, if the user selects a type of system that supports CHAP, selects Single or Mutual from the Session CHAP Mode drop-down menu, and then changes the type of system to PowerMax/VMAX, some of the CHAP username and password fields may remain active, even though these CHAP settings are not supported on the PowerMax/VMAX.

| Click Cancel and Discard to close the

Add Remote System panel. Then, click Add Remote System again and select PowerMax/VMAX from the Type drop- down menu.

---|---|---|---
MDT-356233| Import| The Inband Migration Tool (IMT) Copy Progress Percentage and Estimated Completion Time display an incorrect progress of 0% on the File Imports tab of the Import External Storage page in PowerStore Manager.| Use the PowerStore REST API command

/file_import_session/{id} to view the current_operation_progress_percentage and estimated_completion_timestamp for a file import session. If needed, contact your service provider for assistance.

MDT-295614| Import| Performing a non-disruptive import using a Windows host plug-in may fail on a PowerStore cluster with multiple storage networks. If multiple storage networks are configured, all the iSCSI portals are returned on discovery, but iSCSI portal filtering is not supported by PowerStore.| Use one of the following workarounds:

·         If possible, remove all the storage networks except for the storage network on which the Windows host resides.

·         Perform an agentless import.

MDT-252560| Import| Importing a volume may fail with the following error:

Import failed because the mirror service could not be enabled on the Destination volume

| Run the verify operation on the remote system to recover data connections.

When the verify operation on the remote system completes successfully, retry the import operation.

MDT-562457| Install or Upgrade| When performing a Data In Place (DIP) upgrade in a PowerStore cluster with two appliances, REST commands may fail and PowerStore Manager access to the affected cluster may become unavailable if the secondary appliance is powered down.| Power up the secondary appliance to restore PowerStore Manager access and complete the DIP upgrade.
MDT-534404| Install or Upgrade| The Pre-Upgrade Health Check may fail if the configuration capture request takes more than four minutes to complete.| Wait 10 minutes and rerun the Pre-Upgrade Health Check.
MDT-461643| Install or Upgrade| Performing a Pre-Upgrade Health Check on a PowerStore cluster with NVMe expansion enclosures may fail with the following error:

pre fw upgrade state is FW_ UPGRADED_NEEDED_NOT_AVAILABLE

| Check the status of the Access Modules (AM) on the expansion enclosures. If there are no alerts, retry the Pre-Upgrade Health Check. If the problem persists, contact your service provider for assistance.
MDT-401600| Install or Upgrade| If the Health Check package was uploaded on the PowerStore cluster before a new appliance was added to the cluster, the Health Check package is unable to run for the newly added appliance.| Install the same Health Check package again.
MDT-337169| Install or Upgrade| NAS services might fail to start during the

initial installation and configuration of PowerStore appliances or during a PowerStore software upgrade.

| If NAS services fail to start during the initial installation and configuration of PowerStore appliances, try to install NAS again as described in step 4 of Dell Knowledge Base Article 000130232: PowerStore: PowerStore Manager displays ‘NAS Installation has failed’ when trying to configure PowerStore appliance(s).

If NAS services fail to start during a PowerStore software upgrade, contact your service provider for assistance.

---|---|---|---
MDT-265301| Install or Upgrade| When performing an NDU on a PowerStore T model appliance running in Unified mode, the inter-node iSCSI connection might fail when one node is rebooted, which can cause the NAS to panic.| Allow NAS HA services to take care of uninterrupted NAS operations during an NDU. If data unavailability occurs after the NDU is complete, contact your service provider for assistance.
MDT-156905| Install or Upgrade| After performing an NDU, the NAS servers may not fail back automatically to the original node.| Manually fail back the NAS servers using PowerStore Manager.
MDT-135505| Install or Upgrade| Drive firmware updates may not get applied to a single drive or multiple drives within the appliance. This problem may occur because of a timing issue between the nodes of the appliance.| Download and install the drive firmware updates again to ensure it gets updated on the drives that were missed in the previous attempt.
MDT-508449| Internal Migration| The Migration Tool (IMT) import session progress might stop updating if a high availability (HA) event occurs on a multi- appliance PowerStore cluster.| Monitor the progress of the import in PowerStore Manager using the Import External Storage page or import logs.
MDT-484929| Internal Migration| When VMware vSphere 8.0 hosts are connected to a PowerStore cluster, attempting to migrate powered-off vVol- based VMs that use NVMe storage may fail.| 1.        Power off the affected VMs.

2.        Unregister the VMs from vCenter Server.

3.        Migrate the VMs.

4.        Re-register the VMs with the vCenter Server.

5.        Power on the VMs.

MDT-447218| Internal Migration| Because of VMware DCPN case number 00105337, vVol migration may fail if a vVol is migrated to an appliance on which it was previously located.| For more information and steps to avoid this issue, see Dell Knowledge Base Article 0000204524: PowerStore: When using PowerStoreOS 3.x, vVol Migration may fail if a vVol is migrated to an appliance on which it was previously located.
MDT-359565| Internal Migration| PowerStore Manager does not prevent a user from clicking the Delete button during a migration cutover. However, deletion during cutover is not supported and the migration session is not deleted. The delete operation results in a backend error.| None
MDT-285903| Internal Migration| If a migration fails due to a high availability (HA) event, a system snapshot may remain on the migration source object.| Contact your service provider to clean up the remaining snapshot.
MDT-239624| Internal Migration| When a VM is being deployed from a VM template, vSphere uses the clone operation to provision vVols of the newly created VM. PowerStore internally translates it into space-efficient clones, which is the same process that is used for vVol snapshots. When a Data Path (DP) family size limit exists, a base vVol cannot have more than a 1000 derivatives, such as snapshots and space-efficient clones. This fact results in the following limitations:

·         Cloning from a VM template fails if a base vVol of the VM template already has 1000 derivatives.

·         When the maximum number of family members limit is reached, vVol migration cannot occur because it must create an internal vVol snapshot, which fails.

·         All vVols are placed to the same appliance in a multi-appliance cluster, so the load is not balanced between appliances.

| Use several VM templates created from the same OVF template so that each VM template is used for a reasonable number of VM clones to avoid reaching the DP family limit. This action allows vVols to be distributed between appliances in a multi- appliance cluster for load balancing.
---|---|---|---
MDT-140927| Internal Migration| After a volume is migrated, a decrease in capacity and performance metrics reporting may occur for that volume at the end of the migration operation.| None
MDT-533093| Monitoring| During an NDU, a large increase in bandwidth and IOPS may be displayed on the cluster performance chart while performance metrics are recalculated.| None. The bandwidth and IOPS values in the cluster performance chart return to normal after the NDU is completed.
MDT-249810| Monitoring| If the total system capacity is below 10 TB, free page tables are not replenished as this action would consume space that might not be required and waste free pages. The lack of free pages in the table causes a performance impact for writes. For better performance, more capacity should be added.| Ensure that the system capacity is higher than 10 TB.
MDT-55667| Monitoring| Storage container capacity metrics are not available until I/O operations begin.| Look at the capacity metrics again after I/O operations on the storage container have been performed.
MDT-454979| Notifications and Alerts| If a PowerStore cluster has a very large number of initiators, hardware-related alerts may be processed more slowly than normal. This issue does not affect PowerStore Manager operation.| None. A 10–30 minute delay may occur when hardware-related alerts are raised and cleared.
MDT-437349| Notifications and Alerts| An alert regarding node failover during an NDU may fail to clear automatically.| Contact your service provider for assistance with manually clearing the alert.
MDT-566899| PowerStore

Manager

| If appliances with different branding are used to create a cluster, PowerStore Manager may display different branding after a failover of the primary appliance in the affected cluster.| Contact your service provider before

creating a cluster.

---|---|---|---
MDT-564033| PowerStore Manager| Filtering replication objects in the Source Resource column of the Replication page is not supported in PowerStore Manager.| Instead of filtering the objects in Source Resource column, sort the objects in the Source Resource column by clicking the column heading
MDT-546968| PowerStore Manager| In rare cases, a Hard limit reached event may be raised incorrectly, instead of a Hard limit exceeded event.| None
MDT-545229| PowerStore Manager| If a huge number of User Quotas are defined on a PowerStore cluster, it may take time for these entries to be displayed in PowerStore Manager.| None
MDT-544803| PowerStore Manager| When hovering over the name of a rule in the Replication Rule column on the Protection Policies page, PowerStore Manager may incorrectly display the tooltip [object Object] if the name of the rule is wider than the Replication Rule column.| To view the full name of the rule, increase the width of the Replication Rule column.
MDT-496618| PowerStore Manager| The Volume Mappings counter, which is displayed on the Mapped Volumes tab of the Host Information page, may fail to update after a mapping or unmapping operation is performed. However, the mapping or unmapping operation is performed correctly, and the information that is displayed in the table on the Mapped Volumes tab is correct.| Go back to the Host Information page, and then click the host again to display the updated Volume Mappings counter on the Mapped Volumes tab.
MDT-362809| PowerStore Manager| When a certificate file is opened in some text editors, a newline character (\n) is appended to the end of the certificate. A certificate becomes invalid when a newline character (\n) is appended to the end of the certificate.| Open a certificate file using a text editor such as Windows Notepad, which does not append a newline character (\n) to the file.

NOTE: If you open the certificate file using an editor such as Notepad++, you must manually remove the newline character (\n) from the certificate file.

MDT-261523| PowerStore Manager| If Mozilla Firefox is used to upgrade a PowerStore cluster to PowerStoreOS 2.0 or later, a browser cache issue causes the names of some buttons and labels to be displayed as object names instead of the correct button or label names.| Use a different browser such as Google Chrome or Microsoft Edge, or clear the browser cache in Firefox. For instructions, see How to clear the Firefox cache on the Mozilla support site.
MDT-118394| PowerStore Manager| Uploading a software upgrade package that is larger than 4 GB may fail when using an older version of Microsoft Edge.| Use a different supported browser to upload the software upgrade package.
MDT-622297| PowerStore

REST API

| The PowerStore REST API returns that PowerStore drives are Level 2 FIPS 140-2 compliant, when the drive may not be Level 2 FIPS 140-2 compliant.| See the PowerStore Simple Support

Matrix for federal compliance of the PowerStore drives.

---|---|---|---
MDT-558612| Storage – Block| In rare cases, a known issue in Python 2.7 may cause a kernel dump to be unexpectedly generated when an internal code exit occurs.| None
MDT-528613| Storage – Block| A node may reboot unexpectedly if two drives fail while the space utilization is very high.| None. The node is expected to recover successfully after it reboots.
MDT-284475| Storage – Block| After mapping a LUN or creating a VM, the LUN or VM appears correctly, but it is inaccessible to hosts.| Unmap the non-working LUN and map it again. For VMs, delete the non-working VM and recreate it.
MDT-242368| Storage – Block| Unable to delete a volume group after deleting a replication session, due to replication session system snapshots that have not been deleted.| Contact your service provider for the procedure to delete the snapshots.
MDT-545358| Storage – File| PowerStore does not enforce the NFS exports advisory limit of 8000 exports per appliance. When this limit is exceeded, the user may experience slowness while accessing these NFS exports.| Remove NFS exports to reduce total count and keep the number NFS exports below the advisory limit of 8000.
MDT-524910| Storage – File| Performing a bulk delete of file system snapshots may fail.| Retry the bulk delete operation.
MDT-515997| Storage – File| If a PowerStore cluster is under a significant load, gaps may occur in the SDNAS historical metrics that are displayed in the PowerStore REST API and PowerStore Manager.| None
MDT-458871| Storage – File| A file system replication session may become out of sync because of a missed recovery point objective (RPO).| Pause and resume the replication session.
MDT-361933| Storage – File| File system asynchronous replication may use an existing destination NAS server in non-production mode as replication target if it shares the same name as the source NAS server.| Rename the destination NAS server before setting up remote protection for the new NAS server that has the same name. Also, ensure that a NAS server in non-production mode with the same name does not exist.
MDT-314995| Storage – File| In a Unified appliance cluster, after the system exits from an out of space scenario, some of the NAS Servers may fail to return to Read-Write mode when the Control Path (CP) restarts.| Contact your service provider for assistance.
MDT-195652| Storage – File| Unable to remove the description from an NFS Export.| To change the description, overwrite the current description with a blank character space. Save the NFS Export with no description, and then go back and update the description.
MDT-147688| Storage – File| SMB clients cannot connect to a NASserver when the NAS server name is more than 15 characters. This issue occurs because the SMB server name takes the NAS server name by default.| Specify a NAS server name with 15 characters or less or specify an SMB server name with 15 characters or less. To modify the name, select all the characters in PowerStore Manager, and type the new name.
---|---|---|---
MDT-139095| Storage – File| The following error message was returned while trying to create or delete a snapshot:

Addition [or deletion] of NFS Export failed due to [The path [path_name] was not found on the system and cannot be exported.]

| Wait for several minutes and try the action again.
MDT-137232| Storage – File| In rare cases, NAS services could be disrupted if the appliance is close to reaching the maximum used data capacity.| Take one of the following actions on the appliance where the NAS service is installed:

·         Add capacity to the appliance by adding drives or expansion enclosures.

·         Increase free capacity on the system by deleting unused snapshots or volumes.

·         Migrate data off the system.

MDT-116676| Storage – File| The major alert NAS node is down may be seen after changing the Cluster MTU value.

Changing the MTU value may cause degradation or disruption in NAS file services.

| Wait for the alert to clear before performing any further NAS file operations.
MDT-573046| Support| The service command svc_diag list–battery may report an incorrect Fault Status Register (FSR) status for a BBU in a PowerStore 500T model appliance.| Contact your service provider to run an internal command to view the correct BBU status.
MDT-345741| Support| The upload support materials bundle (svc_dc upload) service script may fail with the following error:

No path was found for the data collection

| Wait five minutes and retry the svc_dc upload service script.
MDT-289213| Support Connectivity| If a node failover or node reboot occurs while support data is being uploaded, it may be unclear if the support data was successfully uploaded.| Upload the support data again after the failover or reboot is complete.
MDT-256673| Support Connectivity| Cannot enable the Secure Remote Services (SRS) gateway on an appliance. Version 3.48 of the SRS gateway has an issue with connectivity from some PowerStore systems.| Use a different version of the Secure Remote Services gateway.
MDT-148608| Support Connectivity| Attempting to send a collection of support materials fails when there is insufficient free space on the system.| Delete old data collections to increase the available space on the system so Support Connectivity can make a copy of the support materials to upload.
MDT-547360| Virtualization| vVol datastores may be inaccessible in an IPv6 environment if the vCenter Server and ESXi hosts are running vSphere 8.0 Update 1.| Upgrade the vCenter Server and ESXi hosts to vSphere 8.0 Update 2 or later.
---|---|---|---

Limitations

The following limitations exist in this release:

Table 5: Limitations in PowerStoreOS 3.6.1.0

16 import sessions is supported in the Ready-For-Cutover state. When several dozen import operations are run back-to-back, intermittent failures of alternate import sessions may occur. If this issue occurs, do the following:

·         Remove the remote system and then add it again.

·         Run fewer sets of imports (16 or less) at a time. It is recommended to start all these import sessions with automatic cutover turned off.

·         Once all imports have reached the Ready-For-Cutover state, do a manual cutover.

After one set of imports is completed, run the next set of imports after a delay of ten minutes. This delay allows enough time for the system to cleanup any connections to the source system.

Install or Upgrade| Performing an NDU from a Solaris host is supported with Solaris native MPxIO starting in Oracle Solaris 11.4 SRU 35. Contact your service provider for upgrade support on earlier Solaris versions or Solaris updates.
Internal migration| Before performing an internal migration, see Dell Knowledge Base Article 000105896: PowerStore: Supported Host OS for Non- Disruptive migration of Storage resources for a list of supported configurations and limitations.
Monitoring| Starting a new volume mapping job, while another volume mapping job is still in progress, may cause mapping failure due to duplicate Logical Unit Numbers.
PowerStore Manager and PowerStore REST API| Use the PowerStore cluster management IP address to access the PowerStore Manager and the PowerStore REST API. The appliance management IPs are intended primarily for service use.
Storage – File| Customers that use an SFTP client to access file data on a PowerStore T model cluster may experience access issues after upgrading to PowerStoreOS 3.0 or later because support for the following insecure key exchange (KEX) algorithms have been removed:

·         diffie-hellman-group-exchange-sha1

·         diffie-hellman-group14-sha1

The access issue occurs if the customer is using an outdated SFTP client that only supports the KEX algorithms that were removed from the PowerStoreOS. To resolve this issue, the customer must upgrade to a newer version of the SFTP client.

Storage – File| The maximum supported number of Access Control Entries (ACE) that can be added to an Access Control List (ACL) for an SMB share is 650.
Storage – File| If an Access Control List (ACL) is set on an SMB share using a security identifier (SID) that can be resolved to a user, the SID is converted to the format domain\username.
Storage – File| Cloning write-once read-many (WORM) files on a File-Level Retention (FLR) enabled file system is not supported.
Storage – File| PowerStoreOS 3.5 and 3.6 support up to 125 File Systems per NAS Server. If you are upgrading from an earlier version of PowerStoreOS to PowerStoreOS 3.5 or 3.6, and you have more than 125 File Systems on a NAS Server, PowerStore supports up to the number of File Systems that you have created. However, if you delete a File System from the NAS Server, you cannot create another File System on the NAS Server until there are less and 125 File Systems.
---|---
Support Connectivity| Support Connectivity cannot be enabled on PowerStore models that are configured with IPv6 for the management network. Support Connectivity is not supported over IPv6.
Virtualization| Version 2.0.0.33 of the PowerStore Storage Replication Adapter (SRA) for VMware Site Recovery Manager is not supported with PowerStoreOS 3.5 or later. For more information, see Dell Knowledge Base Article 000213328: PowerStore: SRM discover devices failed when SRA version 2.0.0.33 is used with PowerStoreOS version 3.5 or later.

Environment and system requirements

Host configuration requirements

• For host connectivity to PowerStore appliances, see the E-Lab Navigator Host Connectivity Guides.

Simple Support Matrix

• For compatibility and interoperability information, see the PowerStore Simple Support Matrix, which can be downloaded from dell.com/powerstoredocs.

Installation and upgrade considerations

Installation considerations

Before installing a PowerStore cluster, review the PowerStore Deployment Checklist, which can be downloaded from dell.com/powerstoredocs. You are ready to install your system when you have completed all the planning steps in the deployment checklist.

Upgrade considerations

NOTE: PowerStoreOS 3.6 is not supported on PowerStore X model appliances.

Before upgrading the PowerStore cluster, review the PowerStore Software Upgrade Guide, which can be downloaded from dell.com/powerstoredocs. For a full list of supported PowerStore upgrade paths, see Dell Knowledge Base Article 000175213: PowerStore: PowerStoreOS Matrix.

For systems with Support Connectivity enabled, upgrade packages can be automatically downloaded to the system when they become available on Support. The PowerStore upgrade packages are also available for download from the Support site. However, the upgrade packages can only be accessed when you are logged into the Dell Support Site using an account that is associated with the PowerStore cluster.

NOTE: When using the PowerStore CLI, it is recommended that you use the PowerStore CLI version that accompanies each major release of PowerStoreOS. For example, use PowerStore CLI version 3.0 with PowerStoreOS versions 3.0 and 3.2, use PowerStore CLI version 3.5 with PowerStoreOS version 3.5, and use PowerStore CLI version 3.6 with PowerStoreOS version 3.6.

Download the upgrade package for the PowerStore T model cluster

If automatic download is not enabled, download the PowerStore T upgrade package from the Support site.

1. Access the Support site by opening a browser and going to the Dell Technologies Support site.
2. Sign into your support account that is associated with the PowerStore cluster.
3. Go to the product page for your PowerStore T model, and then go to Drivers & Downloads.
4. Identify the PowerStore upgrade package to download by carefully reading the package descriptions. The PowerStore T upgrade package is named PowerStore T OS Upgrade 3.6.1.0-2261541.
5. Click Download to download the PowerStore upgrade package. The name of the downloaded PowerStore T upgrade package is PowerStoreT-3.6.1.0-2261541-retail.tgz.bin.
6. See the PowerStore T model Software Upgrade Guide for instructions on how to upgrade your PowerStore T model cluster.

Where to get help

• The Dell Technologies Support site contains important information about products and services including drivers, installation packages, product documentation, knowledge base articles, and advisories.
• A valid support contract and account might be required to access all the available information about a specific Dell Technologies product or service.

Notes, cautions, and warnings

• NOTE: A NOTE indicates important information that helps you make better use of your product.
• CAUTION: A CAUTION indicates either potential damage to hardware or loss of data and tells you how to avoid the problem.
• WARNING: A WARNING indicates a potential for property damage, personal injury, or death.

© 2024 Dell Inc. or its subsidiaries. All rights reserved. Dell Technologies, Dell, and other trademarks are trademarks of Dell Inc. or its subsidiaries. Other trademarks may be trademarks of their respective owners.

Read User Manual Online (PDF format)

Read User Manual Online (PDF format)  >>

Download This Manual (PDF format)

Download this manual  >>