Resolved Issues in 5.1.0-SP60

• ORION-207864: Updated cluster upgrade routines to prevent removal of user-supplied monitored properties in an event definition during an upgrade.

• ORION-206577: Made updates to automatically enable OpenSM service upon completion of the VAST Easy Install procedure for clusters with InfiniBand internal networking.

• ORION-201844: Updated the ‘wrong count of dnode ipmi ips error message displayed if you supply an incorrect number of DNode IPMI IPs during VAST Easy Install to correctly state the suggested number of IPs.

• ORION-201576: Resolved an issue that could cause a Failed to copy config source /etc/sysconfig/network-scripts/ifcfg-*. Error: cp: target '/tmp/inactive//etc/sysconfig/network-scripts' is not a directory error during an OS upgrade on a cluster where IPv6 addressing was used for some of the DNodes.

• ORION-201373: Improved validation of IP addresses entered in the VAST Easy Install fields to prevent entry of duplicate IPs.

• ORION-182057: Resolved an issue that could cause the cluster upgrade task to fail with the time data '' does not match format '%Y-%m-%dT%H:%M:%S' error.

• ORION-232011: Resolved an issue where a DTray replacement procedure could not be completed due to the 'NoneType' object has no attribute 'target_version' error occurring during BMC upgrade.

• ORION-210991: Added the IPMI netmask field to the Configure new DBoxes step in the VAST Web UI’s DBox migration wizard so that you can specify an IPMI subnet for the IPMI gateway.

• ORION-204099: Updated the DNode replacement logic to ensure that it correctly determines the node to be replaced based on the node's IPv6 address.

• ORION-162385: Resolved an issue where during a DNode replacement procedure, a failed to run configure network on <SSH:_host:"ipv6%port;_port:22,_timeout:30 error occurred when trying to rerun the cluster networking configuration script (configure_network.py) after the node reboot.

• ORION-214087: Improved the way VMS checks for the OpenSM service on a cluster where both external and internal interfaces are InfiniBand to properly handle configurations where the OpenSM master is outside the cluster.

• ORION-207867: Resolved an issue that could, in some specific cases, cause a vast-int-vlans.service assertion error when running the cluster networking configuration script (configure_network.py) on Ceres CNodes.

• ORION-203770: Updated the cluster networking configuration script (condifure_network.py) to prevent failures due to the unrecognized arguments: --ext-slave-ifaces error.

• ORION-203289: Improved port status checks to eliminate a flow where some hosts could encounter NFS connectivity issues following a switch replacement.

• ORION-202586: Enhanced the logic that assigns the technician port IP during cluster networking configuration to block assignment of a technician IP that conflicts with the existing customer management network.

• ORION-233592: Resolved an issue that could cause multiple CNode containers to restart with the assertion failed: (pre_compression_encode == current_chunk->_dr_header->pre_compression_encode)  error followed by a DATA_REWRITE deny list alert on an attempt to run data rewrite after having enabled similarity-based data reduction.

• ORION-211569: Eliminated a flow that could end up with an ESTORE MAINTAIN_MD_BLOCKS deny list due to the timeout expired for @life_type=...>13,life_gen=<...> (CONTENT_DEFRAG_MD_READ) error.

• ORION-205723: Resolved an issue that could cause an ESTORE TREE_UNLINKER deny list alert due to a lifecycle configuration size mismatch, read_size=0 expected_size=124 error.

• ORION-204347: Resolved an issue that could cause an ESTORE CPOI_SCRUBBER deny list alert on the cluster.

• ORION-199248: Made updates to eliminate a flow that could result in a HANDLE_OPTIMIZE deny list alert on the cluster.

• ORION-193665: Enhanced handling of scenarios where a directory is being deleted by both the Trash Folder feature and the Lifecycle Rules feature at the same time, which could previously result in a ESTORE TREE_UNLINKER deny list alert.

• ORION-207806: Updated quota calculation routines to avoid throwing an Error: division by zero error on attempts to display a quota with a hard limit of 0 (zero).

• ORION-206297: Resolved an issue that could prevent VMS to correctly update the quota capacity percentage (Element Store -> Quotas) although this information was reported correctly in Uplink.

• ORION-215532: Updated the logic behind the Tenant field displayed for a lifecycle rule (in VAST Web UI: Element Store -> Lifecycle Rules, in VAST CLI: lifecyclerule list and lifecyclerule show command output) to ensure that the field shows a tenant. Prior to this change, the field could remain empty.

• ORION-223138: Removed the option to set total QoS limits from QoS settings in VAST Web UI.

• ORION-216774: Resolved an issue that could cause incorrect setting of the directory owner for child directories of a parent that had no default ACL on views with the SMB and S3 protocols enabled and the Mixed Last Wins or SMB security flavor set.

• ORION-216063: Resolved an issue which could cause a client connecting through the VAST NFS Driver to occasionally get stuck when trying to read a file.

• ORION-169644: Resolved an issue that could cause a delete pending is already set in remove flow for phandle=<...> error on a CNode when performing an NFSv4 delete operation.

• ORION-218864: Resolved an issue where the audit log for an NFSv3 SETATTR request changing directory permissions showed the object type as UNKNOWN.

• ORION-212855: Updated handling of user-supplied entries in a share-level ACL to avoid a flow where an Everyone group entry could disappear if placed after an existing group entry in the ACL.

• ORION-210311: Improved handling of group historical SIDs to avoid a situation where following a domain migration, some of the users were denied access while others could proceed as expected.

• ORION-209149: Updated the logic behind the VAST Web UI and VAST CLI options that enable or disable mandatory SMB signing so that the options can be used to change the feature state as expected.

• ORION-203797: Enhanced handling of group SIDs when using SMB native authentication to resolve an issue that could cause permission denied errors when trying to access an SMB share with share-levels ACLs enabled.

• ORION-197355: Enhanced SMB session management routines to avoid raising multiple NV hash remove failed alerts when processing two subsequent SMB sessions for the same durable or persistent handle.

• ORION-196448: Enhanced handling of historical SIDs to eliminate a flow that could result in a change of the group referenced in the share’s ACE upon migration of the user group from one Active Directory domain to another.

• ORION-217661: Removed a restriction that caused VAST Cluster to reject zero-sized parts in a multipart upload with a 400 Bad Request error.

• ORION-204776: Optimized processing of S3 listing requests to avoid increased read latency at the time when the cluster is busy handling multiple listing requests with a very large number of items.

ORION-200613: Eliminated a flow where attempts by some users to disable Access-Based Enumeration (ABE) for a view could result in an UserDBResultCode.INVALID_ARGUMENT error.

• ORION-197543: Optimized ABE routines to avoid performance degradation during ABE permission checks.

• ORION-216244: Made updates to prevent generating duplicate entries for the same item in VAST Catalog.

• ORION-201923: Resolved an issue that could cause a Please collect a bundle and three 30-mins hubbles: with row_id=<...> handle=<...>, and handle=<...> alert when the cluster has VAST Catalog enabled.

• ORION-173787: Added a limitation on the number of filesystem snapshots that can be created per VAST Catalog protection policy. A maximum of 500 snapshots are allowed. The limitation helps eliminate a flow that could result in a reaching halt_writes splittability issue alert followed by the BIG_CATALOG deny list imposed on the cluster.

• ORION-228493: Resolved an issue that could prevent snapshots from being created or deleted following a failed upgrade attempt.

• ORION-209594: Resolved an issue where following an upgrade to version 5.0, a protected path could not be created with a remote cluster that was still running version 4.6.

• ORION-213065: Resolved an issue that could cause the CNode container to restart with the assertion failed: (is_valid_communication_id(communication_id)) error while trying to set up connection to an asynchronous replication peer.

• ORION-204355: Enhanced VAST Catalog defragmentation routines to prevent a flow that could cause an assertion failed: (!is_null()) counter: 0 alert followed by the BIG_CATALOG deny list alert.

• ORION-203861: Made updates to eliminate an issue that could cause an attempt to set up a non-VAST S3 replication peer to fail with the SignatureDoesNotMatch error (HTTP status code 403).

• ORION-201912: Resolved an issue that could cause a NATIVE_REPLICATION deny list alert on multiple CNodes due to the assertion failed: (false) UDBTranslate::convert_vid_to_user_identity failed to translate vid=<...> error.

• ORION-212816: Updated handling of tenant IDs during identity policy updates to resolve an issue that could cause a The policy and the group must belong to the same tenant error on an attempt to query a user group that has been updated by assigning an identity policy in VAST Web UI.

• ORION-209184: Resolved an issue where a query by group login name could return valid results in the Active Directory or NIS context but not in the aggregated context.

• ORION-208561: Enhanced LDAP health check routines to prevent the process from getting stuck, causing errors when trying to join Active Directory.

• ORION-198358: Resolved an issue where an error during an s3_list_all_identity_based_policies operation could cause identity policies to be automatically recreated with different IDs.

• ORION-198357: Updated S3 replication routines to avoid overwriting an existing identity policy if it is identical to the replicated identity policy. This ensures that the policy ID remains the same, preventing failed to fetch s3 policies: list s3 identity policies error: ErrorCode.InternalError VMS errors that could be encountered following events such as cluster migration.

• ORION-211903: Resolved an issue that prevented VMS from sending email notifications about SSD failures.

• ORION-208184: Eliminated a false could not fetch next ip alarm that could be encountered when using VAST DNS with some specific workloads.

• ORION-207343: Improved VMS metrics cache to resolve an issue where following an upgrade, some customized analytics could not be displayed in VAST Web UI although they were available in Uplink.

• ORION-205162: Enhanced the way VMS reports link state changes to avoid false alarms.

• ORION-203155: Made updates to prevent garbage at the end of the link width alarm message Unexpected width, actual link width is <...>.

• ORION-195241: Improved the logic used to generate NIC status alarms to prevent false cable disconnection alarms.

• ORION-194297: Resolved an issue that could in some cases result in a Data Flow page not presenting any data.

• ORION-226583: In the Dbox Migration wizard, relabelled the field intended for entering the IPMI gateway IP address to IPMI gateway.

• ORION-224820: Updated the logic behind the Acknowledge option in the Alarms page (Alarms & Events -> Alarms -> right-click an alarm) so that it acknowledges the alarms listed on the page.

• ORION-211547: Updated the logic behind the (+) button in the Users Accounting tab of quota settings (Element Store -> Quotas -> choose to create or update a quota) so that clicking the button opens the User Rules tab for the user.

• ORION-209993: Updated the CNode Replication Bandwidth Limit field (Settings -> Cluster -> General) so that it always displays the value in effect.

• ORION-205441: Updated the logic behind the Acknowledge option in the Alarms page (Alarms & Events -> Alarms -> right-click an alarm) so that it acknowledges only those alarms that are currently listed on the page, skipping those that are hidden due to a filter set.

• ORION-201828: Updated the tooltips for External Netmask and IPMI Netmask fields in VAST Easy Install and Configure Network dialogs to show a valid subnet mask.

• ORION-195778: Updated API documentation for the /capacity endpoint to display the Description and some other fields as appropriate.

• ORION-194544: Resolved an issue where the cluster-wide SCM Read IOPS metric, when retrieved via the API, could occasionally be reset to a lower non-zero value.

• ORION-193392: Resolved an issue due to which the /api and /docs endpoints could be intermittently unreachable.

• ORION-191918: Updated the logic behind the /iodata/ endpoint to properly return values for host, user, vip, vippool, cnode and view fields. Prior to this change, object IDs could be returned instead of actual values.

• ORION-225866: Improved handling of substripes during defragmentation to eliminate a flow that could cause defragmentation to get stuck, resulting in a Cluster remaining_stripes_state changed from ABUNDANT to SCARCE alert on the cluster.

• ORION-225524: Resolved an issue that could cause a CNode container to restart due to an out-of-stack error condition.

• ORION-215243: Optimized the mechanism of retrieving information about detached devices during slot migration to eliminate a flow that could end up with service disruption.

• ORION-214537: Resolved an issue that could cause multiple allocated 90% of mooktze buffers! top consumer is DELETE_WORKER_PENDING_DELETE_QUEUE alerts on the cluster.

• ORION-212557: Resolved an issue that could cause some SSD drives to fail with the XRQ NVMF backend ctrl timeout error.

• ORION-211063: Resolved an issue that could cause multiple CNode containers to restart with the assertion failed: (pre_compression_encode == current_chunk->_dr_header->pre_compression_encode)  error followed by a DATA_REWRITE deny list alert on an attempt to run data rewrite after having enabled similarity-based data reduction.

• ORION-210333: Eliminated a flow that could cause a signal=11 <...> [Address not mapped to object] alert on the cluster.

• ORION-205440: Made updates to correctly configure CPU isolation (isolcpus) for an IceLake CBox with 72-core CPUs.

• ORION-205410: Improved handling of IPMI command timeouts to avoid a flow that could result in a DNode container restart due to the BMC not responsive error.

• ORION-205041: Made improvements to prevent increased latencies for specific workload types.

• ORION-204265: Resolved an issue that could cause a CNode container restart with the assertion failed: (acl_buf == nullptr) file=src/proto/common/allocated_acl.hpp:28 function=void *Proto::EncodedAceBuffer::alloc(uint32_t) silo=<...> error.

• ORION-203542: Resolved an issue that could cause multiple CNode containers to restart with the assertion failed: (ranges->count() == 1) <...> function=EStore::ColumnPredicate::ColumnPredicate(<...>) error.

• ORION-200557: Resolved an issue that could cause multiple CNode containers to restart with the silo is stuck error when deleting a very large VAST Database table.

• ORION-200149: Added a mechanism to prevent scenarios where a hardware issue could result in a mix of two internal metadata blocks during reactivation of a DTray, causing multiple deny lists and CNode container restarts.

• ORION-199628: Updated BMC firmware validations to avoid raising the DNode BMC Version is not up to date alert when firmware version 11.17 is used on all DNodes.

• ORION-198928: Resolved an issue where during recovery from drive failures, the CNode container restarted with the assertion failed: (!assert_empty) list is empty error, followed by a ESTORE MIGRATE deny list alert.

• ORION-194336: Resolved an issue that could cause a CNode container restart with the killing pid=<...> at time=<...>: silo=<...> is stuck for <...> ms error.

• ORION-190564: Made updates to automatically rotate logs in the /var/log/wtmp directory so that the directory size does not exceed 2GB.

• ORION-158539: Updated the Hardware Layout for the CERES DBox so that the the back view shows each data port in the left or right position as appropriate.