Resolved Issues in 5.2.2

ORION-258402: Made updates to ensure that during NDUs, the cluster always preserves the creation time property value for the existing S3 keys.

ORION-249921: Resolved an issue that could occur on some clusters with lifecycle rules configured, where the leading CNode could not start due to the failed to upgrade schema table for export <export> error, causing the upgrade task to hang.

ORION-248621: Resolved an issue where the an EBox deployment task failed due to the cannot access '/vast/deploy': No such file or directory error although the directory existed on all the nodes.

ORION-221554: Resolved an issue where an attempt to update a tenant during cluster upgrade caused CNodes containers to restart.

ORION-259677: Resolved an issue where the value specified in the optional Hostname Prefix field in the DBox and CBox Expansion dialog (Infrastructure -> CBoxes or DBoxes -> choose to create a new box) was not used during expansion, resulting in the nodes being configured with the localhost.localdomain hostname instead.

ORION-251666: Resolved an issue where an attempt to add the DBox through VAST Web UI could not succeed due to the External IP address is not valid error.

ORION-255602: Improved the DBox replacement procedure to prevent drive location assignment errors when replacing a Sanmina DBox with a Mavericks MLK DBox.

ORION-241559: Added support for the MAVERICKS_15_TO_SIXTY_TB_DELTA_LAYOUT layout during the DBox migration procedure.

ORION-250417: For clusters with InfiniBand internal networking, optimized flows where VMS needs to acquire information from the OpenSM service in order to reduce the load on the OpenSM service.

ORION-248412: Updated the logic that sets SSH listening ports to prevent a flow where it could ignore the B2B template deployment parameter for clusters with B2B IPMI networking, resulting in incorrect technician port configuration on the cluster nodes.

ORION-241045: Made updates to prevent the VAST cluster from generating interface names that are longer than 15 characters, which could previously cause multiple CNode containers to restart with the (len < (int)sizeof(label_with_device) && len > 0 && len >= MAX_LABEL_LENGTH) error.

ORION-208862: For clusters with internal IB networking, enhanced OpenSM validations with a mechanism to periodically update the ibdiag.conf file on the CNodes based on the current interface availability.

ORION-251518: Resolved an issue that could cause CNode container restarts with the assertion failed: (!this->_is_set) error followed by an ESTORE TREE_UNLINKER deny list alert.

ORION-113279: Resolved an issue that could cause an ESTORE MIGRATE denylist on the cluster due to the waiting for a lock for too long (195000ms). aka- passed max compare and swap retries  and assertion failed: (!name_a->_snap_line.equal(name_b->_snap_line)) Two entries for same hash and snap after transactions cleanup?! errors.

ORION-213073: Updated handling of internal quota IDs to eliminate some flows where quotas could not be displayed or managed (for example, deleted) because the VMS was internally using an incorrect quota ID.

ORION-208221: Updated quota notification emails to indicate blocked users and blocked groups separately. Prior to this change, both users and groups were listed in the Blocked users field.

ORION-246038: Resolved an issue that could occur for views with two or more lifecycle rules defined and cause a delete_lifecycle_rule returned an error: ErrorCode.NoSuchBucket error when trying to delete a second lifecycle rule after the first one has been deleted.

ORION-245323: Enhanced routines behind lifecycle rules to eliminate a flow that could cause a CNode container to restart with the ((consecutive_failures) < (fail_handling->failures_till_panic)) (2400 < 2400) failed op=TREE_UNLINKER error when a lifecycle rule is set to force deletion of a VAST database (which is prohibited).

ORION-253118: Added validations to prevent configuring minimum QoS in newly created QoS policies.

ORION-250829: Made updates to allow for deletion of the QoS policy in case the attempt to create a view with this QoS policy fails.

ORION-261047: Added a caching mechanism to prevent hotspots when checking permissions for an S3 workload on a view controlled with the NFS security flavor.

ORION-245291: Resolved an issue that could cause an allocated 90% of mooktze buffers! top consumer is NAME_TRAVERSAL_CONTEXT_ACTUAL_NAMES_POOL alert when trying to list contents of an S3 bucket with an extremely large number of S3 objects (ending with a slash) from NFS.

ORION-252534: Changed the way the VAST cluster handles NFSv3 create requests that include the atime for a file on a WORM-enabled view. Prior to this change, such requests could be intermittently rejected with a permission denied error by the WORM retention logic. Starting with this release, for this type of requests, the VAST cluster ignores the incoming atime if it is less than or equal to the current time.

ORION-224594: Resolved an issue that could cause false GSS Context allocations reached 96% alarms when the cluster was busy with NFSv4 workloads.

ORION-255064: Enhanced handling of SMB lease break operations to avoid a flow that could cause a CNode container restart.

ORION-245517: Enhanced handling of bursts of NTLM authentication operations to prevent latencies and avoid impact on other tasks running on the cluster.

ORION-241304: Updated SMB lease break event notifications produced by the VMS to remove unnecessary Existing phandle <...> in lease break map differ from requested phandle <...> for lease <...> alerts.

ORION-216381: Made updates to support CIFS mounts without the noserverino flag.

ORION-243166: Resolved an issue that could cause an S3 request to fail with the SignatureDoesNotMatch error in some cases when there are multiple headers with the same prefix and one of the headers is signed.

ORION-242283: Made updates so that an invalid ACL is counted as a client error in the bucket metrics.

ORION-235815: Improved handling of S3 requests signed with Signature v4 to avoid returning status code 403 due to signature mismatch in case the request contained multiple Content-Encoding headers.

ORION-235044: Resolved an issue that could cause caching inconsistency when processing S3 requests.

ORION-254660: Made updates to ensure that the audit log records for SMB requests against a non-default tenant configured to use SMB native authentication, contain non-empty LoginName, UID and SID field values.

ORION-253107: Added a validation to prevent configurations that result in creating a snapshot on a path used by a different snapshot.

ORION-248109: Resolved an issue where the cluster was raising RPO miss alerts although the replication was running smoothly and in accordance with the protection policy.

ORION-241340: Added an ability to edit the destination peer in the duplicate protection policy being added using the Duplicate and Edit action in the Data Protection -> Protection Policies page in VAST Web UI.

ORION-218211: Resolved an issue where enabling global snapshot clones caused CNode containers to restart with the Fiber=1c00eb group=W_GN_MD_PREFETCH_WORKER had a suspension timeout error.

ORION-211566: Resolved an issue that could cause an error when trying to create a local protected path or a global snapshot clone on an S3 bucket view for which Write Once Read Many (WORM) was enabled.

ORION-178696: Removed a validation that could prevent configuring a path as a local protected path and a replication protected path at the same time.

ORION-250171: Enhanced communications between replication peers to resolve an issue where a protected path got stuck in the Initial Sync status for a long period of time.

ORION-250164: Eliminated an asynchronous replication flow where some destination files could be created with a VAST ID (vid) of 0 (zero), resulting in an ESTORE NATIVE REPLICATION deny list alert due to the UDB2 lookup group on vaid=<...> failed to return a token counter error.

ORION-241730: Made updates to avoid reasing a Replication Group <Name> missed it's sync point for 0.00 seconds alert after removing a member from a replication group that used to have three or less members.

ORION-241633: Made updates to prevent some flows where an excessive consumption of metadata space could be encountered on the replication destination cluster.

ORION-229345: Resolved an issue where an attempt to set up replication with mTLS encryption resulted in a certificate validation error due to one of the keys being password-protected.

ORION-220888: Resolved an issue where multiple CNode container restarted with the ((attr->creation_snap.snap_time) <= (_snap_range.end()))  error, causing an automatic deny list alert related to handling of a replication stream.

ORION-220478: Resolved an issue that could cause multiple missed RPO alerts for a VAST Catalog replication stream on clusters with VAST Catalog enabled.

ORION-217663: Resolved an issue that could cause an ESTORE BIG_CATALOG denylist alert due to a can't find base attributes error on a cluster with VAST Catalog enabled.

ORION-252425: Updated handling of semi-sorted projections to resolved an issue which could cause ESTORE TABULAR_PROJ_SORTER deny list alerts due to the (!_traversal_ctx->performing_split(nullptr)) shouldn't be performing split operations during sub tree creation error.

ORION-259236: Made updates to ensure that user impersonation configured for an SMB share controlled with the NFS or Mixed Last Wins security flavor, is also effective for the subfolders within the SMB share top folder.

ORION-248372: Improved handling of the memberOf attributes with ranges in the auth provider responses (memberOf;range=<...>) to resolve an issue where a user was denied access due to a PARSE_FAILED error when parsing the provider query response.

ORION-247119: Resolved an issue that could occur in an environment with Active Directory connected using the GC port, where an attempt to set users or groups as bucket owners in a view policy via VAST Web UI failed with the UDB returned RETRY <...>. Try again alter error, although it was possible to add those via VAST CLI.

ORION-227177: Enhanced LDAP netgroup caching to reduce impact on cluster's read latency.

ORION-223630: Added support for longer Unicode LDAP DN strings (up to 1020 bytes).

ORION-249956: Added an ability to export metrics that report client and server error responses using VAST Prometheus Exporter.

ORION-249747: Added validations to ensure that a VMS manager user belonging to a certain tenant, is not allowed to manipulate views, quotas, view policies, virtual IP pools or QoS policies that are not associated with the tenant.

ORION-238277: Resolved an issue that could in some cases cause discrepancies in the numbers VMS reports for bandwidth consumption in Data Flow diagrams and in the cluster dashboard.

ORION-235779: Added VMS event definitions to provide the ability to monitor the fan status in HPE IceLake CBoxes.

ORION-197137: Starting with VAST Cluster 5.2.2, VAST Prometheus Exporter metrics contain a UID of 0 (zero) for the root user. Prior to this change, an empty root UID field was returned.

ORION-249163: Updated the logic behind the SMB continuous availability field in view policy settings (Element Store -> View Policies -> choose to edit a view policy) so that it does not automatically change from Enabled to Disabled when the policy is reopened for editing.

ORION-246617: The field for setting the unit of measure for the Minimum object size and Maximum object size filters of a lifecycle rule (Element Store -> Lifecycle Rules -> choose to view or edit a rule) is now compatible with the VMS setting that stipulates use of base2 units.

ORION-245782: Updated the view settings dialog (Element Store -> Views -> choose to create or edit a view) so that the WORM tab is displayed when the S3 protocol is disabled for the cluster.

ORION-250634: Improved handling of DNode failure scenarios to prevent a flow that could cause multiple CNode containers to restart due to the ((timer.elapsed_seconds()) < ((uint64_t)P::ACTIVE_PASSIVE_HA_DELAY)) (170 < 170) trying to read a block in address=<...> for too long error.

ORION-249421: Made updates to prevent multiple CNode container restarts due to simultaneous reconfiguration of NTP on the cluster nodes.

ORION-247038: Enhanced DNode failure handling to sustain a rare combination of events which otherwise could result in temporary service disruption after deactivation of a CNode.

ORION-246303: Resolved an issue that could cause a CNode container to restart with the spinlock invalid error when processing SMB workload.

ORION-242594: Made updates to ensure that replacement of a very large number of SSDs at the same time does not result in CNode containers being restarted due to the assertion failed: <...> after deactivation of drive=<...> dbox=<...> will be at risk to have insufficient drives" for mioc state and assertion failed: (load_successful) this might mean that system_format failed earlier <...> or that loading mioc failed due to unavailable dboxes/drives errors.

ORION-242420: Resolved an issue where UDB2 lookup group on gid=<...> failed to return a token alerts were raised on one of the CNodes following a drive failure.

ORION-241261: Resolved an issue that could cause multiple CNode containers to restart due to the Fiber=e2035e09 group=W_DR_DEFRAG had a suspension timeout! error.

ORION-241260: Resolved an issue that could cause multiple CNode containers to restart due to the timeout waiting for deactivation safe error (occurring when the CNode attempts to deactivate a drive).

ORION-239683: Made updates to NVMe namespace management to eliminate a flow that could cause a CNode container to restart with the assertion failed: ((result) == (0)) (3 == 0) error.

ORION-238029: Added support for the Solidigm SBFPF2BV614TOP drive model in VMS.

ORION-237353: Resolved an issue that could cause a ESTORE MULTI_XEYSTORES_SCRUBBER denylist alert on the cluster due to the a block to relocate must exist error.

ORION-235410: Introduced HA enhancements so that the cluster would be able to automatically recover in scenarios involving failures of multiple switches.

ORION-231563: Resolved an issue that could cause a CNode container to restart due to an out-of-stack error condition, followed by a an ESTORE DELETE_SNAP deny list alert.

ORION-225587: Resolved an issue where intermittent hangs or IO errors were encountered, accompanied with (false) inconsistent serial_@a=<...> address_@a=<...> errors and data decompression failure alerts in the cluster logs.

ORION-192203: Increased the number of the user's previous VAST OS passwords checked during password rotation from 5 to 10. A new password can be set only if it does not match any of the 10 previous passwords.

ORION-254123: Improved handling of timeouts that might occur while getting SMB logs from the CNodes to prevent HARVEST_FAILED errors during callhome bundle collection, resulting in the Missing callhome logs bundles for over day alerts on the cluster.

ORION-244957: Resolved an issue that could cause the failed to create temp dir - [Errno 28] No space left on device error during callhome bundle collection.

ORION-230144: Made updates to optimize the size of callhome bundles.