Resolved Issues in 5.3.3

ORION-273493: Resolved an issue that could cause a set_vsettings returned an error: SetVSettingsResultCode.INVALID_KEY error during EBox cluster upgrade.

ORION-269412: Resolved an issue where an attempt to deploy a cluster could not succeed due to the assertion failed: ((res) == (EStoreRes::OK)) error.

ORION-269320: Resolved an issue where during a VAST Cluster Install procedure for an EBox, the VMS raised an Not all EBox IP addresses were filled error although a suitable number of IPs was supplied in the EBox External IP Pool field.

ORION-259918: Added a validation to ensure that the VAST Cluster version currently installed on the cluster allows upgrading to the target version. If the current version does not allow the upgrade, the target version bundle cannot be uploaded.

ORION-251519: Resolved an issue where an upgrade of an IceLake cluster could not be completed due to the failed to allocate sub stripe  error.

ORION-251508: Made updates so that firmware versions are displayed correctly in the Upgrade Cluster dialog displayed for Cisco EBoxes.

ORION-238083: Updated the logic behind the VMS option to power off an EBox to prevent situations where the power off task is reported as complete while the box is inactive but not powered off.

ORION-257789: Made updates to prevent the VMS from re-applying cluster networking configuration changes on active nodes in case the normal network reconfiguration flow (which includes node deactivation, applying the changes, and re-activation) has been interrupted by an unrelated event.

ORION-277767: Improved the logic used to handle IPv6 addresses specified in host-based access rules of a view policy to resolve an issue where addresses with a subnet mask of /96 did not work as expected.

ORION-271186: Resolved an issue that could cause CNode containers to restart due to the ((obj_attr.element_flags &amp; ElementFlags::IS_BUCKET) == 0) nested buckets are not supported, found bucket_name=<...> in path=<...> error following a sequence of user actions which included multiple replicated bucket delete and recreate operations.

ORION-269979: Resolved an issue that could cause multiple CNode containers to restart due to the many block allocation attempts that resulted in non-recalimable blocks, probably reached full truman capacity error.

ORION-269346: Resolved an issue where following a cluster upgrade, an ESTORE RESET_TO_SNAP denylist alert was raised on the cluster due to the ((bytes_consumed_by_deserialize) == (serialized_size)) xeystore value deserialize didn't consume expected number of bytes. error.

ORION-268120: Resolved an issue where an unexpected client disconnection during the TCP connection establishment phase could cause the CNode container to restart with the assertion failed: ((*__errno_location ()) == 11 || (*__errno_location ()) == 11) errno: Software caused connection abort error.

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-268821: Resolved an issue where the cluster did not reject a cut operation on a folder that contained WORM-locked files.

ORION-268555: Resolved an issue that could cause multiple CNode containers to restart after encountering a release_nfs4_connection_state_resources timeout.

ORION-268935: Improved handling of historical user SIDs with SMB native authentication to avoid a scenario where a user could be denied access due to the When using smb native, while building the user token, failed to map group SID=<...> to vaid error.

ORION-270903: Made updates to ensure correct handling of S3 ListObjectsV2 requests where the bucket URL contains ?& as a delimiter between the bucket name and query parameters.

ORION-270585: Resolved an issue where copying objects to an S3 bucket would intermittently fail due to the The specified key does not exist error.

ORION-272263: Eliminated a flow that could prevent timely cleanup of .vast metadata directories for S3 buckets with versioned objects based on the lifecycle policies.

ORION-267111: Added logic to convert two or more consecutive spaces into a single space when creating the CanonicalHeaders string for signature calculation. Prior to this change, the signature was calculated taking into account the multiple spaces, which resulted in SignatureDoesNotMatch errors for the clients.

ORION-265491: Eliminated a flow where an InvalidAccessKeyId could be returned to the client running IO with a valid S3 access key on a multi-tenant cluster.

ORION-254299: Resolved an issue that could cause an ESTORE BIG_CATALOG deny list alert due to a update_rows(&select_iter, batch, ancestry_i_ehandle_col_id) failed error on a cluster that had VAST Catalog enabled and replication running.

ORION-277736: Resolved an issue that could cause a Standalone protected path to be reported as Read instead of Writable on one of the peers.

ORION-275709: Made further enhancements in the way VMS tracks and handles various synchronous and asynchronous replication states to ensure that each state transition is reported accurately and relevant alerts are raised where necessary.

ORION-275477: Resolved an issue where presence of a quota in a OK (calculating) state on the replication destination peer prevented transition of the replication stream to the SYNC state.

ORION-263090: Made updates to help prevent increased latency when running synchronous replication.

ORION-258411: Made updates to eliminate a flow that could cause CNode containers to restart due to inability to establish a secure connection when running replication with mTLS encryption.

ORION-273989: Resolved an issue that could cause repeated LDAP degraded alerts because of a timeout in one of internal DNS resolution routines.

ORION-272868: Fine-tuned internal timings to prevent a flow where an attempt to join Active Directory could fail due to a LDAP query timeout.

ORION-277445: Eliminated duplicate vast_cnode_metrics_RpcRdmaMetrics_active_links entries in VAST Prometheus Exporter response.

ORION-276833: Made updates to prevent exporting duplicate NIC metrics when using the /api/prometheusmetrics/all endpoint of VAST Prometheus Exporter.

ORION-276375: Resolved an issue that could cause view analytics/metrics to occasionally display unrealistically high IOPS and/or bandwidth values.

ORION-274905: Resolved an issue where VMS reported a checksum valid string instead of a valid serial number for one of newly added Cumulus switches.

ORION-265042: Resolved an issue that could cause an Ensure this value is less than or equal to 9223372036854775807 error when trying to use a very large number to manually set a Global Access Cache limit on files and directories (Settings -> Cluster -> General Cluster Setup and Actions -> Global Access Cache Limits -> Limit files and directories -> Set manually).

ORION-263957: Resolved an issue that could make VMS recreate the existing identity policies due to an internal polling timeout.

ORION-254263: Resolved an issue that could cause the VMS to hang with Internal Server Error (500) when listing the views or editing a view.

ORION-244308: Made updates to resolve an issue where occasionally, subdirectories could be displayed in the VAST Web UI Analytics -> Capacity page only after refreshing the page.

ORION-273205: Enhanced the replication configuration dialog to ensure that all labels for user controls are displayed in full.

ORION-268219: Made updates so that toggling the Host button in the Analytics -> Data Flow page does not cause a reset of data flow isolation settings.

ORION-278052: Made updates to enforce denylisting if during DBox expansion, multiple CNode containers keep restarting with the Not enough space for all write buffer error.

ORION-273549: Resolved an issue that could cause CNode containers to restart with the (version-> has_val()) fiber-cluster-version was not set for this fiber error.

ORION-257849: Resolved an issue where a health check error laddr_@a=<...> mirrored_@a=<...> alert was encountered on a CNode.

ORION-257017: Resolved an issue that could sometimes prevent completion of the driver replacement procedure on a Cisco EBox due to a modify-ssd task timeout during the device activation phase.

ORION-249456: Fine-tuned the logic for collecting BMC logs to prevent the process from getting stuck.

ORION-217946: Resolved an issue that could cause the cluster to falsely report a fan speed of 0 RPM on MLK DNodes running certain firmware versions.

ORION-276360: Enhanced handling of support bundle-related failures to eliminate a flow that could cause spamming of worker logs and high VMS memory usage alerts.

ORION-241318: Resolved an issue where some of support bundles could not be collected due to the key=<key name> failed retrieved from DB error.

ORION-199658: Resolved an issue where an attempt to create a support bundle with a preset to debug via VAST CLI failed with the Exception during node harvest: cannot use a string pattern on a bytes-like object error.