New Features¶

• LightOS connector: Introduce new configuration options to customizethe LightOS discovery client’s behavior.

  Operators can now override the default values for:

  • discovery_client_address

  • discovery_client_port

  • discovery_client_dir_path

  These settings allow fine-tuning of the discovery client’s networkendpoint and configuration directory.

Upgrade Notes¶

• LightOS connector: Introduce new configuration options to customizethe LightOS discovery client’s behavior.

  Operators can now override the default values for:

  • discovery_client_address

  • discovery_client_port

  • discovery_client_dir_path

  No changes required. Default values are retained for backward compatibility;no action is needed during upgrade.

• With the fix forBug #2138280, starting fromos-brick version 6.15.0 or later, users must retain the/opt/emc/scaleio/openstack/connector.conf file for volumes that wereattached using the legacy flow prior to the 6.13.0 release, until allsuch legacy volumes are detached. For volumes attached using the newattachment flow, the/opt/emc/scaleio/openstack/connector.conf file isno longer required.

Bug Fixes¶

• Bug #2003179: Fixedissue with the ScaleIO connector where privsep was loggingpassword in plaintext. This was fixed by creating a new logger for theScaleIO connector with default log level set to ERROR.

• NFS driverBug #2122594:Previously we mounted NFS shares always with version 4.1 if supportedby the client/server. Now we attempt to mount with NFSv4.2 first.If that fails due server/client incompatibility we will fall backto older versions(4.1, 4.0, 3.x).

• PowerFlex DriverBug #2138280: Fixed an issue wheredetach operations failed for volumes attached before the 6.13.0release that did not include ansdc_guid in the connector information.From os-brick version 6.15.0 or later, detach operations for these legacyvolumes are handled by os-brick to ensure backward compatibility.

Bug Fixes¶

• Bug #2136842: Fixedissue with server hard reboot instance due to delayed uevent processing.When performing server hard reboot, the uevents for removing multipathdpaths could be delayed and cause device removal at a later stage causingthe device to be inaccessible. Now we wait for multipathd to remove thepaths before marking the disconnect_volume operation complete.