What's new
Release highlights for HCP 8.2
Cross-origin resource sharing
HCP 8.2 introduces support for Cross-Origin Resource Sharing (CORS), an important security mechanism in web environments that allows restricted resources to be requested from a domain outside of the domain from which the original resource was served. CORS configuration can be performed using the System Management Console, the Tenant Management Console, MAPI and these S3 API methods: PUT bucket, DELETE bucket, and Get bucket. CORS run-time is supported through the S3 and REST gateways.
POST object upload API method
You can use the new POST Object Upload method included in the HCP S3 API to upload objects to HCP in web environments. This method allows browser-based uploads of objects into namespaces when the S3 API is enabled. This method of object upload uses HTML form fields to the POST Object Upload API call to pass parameters such as authentication, object key, and policy information to HCP. The API supports both AWS signature v2 and v4 for authentication.
Service performance improvements
HCP 8.2 services have been updated to make computational and storage resources work more efficiently and improve performance. The Storage Tiering service and the Disposition service now perform inline garbage collection. Objects that must be deleted are identified by these two services, offering more available disk space and improved allocation of HCP system resources. Inline garbage collection has also been added to the DELETE and Multi-object DELETE (S3 Bulk DELETE) API methods. When these interactive delete operations are performed, storage space is cleared immediately, so that the Garbage Collection service does not have to run as often as in previous releases.
Ranged read optimization
HCP 8.2 offers improved performance of ranged-read requests on objects that are compressed or encrypted (or both) and located in cloud tiers and HCP-S nodes.
"X-Forwarded-For" (XFF) and "forwarded" header support
"X-Forwarded-For" and "forwarded" headers are supported with REST gateways. When a client issues a request, these headers preserve the original IP address of the client if the request is routed by a Load balancer or a Proxy server. HCP sends the original IP address to the appropriate gateway logs, which can then be used by log parsing applications to detect the origin of the request.
Monitoring improvements
Wear-level monitoring for solid state drives includes alerts and events that are generated from threshold values configured in the drive. System-level APIs provide monitoring and reporting for user and group accounts, as well as roles.
VMware improvements
HCP 8.2 supports ESXi 6.7 U1 and ESXi 6.5 U2 VMWare environments. You can also upgrade from 6.5 U2 to 6.7 U1. This release also supports the use of VMWare distributed virtual switches.
The release also resolves several issues found in previous releases of the product. For more information, see Issues resolved in the current release.
Trademarks and Legal Disclaimer
© 2014, 2019 Hitachi Vantara Corporation. All rights reserved.
Release highlights for HCP 8.1.2
Hitachi Content Platform release 8.1.2 adds HCP G10 node support for an all-SSD primary storage device configuration. This configuration may yield performance gains for many HCP workloads including GETs, PUTs, Services, and indexing.
The release also resolves several issues found in previous releases of the product. For more information, see Issues resolved in release 8.1.2.
Trademarks and Legal Disclaimer
© 2014, 2019 Hitachi Vantara Corporation. All rights reserved.
Release highlights for HCP 8.1
Release 8.1 of HCP fixes issues found in release 8.0 and earlier releases.
Patch-level software updates
As of release 8.1, HCP will make fixes to critical bugs found by customers and overall system quality improvements widely available in the form of software patches. An HCP system can be upgraded to any patch-level software release from release 7.0 or later.
Encrypted ranged-read optimizations
Release 8.1 of HCP improves the performance of ranged-read requests on objects tiered to encrypted storage pools with Amazon S3, Google Cloud, Microsoft Azure, the S3 compatible API, and NFS.
Replication encryption updates
HCP release 8.1 improves replication security by enabling encryption on replication links by default.
Hitachi API for Amazon S3 bulk delete
New with release 8.1 of HCP, you can use the HTTP POST method to delete multiple objects in a bucket with the S3 compatible API.
For information about how to delete multiple objects with the S3 compatible API, see Using the Hitachi API for Amazon S3.
S3 compatible retention headers
With release 8.1 of HCP, you can set object retention values and perform privileged deletes on objects with the S3 compatible API.
For information about how to use S3 compatible retention headers, see Using the Hitachi API for Amazon S3.
S3 compatible GET bucket v2
Release 8.1 of HCP supports GET bucket v2 operations using the S3 compatible API. With this added support, you can use the HTTP GET method to list the contents of a bucket with v2 of the S3 compatible API.
For information about how to use GET bucket v2 operations with the S3 compatible API, see Using the Hitachi API for Amazon S3.
VMware ESXi 6.5 U1 support
As of release 8.1 of HCP, HCP software can be deployed on virtual machines in VMware ESXi 6.5 U1 environments.
For information about how to deploy and configure HCP virtual machines in VMware ESXi 6.5 U1 environments, see Deploying an HCP-VM System on ESXi.
HCP-VM management port network
With release 8.1 of HCP, a separate management port network can be configured on HCP-VM systems, which can be used to isolate management access from client access.
For more information about the HCP-VM management port network, see Deploying an HCP-VM System on ESXi.
For information about how to configure the management port network using the System Management Console, see Administering HCP.
Signal monitoring enhancements
HCP release 8.1 expands the signaling capabilities of HCP, which allows HCP nodes and services to be monitored and examined more effectively. This feature also allows you to use the HCP management API to retrieve statistics regarding the nodes in your HCP system and statistics regarding the services used by your HCP system.
For information about how to use the management API to retrieve node statistics and service statistics, see HCP Management API Reference.
BMC/BIOS firmware updates
Release 8.1 of HCP updates the BMC Firmware Revision to 3.66.00 and updates the BIOS Version to S2B_3B10.07.
VMware vSAN support
Release 8.1 of HCP provides support for VMware vSAN 6.5 in VMware environments.
Resolution of cross-site scripting vulnerabilities
HCP release 8.1 includes security updates regarding persisted cross-site scripting, reflected cross-site scripting, and cross-site request forgery.
Documentation for configuring HCP as a CIFS Vault Store Partition within EV
Release 8.1 of HCP provides new documentation on how to create an HCP Vault Store Partition within Veritas Enterprise Vault using the CIFS namespace access protocol.
For more information about this and other configurations using CIFS, see HCP Streamer Adapter for Veritas Enterprise Vault v1.4.4 User Guide.
Dedicated database volume for HCP-VM
New with HCP release 8.1, you can optionally configure a dedicated database volume for an HCP-VM system if each virtual machine is configured with three or more data disks, at least one of which is greater than 50 GB. By configuring a dedicated database volume for an HCP-VM system, you can separate the storage of user data and metadata from the HCP database.
For information about how to configure a dedicated database volume for an HCP-VM system, see Deploying an HCP-VM System on ESXi.
Windows 10 CIFS validation
As of release 8.1 of HCP, Windows 10 can be used for CIFS ingest.
SSD expansion support
With HCP release 8.1, you can expand SSD drive capacity in an HCP system by using the HCP Service Menu.
AD cipher support expansion
With release 8.1 of HCP, you can integrate HCP into Active Directory environments that permit the use of only AES-128 and AES-256 Kerberos authentication with AD.
SMBv1 disable option
As of HCP release 8.1, you can use the HCP System Management Console to prevent users from using the SMBv1 protocol for data access with CIFS.
For more information about this and other network security options, see Administering HCP.
Health check reporting
New with release 8.1 of HCP, you can perform health checks on the nodes in your HCP system and download the health check reports in JSON format.
For information about how to prepare and download health check reports using the HCP System Management Console, see Administering HCP.
For information about how to prepare and download health check reports using the HCP management API, see HCP Management API Reference.
Increased log download speed
Release 8.1 of HCP increases the speed of log collection and log downloading on nodes in the HCP system.
High load/swap event improvement
As of release 8.1 of HCP, you can specify how much time the HCP system should allow high load events and disk swap events to persist before raising the high load alarm and the disk swap alarm, respectively.
Replication link timeout
With release 8.1 of HCP, you can specify how long the HCP system should wait before reporting a replication link connectivity failure.
For information about how to change link connectivity failure reporting using the HCP System Management Console, see Replicating Tenants and Namespaces.
For information about how to change link connectivity failure reporting using the HCP management API, see HCP Management API Reference.
Replication before tiering
In HCP release 8.1, you can specify whether objects must be fully replicated before transitioning from one storage tier to another.
Event-based retention
In HCP release 8.1, you can specify a Minimum Retention after Initial Unspecified value. When this value is set, retention updates on Initial Unspecified objects in a namespace must be greater than the Minimum Retention after Initial Unspecified value.
VMware repackaging
Release 8.1 of HCP retires the former OVF deployment method for HCP-VM systems. HCP virtual machines are now deployed by using the HS222 ISO file.
For information about how to deploy HCP virtual machines using the HS222 ISO file, see Deploying an HCP-VM System on ESXi.
New fields in the HCP MIB
The HCP 8.1 MIB contains the new fields shown in the table below.
Field | Description |
---|---|
Storage pool information |
|
storagePoolCompression | The Compress Data at Rest setting for the storage pool |
storagePoolEncryption |
The Encrypt Data at Rest setting for the storage pool |
storagePoolContentVerification |
The Content Verification setting for the storage pool |
Trademarks and Legal Disclaimer
© 2014, 2019 Hitachi Vantara Corporation. All rights reserved.
Release highlights for HCP 8.0
Release 8.0 of HCP fixes issues found in release 7.3.3 and earlier releases.
Erasure coding for geographically distributed data protection
Release 8.0 of HCP introduces erasure coding as an alternative method for geographically distributed data protection. Until now, the only supported method for geo-protection has been the replication of whole objects to one or more other HCP systems in a replication topology. With erasure-coded protection, the data for each object in a replicated namespace is encoded and broken into multiple chunks. An additional chunk contains parity for the data chunks. The data and parity chunks are distributed across the systems in an erasure coding topology such that, for any given object, each system stores one chunk.
An erasure-coded object can be read from any system in the erasure coding topology. If one system becomes unavailable, the distribution of data and parity chunks ensures that the object can still be read from any available system.
You can assign different protection methods to different namespaces. The main tradeoff is between the increased storage efficiency that comes with erasure-coded protection and the ability of whole-object protection to provide protection against concurrent system failures.
Multipart upload with the HS3 API
New for the HS3 API in HCP release 8.0, you can use multipart uploads to store large objects — as large as five TB. Multipart upload is the process of creating an object by breaking the object data into parts and uploading the parts to HCP individually. The result of a multipart upload is a single object that behaves the same as objects for which the data was stored by means of a single PUT object request.
To perform a multipart upload, you first initiate the upload, then you upload the parts, and, finally, you complete the upload. By uploading multiple parts concurrently, you can shorten the amount of time required to store a large object. The part order is determined by the number you assign to each part in the upload requests, not by the order in which the parts are uploaded.
With a multipart upload, you don’t need to know the full size of the object data before you start uploading the data to HCP. Thus, multipart uploads support storing streaming data in real time.
By uploading object data in parts, you avoid the need to repeat a large upload operation when the connection is lost while the upload is in progress. Because each part you upload can be as small as one MB, and the last part can be even smaller, the time required to repeat the upload of a part can be short.
HCP management network
With release 8.0 of HCP, HCP systems can be configured to support a management network. The [hcp_management] segregates system administration, tenant administration, management API, SNMP, syslog, outgoing SMTP, and SSH traffic from the [hcp_system] network.
HCP licensing updates
With release 8.0 of HCP, there are three types of available licenses, basic, premium, and extended. Basic and premium licenses allow for object storage on HCP or on S Seriess. Extended licenses allow you to tier to an extended storage tier.
A license specifies the amount of HCP storage space and the HCP features you can use. For a list of the HCP features available for each license, see HCP System Management Help.
HTTP DELETE version ID and timestamp
With release 8.0 of HCP, you can issue an HTTP object DELETE command using version ID or timestamp. You can delete a single object or a range of objects.
HCP-VM running on KVM host
With release 8.0 of HCP, HCP software can be deployed on virtual machines resting on a KVM host.
Active Directory join without Samba
With release 8.0 of HCP, HCP does not require authenticated CIFS support to join AD. As a result, HCP join time to AD has been improved.
Active Directory authentication tokens with AWS Java SDK
With release 8.0 of HCP, AD authentication tokens can be used with the AWS Java SDK.
Active Directory credentials with HSwift temporary authentication
With release 8.0 of HCP, AD credentials can be used for HSwift temporary authentication.
Generate User Account authorization token
With release 8.0 of HCP, system-level users can generate user account authorization tokens in the System Management Console.
Sending management API log messages to the syslog servers
As of release 8.0 of HCP, system-level users can choose to send log messages about management API requests to specified syslog servers.
New and changed fields in the HCP MIB
The HCP 8.0 MIB for use with SNMP contains the new fields shown in the table below.
Field | Description |
---|---|
Storage licensing |
|
storageUsedCapacity | The total used storage on HCP primary active storage and HCP S Series Nodes |
storageLicensedCapacity |
The total storage capacity on HCP primary active storage and HCP S Series Nodes |
storageLicensePremiumEnabled |
An indication of whether the HCP system has a premium storage license |
Node information |
|
nodeMgmtIP |
The node management network IPv4 address |
nodeMgmtIPv6 |
The node management network IPv6 address |
nodeMgmtIPv6Sec |
The secondary node management network IPv6 address |
nodeMemoryTotal | The total amount of memory on the node, in bytes |
Patch version |
|
hcpPatchVersion | The HCP system patch version |
Management API access logs |
|
syslogSendMapiAccessLogs |
An indication of whether management API access logs are sent to syslog servers |
mapiLogFacility |
The name of the management API log facility that receives management API access logs |
HTTP access logs |
|
httpLogFacility |
The name of the HTTP access log facility that receives HTTP access logs |
Multipart object information |
|
multipartObjectCount | The number of multipart objects |
multipartObjectPartCount | The total number of parts of multipart objects |
multipartObjectSize | The size, in bytes, of multipart objects |
multipartUploadCount | The number of in-progress multipart uploads |
multipartUploadPartCount | The number of parts of in-progress multipart uploads |
multipartUploadSize | The size, in bytes, of parts of in-progress multipart uploads |
HCP information |
|
erasureCodedObjectCount |
The number of erasure-coded objects |
Tenant information |
|
tenantErasureCoding |
An indication of whether the tenant can erasure code its namespaces |
HCP services |
|
ecTopologyTable | The table of erasure coding topologies |
ecTopologyTableEntry | The information about each erasure coding topology table entry |
ecTopologyTableIndex | The erasure coding topologies table index |
ecTopologyName | The erasure coding topology name |
ecTopologyState | The erasure coding topology state |
ecTopologyProtectionStatus | The erasure coding topology protection status |
ecTopologyReadStatus | The erasure coding topology read status |
ecTopologyDataChunks | The number of data chunks in the erasure coding topology on this HCP system |
ecTopologyParityChunks | The number of parity chunks in the erasure coding topology on this HCP system |
ecTopologyType | The erasure coding topology type |
ecTopologyErasureCodedObjectCount | The number of erasure-coded objects in the erasure coding topology on this HCP system |
Trademarks and Legal Disclaimer
© 2014, 2019 Hitachi Vantara Corporation. All rights reserved.