Release Highlights

This page contains an overview of the new features in each release. For a full listing of the features, bug fixes and upgrade notes, please see the release notes.

4.2.10

Bug fixes

  • Storage rule supervisor repeatedly restarting due to a long running transaction, causing copy jobs not to start.
  • High JMS connection usage from storage supervisor.
  • Slow file system metadata updates causing growing imports to end early (Tunable growing file timeout.)
  • Slow library search due to rollbacks.
  • Cached search bigtext entries not always removed.
  • Solr searchers keeping GlassFish from shutting down.
  • Quote characters in a collection name generates broken access DOT graph.
  • Checksum not computed for file with %2B in name.

Transcoder fixes

  • Interlace flag not working for H.264.
  • Incorrect subclipping of MXF OP1a with index in footer.
  • Incorrect bit depth for AES3.
  • 32bit lpcm detected as 16bit.
  • Burnt in timecode text offset from text box.

4.2.9

This is a bug fix release with a large number of fixes. Notably:

  • Cached search documents not being removed from the big text table, causing it to grow over time.
  • The periodic update of a large library could cause library API requests and index updates to halt due to a locking issue.
  • New Relic users will now see web requests with transaction names that match the path used in the request, instead of the automatic name assigned by New Relic which tends to group requests to different endpoints together.
  • Speed ups when faceting on many fields with many matching terms.
  • When using SolrCloud the connection manager could abort long running requests, typically causing updates to Solr to fail.
  • Various bug fixes and improvements.

For the full list of changes, see the release notes.

4.2.8

This is a minor bug fix release that also adds two new job parameters.

New job parameters:

Fixes:

  • Storage not marked as offline if MatrixStore vault goes offline.
  • Poster format settings not working.
  • Installer ignoring required but failing HTTP requests.
  • Certain transcodes/render jobs crashing the transcoder.
  • Various bug fixes and improvements.

For the full list of changes, see the release notes.

4.2.7

This is a minor bug fix release. See the release notes for details.

4.2.6

Security notice

Apache POI has been updated from version 3.8 to 3.11 to mitigate XXE vulnerabilities (CVE-2014-3529 and CVE-2014-3574) when extracting metadata from Office documents. This metadata extraction is only done if enabled using parseFileMetadata.

XXE has also been disabled in the XML parser used by Vidispine, to address vulnerabilities when parsing XML API requests and sidecar files for example.

Performance improvements

This release contains a number of performance improvements.

For the API:

  • Faster collection update and delete.
  • Faster group update and delete.
  • Faster merged access retrieval.
  • Faster storage-rule creation.

For items:

  • Faster poster generation.

Changed defaults

Multi-site processing is now disabled by default. See disableSiteCrunching.

Other

  • Indexing of items with large metadata text fields could cause indexing to halt. This has now been fixed.
  • Field groups in the item metadata are now indexed less often.
  • Additional audio and video settings from the transcode preset are now supported when rendering or conforming.
  • More control over burnt-in subtitle placements. See Subtitle metadata fields and groups.
  • Various bug fixes and improvements.

4.2.5

Temporary transcoder path

It is now possible to use another directory for temporary files for the transcoder. This is controlled via the <tempPath> element in the transcoder. This element is controlled via local configuration file or transcoder resource definition or by changing the configuration for all transcoders.

Optional hit count

The hit count can now be omitted from the results when searching. This can be done to reduce the response time of search requests.

GET API/item?count=false

Other

  • Buckets with files in subfolders are now properly scanned and will now appear on your S3 storages.
  • Proxy services can be used for cloud licensing.
  • A number of issues with the multi-site sync have been fixed.
  • Various bug fixes and improvements.

4.2.4

FTP connection pooling

If you perform a large number of imports or exports over high latency FTP connections then you can now create a FTP connection pool to reduce the overhead of establishing a new connection each time.

PUT /configuration/ftp-pool
Content-Type: application/xml

<FtpPoolConfigurationDocument xmlns="http://xml.vidispine.com/schema/vidispine">
  <pool>
    <minSize>0</minSize>
    <maxSize>-1</maxSize>
    <evictionInterval>30000</evictionInterval>
    <minIdleTime>60000</minIdleTime>
  </pool>
<FtpPoolConfigurationDocument/>

Find collections with specific items

An item subquery can now be used when searching for collections to find collections based on the items that they contain.

PUT /collection
Content-Type: application/xml

<ItemSearchDocument version="2" xmlns="http://xml.vidispine.com/schema/vidispine">
  <operator operation="OR">
    <field>
      <name>title</name>
      <value>Peach</value>
    </field>
    <item>
      <field>
        <name>title</name>
        <value>Peach</value>
      </field>
    </item>
  </operator>
</ItemSearchDocument>

See Searching for collections with specific items.

Transcoder transfer and hash computation

Available transcoder with direct file access to media can be used to compute hash sums of files, or to do filesystem-to-filesystem transfers. This will offload the application server.

To enable this, use the configuration variables enableTranscoderHashing and enableTranscoderTransfer.

Database purging

Automatic routines for trimming two of the largest Vidispine tables are now included. For information about how to enable this, see Database purging.

4.2.3

StatsD metrics

Metrics related to operations performed by both Vidispine and the transcoders can now be exposed using the StatsD protocol. It is also possible to access the Vidispine metric statistics using JMX.

$ node stats.js config.js
17 Dec 12:23:31 - reading config file: config.js
17 Dec 12:23:31 - server is up
17 Dec 12:23:31 - DEBUG: Loading backend: ./backends/graphite
17 Dec 12:23:34 - DEBUG: vs.job.total.aborted:99|g
17 Dec 12:23:34 - DEBUG: vs.job.total.aborted_pending:0|g
17 Dec 12:23:34 - DEBUG: vs.job.total.failed_total:520|g
17 Dec 12:23:34 - DEBUG: vs.job.total.finished:50491|g
17 Dec 12:23:34 - DEBUG: vs.job.total.finished_warning:3|g
17 Dec 12:23:34 - DEBUG: vs.job.total.ready:0|g
17 Dec 12:23:34 - DEBUG: vs.job.total.started:1|g
17 Dec 12:23:34 - DEBUG: vs.job.total.waiting:0|g
17 Dec 12:23:34 - DEBUG: vs.service.load.5:0.02|g
17 Dec 12:23:34 - DEBUG: vs.service.load.60:0.02|g
...

Multithreaded transcoder pipeline

The decoding part of the transcoder is now multithreaded for I-frame-only content. This means that for content such as ProRes, D10/IMX, etc, you will see a speed-up, especially if the input material is in high resolution.

FileCatalyst transfers

FileCatalyst is now available as a transfer method between storage locations. The Vidispine application acts as a FileCatalyst client which can communicate to FileCatalyst servers for transferring files.

In order to register a FileCatalyst server for a storage, add a new transfer method to the storage.

<method>
  <uri>filecatalyst://fc:s3cret@localhost:2100/incoming/</uri>
  <type>TRANSFER</type>
</method>

For more information, see FileCatalyst Integration.

StorNext file information

For files residing on a Quantum StorNext file system, Vidispine can now show metadata on the file. In order to use this, enable web services on StorNext and add a HSM method to the storage.

<method>
  <uri>stornext://webservice:webservice@localhost:81/stornext/snfs/</uri>
  <type>HSM</type>
</method>

For more information, see StorNext Integration.

Standalone metadata

It is now possible to create standalone documents with arbitrary metadata. If you are using global metadata but need to store a large amount of data, leading to a large metadata documents, then consider splitting it up into smaller documents, for example by entity or group of entities.

PUT API/document/company_a
<MetadataDocument xmlns="http://xml.vidispine.com/schema/vidispine">
  <timespan start="-INF" end="+INF">
    ...
  </timespan>
</MetadataDocument>

Filters and facets

Search filters have been added to replace the facet filters. They support arbitrary queries compared to facet filters that only allow a single field to be queried.

Filters can also be excluded from certain facets. This can be used to reduce the number of search requests need to display search pages that uses multiple facets and multiple drill down options.

PUT API/item
Content-Type: application/xml

<ItemSearchDocument xmlns="http://xml.vidispine.com/schema/vidispine">
  <filter name="typeFilter">
    <field>
      <name>mediaType</name>
      <value>audio</value>
    </field>
  </filter>
  <facet count="true">
    <field>mediaType</field>
    <exclude>typeFilter</exclude>
  </facet>
</ItemSearchDocument>
<ItemListDocument xmlns="http://xml.vidispine.com/schema/vidispine">
  <hits>1</hits>
  <item end="+INF" id="VX-361763" start="-INF">
    <timespan end="+INF" start="-INF"/>
  </item>
  <facet>
    <field>mediaType</field>
    <count fieldValue="none">1867</count>
    <count fieldValue="image">33</count>
    <count fieldValue="video">10</count>
    <count fieldValue="audio">1</count>
    <count fieldValue="data">1</count>
  </facet>
</ItemListDocument>

Other

  • It is no longer necessary to use the application server’s /tmp directory to store output files for Azure, S3, FTP destinations. Instead, this can be handled using segment files on the destination storage.
  • The timestamp handling for generating MP4/H.264 files have been rewritten. This means that proxy files are frame accurate without without the previous tweaks (useDTSmode et al).

For the full list of changes, see the release notes.

4.2.2

Thumbnails on cloud storages

Thumbnails can now be stored on cloud storages such as Amazon S3 and Azure. Thumbnails will be stored using one file per thumbnail.

<ResourceDocument xmlns="http://xml.vidispine.com/schema/vidispine">
  <thumbnail>
    <path>direct+azure://:+kLZrqckLZrqckLZrqckLZrqc==@mystorage/my-container/</path>
  </thumbnail>
</ResourceDocument>

Job pools

Use job pools to ensure that long running low priority jobs don’t block high priority jobs from running.

PUT /configuration/job-pool
Content-Type: application/xml

<JobPoolListDocument xmlns="http://xml.vidispine.com/schema/vidispine">
  <pool>
    <priorityThreshold>HIGH</priorityThreshold>
    <size>2</size>
  </pool>
  <pool>
    <priorityThreshold>LOWEST</priorityThreshold>
    <size>3</size>
  </pool>
</JobPoolListDocument>

Joins

Support for joins has also been added to allow for cross-entity search between items, shapes and files.

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<ItemSearchDocument version="2" xmlns="http://xml.vidispine.com/schema/vidispine">
  <text>peach</text>
  <shape>
    <field>
      <name>language</name>
      <value>en</value>
    </field>
  </shape>
</ItemSearchDocument>

Platform

This release brings support for:

  • PostgreSQL 9.3.
  • Java 7 update 67.

Other

  • Signiant can now be used to transfer files between storages.

For the full list of changes, see the release notes.

4.2.1

WADL

A number of improvements have been made to the WADL file. Missing parameters have been added and duplicate parameters have been removed for example. It has been updated to also include:

  • Parameter options.
  • Markers for repeating parameters.

The WADL file can be obtained using GET API/application.wadl.

Other

  • Support for copying of DTV 608/708 Closed Captions.
  • Files can now also be sorted by file extension.

For the full list of changes, see the release notes.

4.2

API documentation

The API documentation has been moved from the Vidispine wiki into this documentation that you are now reading.

It is available online at http://apidoc.vidispine.com/latest/ and also on your local installation at /APIdoc.

Java 7

The 4.2 series now requires Java 7, specifically Java 7 update 25 as later versions have known bugs with GlassFish 3.x.

Efficient file I/O

The new Java 7 File API is used to reduce the number of file system operations that are used when scanning local storages.

Indexing

The Solr configuration now exists in the indexing configuration, but note that the Solr configuration properties are still supported.

This configuration can also be used to specify which fields should be included in the full text index, unless specified explicitly for a specific field.

PUT /configuration/indexing
Content-Type: application/xml

<IndexingConfigurationDocument xmlns="http://xml.vidispine.com/schema/vidispine">
  <solrPath>http://localhost:8088/solr</solrPath>
  <fieldDefault>
    <name>xmp_*</name>
    <fullText>false</fullText>
  </fieldDefault>
</IndexingConfigurationDocument>

Detect renamed files

Renamed files can be detected and and re-associate based on the file checksum. Enable it using the detectRenamedFiles storage property.

Platform

This release adds support for Ubuntu 14.04, Windows 2012 R2 and MySQL 5.6. At the same time, support for PostgreSQL 8.x, MySQL 5.1 and Java 1.6 has been discontinued.

Other

For the full list of changes, see the release notes.