Squirro 2.4.5 - Aspen - Release Notes

Owned by Felix Hürlimann (Deactivated)
Last updated: Aug 24, 2017 by Oliver Seeliger (Unlicensed)
4 min read

Released 18th January 2017

New Features

  • Secure your Configuration: protect critical information by storing Squirro configuration files in encrypted form - see our document on Securing your Configuration Files for details
  • Redhat 7 Readiness: stand-alone Squirro installations now support RHEL 7 and Centos 7 (state: preview release). Support for cluster coming soon
  • More flexible dashboards: we introduce "Dashboard Layering" allowing parts of dashboard to be displayed depending on conditions other parts of the dashboard (currently only available by editing the raw dashboard JSON definition). More details here.
  • Automate loading dashboards into projects: using the squirro_asset tool it's now possible to upload raw JSON dashboards into projects. Great for automating project setup. More details here.
  • A new, more powerful dataloader: a big update to the dataloader incorporating many of things we've learned working on countless Squirro projects for our customers. See below for full details
  • Upgrading clusters of 2 or more Squirro servers is a breeze: Squirro services proactively react to changes on other servers to keep themselves available. Now clusters of Squirro servers upgrade faster. (Added 31th January 2017)

Improvements

  • Faster, Faster, Faster, Faster: we overhauled the core of Squirro improving search response times by as much as four times

  • Bing Bling: we updated the search provider to support the latest version of Microsoft's Bing API, now running on Azure
  • Known Entity Extraction (KEE) keeps getting better: now with customer tokenizers, the ability to download keywords into fixtures and many more small fixes and improvements
  • Better in Bulk: the filtering service (and trend detection) now work with bulk import 
  • More flexible typeahead: now supports query templates
  • Pick the time in all your dashboard designs: the time selector drop-down can now be displayed even when space around widgets is limited
  • Size doesn't matter: now easier to pause or edit data sources on smaller screen sizes
  • It's all relative: trend detection now supports relative time ranges
  • Keep turning pages: paging through results is now faster and more reliable
  • Smarter defaults: Read and starred flag aggregation is disabled by default. You can re-enable it by configuring 'enable_flags_for_project_ids' in the topic.ini
  • Ability to upgrade to the latest Java JDK security fixes in jdk 1.8.0 release 144 from July 2017 (added August 24, 2017)

Bug Fixes

  • Fixed memory leak in 2.4.3 feed provider service 

  • Fixed issue with Facet Table detail view failing to close on click-again 

  • Fixed issue with unreliable language detection in multi-page documents 

  • Fixed issue where deduplication step removes S3 attachments 

  • Respect custom Facet Table column widths on detail

  • Fixed issue with Squirro sdk executables, e.g. pipelet, squirro_widget broken in 2.4.4 by RPM dependencies problem 

  • Fixed issue where fingerprint display fails if training items have no title 

  • Fixed issue where fingerprint creation times out after 55 seconds 

  • Fixed issue where data loader throws wrong error about missing map-title and map-body when templates are present for both 

  • Fixed issue where facets with value zero are passed as undefined when navigating with them from facet selector (or the left) 

  • Fixed issue where pie chart legend not shown when maximised 

  • Fixed issue where queries that request more than 10000 items break (problem since release 2.3.3) 

  • Fixed problem with Monit making unwanted restarts of Squirro services 

  • Fixed issue with facets of type “Auth" getting recommended for typeahead 

  • Fixed problem where it was no longer possible to disable pipelet validation 

  • Fixed various issue with cache failing to expire under certain situations 

Dataloader Update

Numerous improvements and fixes including...

  • Added support for parallel uploading
  • Batch size handling for multiprocessing
  • Encoding handling for Oracle
  • Better progress logging
  • Avoid splitting for non-string values
  • Better error reporting for value errors
  • Smarter handling of Unicode
  • More useful error logging
  • Added --dry-run option to test data loading without actually sending the data
  • Handling of pipelets that yield multiple items
  • Handling of empty mapped values
  • Allow setting of "analyzed" facet property
  • Fixed updating of facet config
  • Fix wildcard expansion in facets config
  • Add support for pivoting data
  • Auth columns now supported
  • Handle deletions in new model
  • Remove legacy arguments
  • Fix import for large CSV lines
  • Retain column order in Excel sources
  • Allow hard-coded job IDs


Fresh Installation Instructions

Please follow the regular installation steps.

Upgrade Instructions

To upgrade to version 2.4.5 of Squirro, please ensure that your current version is at least version 2.4.3 or higher. This is because of squirro rpm version number changes from "0.1" to "2.4.4". If you are on a version older than 2.4.3, please contact support.

Additionally if you are using Squirro in a Box, additional steps are involved. In this case we also ask you to contact support.

From version 2.4.4

1. Upgrade Storage Nodes and Cluster Nodes collocated on the same machine/VM

[squirro@singlenode ~] sudo yum update squirro-storage-node-users
[squirro@singlenode ~] sudo yum update squirro-storage-node
[squirro@singlenode ~] sudo yum update squirro-cluster-node-users
[squirro@singlenode ~] sudo yum update squirro-cluster-node

2. Upgrade Storage Nodes (separate from Cluster Nodes)

Upgrade all storage nodes one at a time by running:

[squirro@storagenode01 ~] sudo yum update squirro-storage-node-users
[squirro@storagenode01 ~] sudo yum update squirro-storage-node

3. Upgrade Cluster Nodes (separate from Storage Nodes)

[squirro@clusternode01 ~] sudo yum update squirro-cluster-node-users
[squirro@clusternode01 ~] sudo yum update squirro-cluster-node

Upgrade the Java JDK 1.8.0 to release 144

1. Upgrade Storage Nodes and Cluster Nodes collocated on the same machine/VM

[squirro@singlenode ~] sudo yum install jdk1.8.0_144
[squirro@singlenode ~] sudo service elasticsearch restart
[squirro@singlenode ~] sudo service sqpipelined restart

# on centos 6 and redhat 6 only:
[squirro@singlenode ~] sudo service zookeeper restart

[squirro@singlenode ~] sudo yum erase jdk1.8.0_111

2. Upgrade Storage Nodes (separate from Cluster Nodes)

Upgrade all storage nodes one at a time by running:

[squirro@singlenode ~] sudo yum install jdk1.8.0_144
[squirro@singlenode ~] sudo service elasticsearch restart
[squirro@singlenode ~] sudo yum erase jdk1.8.0_111

3. Upgrade Cluster Nodes (separate from Storage Nodes)

Upgrade all cluster nodes one at a time by running:

[squirro@singlenode ~] sudo yum install jdk1.8.0_144
[squirro@singlenode ~] sudo service sqpipelined restart

# on centos 6 and redhat 6 only:
[squirro@singlenode ~] sudo service zookeeper restart

[squirro@singlenode ~] sudo yum erase jdk1.8.0_111