Features

• We’ve introduced a new hover-over feature in the Executive Summary (PBI 666) - following feedback from clients we’ve introduced hover-overs of the behaviour cards in the Executive Summary – for example, if Air is 50% compliant, you can hover over the card to see why it’s 50% - removes the initial need to navigate to the Air dashboard for the information. You need to be in “behaviours” and look out for the “show details on hover” switch.
• In the Executive Summary, the monthly spend visualisation is currently based on calendar year, so we’ve introduced a switch that enables users to move to a ‘rolling 12 months’ view, in PBI 681.

Bugs

• We had a user-experience fault with regards to downloading filtered data – bug 650 resolves that. If no filters are in place, the option for filtered download isn’t available.
• Bug 705 resolved an issue with Meetings dashboards rendering poorly because of issues with the X-axis. We’ve tilted the numbers so the graph now makes more sense!
• In the Summary dashboard, the monthly spend X-axis labels were incorrect so we’ve fixed it in bug 696.

This upgrade can be completed via Keel. Please get in touch if you have any questions

• In PBIs, 690, 701, 693, 692 and 691 we modified the logic with the Meetings summary cards – only data where the event and supplier status’ are “confirmed” or “cancelled” will the data being calculated.
• We modified the logic relating to preferred locations (PBI 686) in Meetings too – a preferred location is now a preferred property in a preferred location.
• A topic of heated debate within the Unlocked Data ranks! We’ve modified the variation percentages in the summary cards; for the avoidance of doubt, this is the calculation: ((<current reporting period> - <previous year reporting period>)/<previous year reporting period>) * 100.
• A really good update to the Meetings dashboard Top Spend visualisation – we’ve given it an overhaul and it’s now so much more informative! PBI 601 completed!
• We’ve replaced the Lead Time donut in Meetings with Cancellations – PBI 600 created to bring more insights to the dashboard.
• We’ve updated the SupplierStatus data model to include more items in 701 – the data definitions will be updated accordingly. We’ll send the latest specifications out.

• In the Meetings dashboard, the dashboard didn’t deal with missing data properly and the dashboard would fail – if there was no hierarchy data and clients switched to the league table view, there was an ugly error – all sorted in Bug 704.
• We introduced a bug in the Management Portal where the client configuration wasn’t being displayed accurately – that was fixed in Big 702.
• As a part of our annual third-party scan, we had an issue with outdated JS libraries that was addressed in Bug 699.
• Bug 697 solved an ugly problem with the summary cards in the abridged dashboard – filtering in the widget broke the joins between the data and visualisations.
• We had an issue with client-side spend reporting within the Meetings dashboard that we resolved in 695.
• The YTD visualisation didn’t include the previous year for comparison – bug 680 solves the problem.
• We had an issue where the hierarchy wasn’t linked to sustainability reporting, but that’s been solved as a part of 676.
• Bug 652 solved an issue with the behaviour donuts – where there was no data, the donuts wouldn’t populate. 652 sorts this problem out.
• On the Meetings dashboard, we broke the Bookings Trends and Prices user experience – it didn’t work well, so we’ve updated that in Bug 555.
• Testing picked up an issue with geo-coding in the Executive Summary – we solved that in 684.

This upgrade can be completed via Keel. As there are multiple images in this release select all images and approve. The upgrade process is smart enough to ensure that your solution is running the very latest image, so click to approve all images. The Keel dashboard, Pending card, will confirm when all images have been applied.

• We’ve introduced a currency service that will consolidate spend into client defined currencies. For our clients where data is consolidated globally this feature enables configuration at a client level and more accurate financial reporting – related to PBIs 603 (engineering) and 644 (user experience)
• We’ve created a Print Dashboard feature – with a minimal number of clicks from the Abridged dashboard you get a preformatted dashboard deck that be converted to PDF (or printed) for distribution or Exec reporting

• In PBI 689, we added PolicyApprover and TripApprover to the widget for additional filtering.
• Similarly, PBI 657 added in POSCountryCode as a filter item in the widget.
• We’ve removed the sustainability card from Meetings – there was broad consensus from our clients that data wasn’t available to support carbon reporting for meetings and events. In PBI 658, we removed it.
• We’re not actually receiving any Car data from our clients at this time and the Car card is complicating matters – so we’ve decided to remove the Car card from our reporting for the time being. It can be re-introduced any time (PBI 664)
• The Abridged dashboard has additional navigation between cards – the arrows were the only way to navigate between dashboards. In PBI 664 we’ve introduced better navigation by clicking on the cards themselves to get to the relevant dashboard.
• The well-being calculations were complicated, so we created PBI 564 to simplify – and we think its much improved.
• PBIs 562 and 613 are refinements to the Savings dashboards. Improved visualisations, more answers to questions, improved navigation – we’re looking forward to feedback on this. A good addition to the dashboards, made better.

• Bug 687 resolved an issue with geo-coding on the Executive Summary map – dots may have appeared in odd places on the map – a problem that will only occur now, if the correct latitude and longitude have been used.
• We created Bug 688 during this release! Testing identified an issue within the Management Portal when new clients were created.
• We had an issue where Meetings data was being multiplied incorrectly. Bug 659 resolves this issue.
• The Top Level hierarchy in Hotels looks a little bit rubbish due to an alignment problem – that issue has been resolved in Big 656
• The minimum spend slider on the Executive Summary (used when digging into behaviours) was calculating numbers ‘on the fly’ and frankly, it looked and performed badly. Bug 605, sorted this out.

It is recommended (bug fixes, stability, and new features) that all clients upgrade to the latest release of the Nginx controller within their environments.

The Nginx controller is the service that is responsible for routing external traffic correctly.

Please note that this upgrade is a pre-requisite to upgrading to Kubernetes 1.22.4

Azure clients please refer to the points below – all other clients please log a ticket if you need assistance for your platforms.

1. Go to Azure and click to go to your AKS Cluster
2. Click on Workloads and then select the “Deployments” tab
3. Click on the deployment “external-access-ingress-nginx-controller-unlockeddata”
4. Click YAML on the left-hand menu
5. Scroll through the YAML until you reach the image tag, as per below (the line should be on, or around, 223)
6. Replace all the text starting “k8s.gcr.io…” (in example above, line 224) and replace with the following text:
   k8s.gcr.io/ingress-nginx/controller:v1.1.1@sha256:0bc88eb15f9e7f84e8e56c14fa5735aaa488b840983f87bd79b1054190e660de
7. Click “Review and Save”
8. Tick to “Confirm manifest changes” and click “Save”
9. This will automatically upgrade the Nginx controller (the virtual appliance that controls access to the solution) to the latest version (v1.1.1).

In response to the bug notification on Monday 17th January (related to Executive Summary dashboard) wee have resolved the issue and version 2.3.34 is now available in Keel. As always, the process ensures that your solution is running the very latest image, so click to approve all images. The process should take less than two minutes.

• We picked up on an issue in the Executive Summary where the “Cumulative Travellers” and “Travellers By Month” visualisations were incorrectly showing no travellers.

Please got in to Keel and upgrade images to version 2.3.33. As always, the process ensures that your solution is running the very latest image, so click to approve all images. The process should take less than two minutes.

• Exciting change! We’ve introduced a new function that enables filters to be applied across the dashboards. Any filters applied (date range, traveller, cost centre, and so on) will remain in place across all dashboards.


• We’ve made some big changes to the Meetings data model so the dashboards can visualise even more interesting data to provide improved insights. We’ll be distributing the changes so you can see the details.

This upgrade can be completed via Keel. As there are multiple images in this release (details below), select all images and Approve. The upgrade process is smart enough to ensure that your solution is running the very latest image, so click to approve all images.

• • As we work with different online authentication providers we found a need to introduce more granular logging to help diagnose issues. PBI 646 sees a flag introduced that enables more granular logging when troubleshooting log in issues (image 2.2.16 in Keel, if you’re interested!):
• Moving the namespace key from values.yaml to client-bespoke-values.yaml – we see this as a cleaner way to enable our clients to host multiple namespaces – and therefore solutions - on a single Kubernetes cluster.
  • Logging can be enabled, but should be enabled when troubleshooting is complete – A LOT of data gets captured!
• We needed to introduce a new data field for authentication for Ping. PBI 647 introduces as an additional field that enables Ping authentication. (image 2.2.17)

• We found an issue where new clients created in the Management Porta weren’t always downloading the latest solution image. Bug 642 resolves this problem (image 2.2.12).
• Keel image 2.2.13 resolves an issue associated with 642 above.

Helm Chart Upgrade

• There are minor changes in the 2.1.30 Helm chart, including:
• Moving the namespace key from values.yaml to client-bespoke-values.yaml – we see this as a cleaner way to enable our clients to host multiple namespaces – and therefore solutions - on a single Kubernetes cluster.
  • For example, UAT and Production could reside on the same cluster.
• We’ve made changes to client-bespoke-values.yaml (“bespoke config 7”) that enable us to be smarter with client and server-side client implementations.
• The deployment – nginx controller.yaml file has changes that reflect the changes made above for client and server-side implementations.
• The configmap.yaml file has changes that support client and server-side implementations.>

This release will require a Helm upgrade:

• Please back up Helm chart 2.0.28 as it will contain content specific to you.
• Download and export Helm chart 2.1.30 to a neew folder
• Update the values in the “client-bespoke-values.yaml” file – you can use the backup from your 2.0.28 deployment as reference.
• Bespoke Config 2 will need to be updated to reflect your IDP provider details.
• If you assign static IP addresses to any services, you will need to update Bespoke Config 3.
• You may assign DNS names via Bespoke Config 4 – ensure these are updated in the new file.
• Modification to Bespoke Config 7 – you will need to enter relevant domain information here.
• Perform the Helm upgrade using “Helm upgrade…” command
• Please note, by adding --set abridgedAvailable=”true” will install the abridged client dashboards.
• or Example: helm upgrade -f client-bespoke-values.yaml unlock8s . --set abridgedAvailable="true"

We think that the changes are self-explanatory, but please get in touch prior to the upgrade if you have any questions. We will support before, during and after the implementation.

Software will be automatically deployed in Engineering step above. No Keel upgrade needed

Features:

• We’ve introduced an easy-to-use method of migrating clients from client-side to server-side (and back, if wanted!) via the Management Portal. It’s really quick and very easy thanks to PBI 604.
• To enable greater flexibility for our clients we’ve increased the number of hierarchy fields from 5 to 10. PBI 630 now enables our clients to apply more granular data filters.
• Not the biggest news in this release – we’ve added a specific icon for the Savings dashboard in PBI 607
• We’ve added the ability to filter by point-of-sale country code in PBI 612 – enables dashboard users to see performance by country. A nice addition.

Bugs:

• Line graphs in the Meetings dashboards weren’t displaying values correctly – a lot of congestion. In Bug 577 we rotated the values on the X axis so that they were clearer.
• In the drill-down section of the Air dashboard we hadn’t linked the four tables. Bug 585 now enables full interaction so when selecting a row of data in one table, the other three tables change context.
• We’ve renamed the data filter button from “Cancel” to “Close” in Bug 622.
• The data filter was case-sensitive which gave false results. In Bug 623 we removed that case-sensitivity so that the data filter is less sensitive.
• With the data filter, there’s the potential to stop a user seeing any data. The dashboards error in this instance, so what we’ve done in Bug 624 is simply display an error message and dashboards with no data.
• We noticed that the data import process had case-sensitivity associated with the ClientID field. Bug 629 solves that problem.
• We had some behind the scenes work to do because of an error when migrating from server-side to client-side. Bug 631 removed the value that caused the error.
• There was an error when attempting to assign data filters to users who were not assigned clients. Now, if a user is not assigned to a client, the option to apply a data filter isn’t available thanks to bug 632.
• When interacting with the data filter (hierarchy management) but ultimately not making any changes the data filter would error – we’ve fixed that in bug 633.

Helm Chart Upgrade

• This upgrade sees the introduction of Server-side processing for ‘mega-clients’ where client-side processing performance was too slow. A direct result of this change (PBI 563) is significant performance increases.
• Please note – the migration of a client is a manual process at this time. We will provide instructions and ‘hand-holding’ for those clients that need to move to server-side. Not to worry – we do have a PBI in the backlog that will automate the process.
• Following on from the licensing introduced by Elastic, we are migrating monitoring and alerting to OpenSearch – the open-source community’s derivative of Elasticsearch and Kibana.
• This deployment will ‘trash-and-burn’ the existing Elasticsearch Kibana instance. All log data will be lost and Index’s will need to be recreated.
• The deployment also see us migrate to Azure persistence storage (rather than file share) in order to reduce Shard failures previously observed.
• Additionally, OpenSearch enables alerting and triggers that will enable proactive notifications.
• We’ve tidied up the Helm chart too – removed a now redundant deployment that created a client that was never pertinent.

This release will require a Helm upgrade:

• Please back up Helm chart 1.2.3 as it will contain content specific to you.
• Download and export Helm chart 2.0.21.M
• Copy the “client-bespoke-values.yaml” file from your 1.2.3 deployment to the 2.0.21 templates folder.
• This ensures that your specific configuration is deployed when you upgrade.
• Copy the “deployment – nginx controller.yaml” file from your 1.2.3 backup and paste in to the 2.0.21 templates folder.
• This ensures that your specific TLS settings and ingress rules are included in the upgrade.
• Perform the Helm upgrade using “Helm upgrade…” command
• Please note, by adding --set abridgedAvailable=”true” will install the abridged client dashboards.
• For Example: helm upgrade -f client-bespoke-values.yaml unlock8s . --set abridgedAvailable="true"

Features:

• We’re delighted to announce the introduction of our dashboard authorisation (client hierarchy) via the Management Portal. Using the Management Portal, it’s now possible to set up filtering rules that control what users can – or can’t – see. You can now control who sees what. An exciting development that we’ve introduced following lots of demand.
• Additionally, we’ll be including the Beta release of our Savings dashboard. The dashboard will give you an indication as to what the final release will look like. There’s some refinement planned, but PBI 562 gives exciting insights in to what we’re doing with your data!

Bugs:

• The Executive Summary was light on help icons and text to explain what data was being displayed, so we created bug 374 to provide more context to graphs and tables.
• We had an issue where accepted savings on Air and Hotel weren’t being calculated correctly. This has been resolved via bug 595.
• Bug 538 resolved the issue with the minimum spend slider on the Executive Summary – previously, the slider wasn’t changing the context of the league tables above.
• Responding to feedback, we created bug 329 that introduced control panels on graphs that enabled year-on-year visualisations to be calculated and displayed.
• Previously, we didn’t get the user experience relating to Behaviours quite right. So, we created bug 453 to simplify the user experience so that it’s easier to select and deselect options.
• Behind the scenes (and ‘under-the-hood’) we did have an issue with the LAT and LON enrichment API failing on Linux pods within the solution – an issue that bug 594 resolved.
• We were seeing random “stale file handles” on client APIs that prevented data being acquired, but bug 530 solves the problem.

Keel upgrade (includes previous releases)

The first half of Sprint 15 has been heavy on bug fixing.

Bugs:

• The Executive Summary well-being tables had “Reset” buttons that didn’t work. Bug 574 has resolved the problem.
• We introduced performance improvements so that dashboards loaded faster, but air still had a problem. This has now been fixed in Bug 506.
• When browsing back to the Executive Summary performance was slow – we resolved this in Bug 561.
• Maps on the abridged dashboards weren’t always rendering to the full width of the screen, so we solved in Bug 571.
• The abridged spend map wasn’t displaying useful content, so we’ve introduced aggregated spend by destination city in Bug 572.
• Bug 570 resolved the issue of showing traveller destinations on the traveller map within the Abridged dashboards.
• In Bugs 547 and 544 we had tables labelled incorrectly – these have been corrected.
• The Behaviours interactivity across all dashboards was flawed as we had labels that looked like buttons that should be clicked. This is solved in Bug 480.
• On the Air dashboard our advanced bookings calculations were incorrect – the code applied for Bug 519 now ensures that any bookings made 14 or more days in advance are included.
• Within the Abridged dashboard, the traveller map wasn’t populating travellers correctly – this has been resolved in Bug 570.

Features:

• We’ve introduced new dashboards for clients that need them – in PBI 560 we created a mechanism for making them available as needed.
• Additionally, we wanted to add more controls to some of the graphs, so in PBI 578, we’ve introduced a widget so that graphs can be tailored to what you want to see.

Features:

• As you know, we’ve been working hard on upgrading the solution to the .Net 6 framework so that the solution gets all the benefits with improved security, performance, and reliability of a Microsoft release. This release sees all components move to .Net6!
  • note that the Ahead-of-Time (AOT) features are not included in this release. More to follow later this month.

Added in a new deployment and associated service to assist in improving the enrichment of latitude and longitude data for improved map visualisations. Upon data acquisition the service determines whether latitude and longitude data needs enriching for each row of data, and if so, looks up city against a table of cities and geocodes and enriches.

• Please note: as the enrichment takes place upon acquisition of data, client data will need to be reloaded.

This release will require a Helm upgrade:

• Please back up Helm chart 1.2.1 as it will contain content specific to you.
• Download and export Helm chart 1.2.2.
• Copy the “deployment – nginx controller.yaml” file from your 1.2.1 backup and paste in to the 1.2.2 templates folder.
  • This ensures that your specific TLS settings and ingress rules are included in the upgrade.
• Perform the Helm upgrade using “Helm upgrade…” command.

Keel upgrade (includes previous releases)

• (PBI 551) – Increase rows in Hotels drill down tables.
• (PBI 552) – Traveller count summary cards to (minus 1) to account for missing data.
• (PBI 543) – Add annotations relating to CO2 measurements.
• (PBI 533) – Increase levels of drill down in Hotels tables to provide greater analytical depth.
• (PBI 532) – Add / enrich geocoding for improved map visualisations on all dashboards.
• (PBI 527) – Introduction of an abridged client dashboard. Please get in touch for more details.
• (PBI 531) – Add / enrich geocoding for improved traveller map visualisations.
• (PBI 403) – Redesigned Travellers dashboard within Executive Summary.
• (PBI 542) – Migrate solution to .Net6.
  • Please note – this release does not include ‘ahead-of-time’ (AOT).

We’ve made some engineering improvements to the Helm chart in release 1.2.1 Updates include:

• Added centralised external Azure DNS values in to client-bespoke-values.yaml
  • Removed from nginx controller file and enables creation of DNS and CNAME records
  • Refer to client-bespoke-values.yaml “bespoke config 4
• Simplification of TLS configuration by removing requirement from chart to a single Kubectl command
  • TLS certificate and private key information no longer stored in plain test, in Helm chart files
  • kubectl -n unlockeddata create secret tls --cert c:\unlockeddata\tls\unlockeddata.crt --key c:\unlockeddata\tls\unlockeddata.key unlocked-tls
• Keel approvals deadline set to 168 hours (7 days) by default
• Removal of Service Principal details from Helm chart by running one command
  • kubectl create secret docker-registry imagepull --docker-server=unlockeddata.azurecr.io --docker-username= --docker-password='<Password as provided by UD>' –docker-email=<valid email address>
• Keel access port changed from port 9300 to 80
• Removal of internal NGINX controllers and redundant webhooks to simplify solution
• HSTS enabled / disabled flag
  • Refer to client-bespoke-values.yaml “bespoke config 6”
• Re-introduction of static IP address assignment for Management Portal, Kibana and Keel
  Refer to client-bespoke-values.yaml “bespoke config 3

Additionally, confirmation the solution is supported on Kubernetes version 1.20.7 – you can upgrade your cluster and nodes to this latest version.

Keel upgrade (includes previous releases)

• PBI (516) – Air Dashboard drill down tables increased to 5 rows.
• PBI (525) – Keel Deadline increased to 168 hours (7 days).
• PBI (515) – Rail drill down tables increased to 5 rows.
• PBI (513) – Air dashboard ticketing table increased to fill space.
• PBI (452) – Reposition Behaviours to second in each dashboard.
• PBI (491) – Executive Summary sustainability to replace tree carbon offset. Testing passed, additional remarks suggested to fill space.
• PBI (514) – Removed duplicated “Behaviour” from subtitle of Behaviour League Table.
• BUG (529) – Newly created resources created as ClusterIP.
• PBI (484) – Executive Dashboad sustainability (Air Class and Hotel Carbon) data issue fixed and whole section moved to 2x2 grid.
• PBI (504) – StateProvince data field expanding to 26 characters for data loading.