Changes for page 212 - Failover Fiesta

Last modified by Carlijn Kokkeler on 2024/04/18 13:07

From 114.1 to 113.1 From 186.1 to 185.1

From version 185.1

edited by Carlijn Kokkeler
on 2023/10/23 14:44

Change comment: There is no comment for this version

To version 114.1

edited by Erik Bakker
on 2023/05/09 13:41

Change comment: There is no comment for this version

Raw
Rendered

Summary

Page properties (3 modified, 0 added, 0 removed)

Details

Page properties

Title

@@ -1,1 +1,1 @@
--207 - Aligned State
++197 - Pay Attention

Author

@@ -1,1 +1,1 @@
--XWiki.CarlijnKokkeler
++XWiki.ebakker

Content

@@ -2,100 +2,62 @@
  {{container layoutStyle="columns"}}(((
  [[image:Main.Images.Release Blog.WebHome@release-blog-intro.png]]
--**Hi there, eMagiz developers!** We have done much work for state generation, so that you can derive more information from your data! Examples are enrichment, aggregation, change detection and duplicate detection. State generation functionality is only accessible for models with an add-on state generation license. Next to state generation, several improvements regarding the alignment of components have been made. Moreover, performance improvements and bug fixes have been implemented. Lastly, a change in the metrics storage duration has been performed.
++**Hi there, eMagiz developers!** In the last few weeks, we have worked hard on improving our next-generation architecture and introduced alerting functionality on top of our event streaming pattern based on user feedback. Furthermore, several smaller feedback items and bugs have been resolved with this release. So without further ado, let us dive into all we have to offer.
--== **State Generation** ==
++== **Event streaming alerting** ==
--//__New components__//
--The following components have been added:
--* Aggregator: reverse of the splitter, used to combine multiple messages into a single one.
--* Infinispan metadata outbound channel adapter: Used to (temporarily) store metadata, which can be used to enrich messages, or make decisions in filters or scripts.
--* Content enricher: used to add properties to the payload of messages, similar to what the header enricher can do for headers.
--* Duplicate detector: Support object used to detect duplicate messages for a certain key on an inbound channel of a flow component. Duplicates can be marked or discarded.
--* Infinispan support objects, supporting the aggregator, metadata outbound channel adapter and content enricher.
++This release expands our alerting functionality into the event streaming pattern. As of now, one new "static" alert has been introduced for all clients (using event streaming), and two "dynamic" alerts have been introduced that you can configure yourself if there is a need for it.
--//__SpEL functions & JSON__//
--Added SpEL functions for:
--* Encoding & decoding Base64, Hex and hmac
--* Date retrieval (currentDateTime, currentEpochMilli, currentEpochSecond)
--* Date conversion (DateTime to Epoch, Epoch to DateTime, DateTime formatter and DateTime parser)
--* Added support for SpEL property accessors for XML and JSON, meaning that you can read message payloads in SpEL expressions easily. For example, to retrieve the value for 'id' in the following JSON String: `{"id":"123"}`, this SpEL expression suffices: `payload.id`.
--* Added support for `#JsonPath` usage in SpEL expressions
++The "static" alert we have added raises an alert (and a subsequent email) when the actual topic size crosses the threshold of 80% of the configured maximum retention size on a topic. This alert provides insights into whether messages on it are deleted due to a size or time constraint. In cases where data is deleted, the topic was too full way before the messages should have been removed due to the retention hours constraint. This alert can indicate that messages might be deleted before consumer groups had the option to consume the messages.
++The two "dynamic" alerts we have added mimic the alerting on the queue level we already offer to the community.
--== **Metrics Storage Duration** ==
--
--
--
++[[image:Main.Images.Release Blog.WebHome@197-release-blog--new-alerting-options.png]]
--== **Feedback Items** ==
++The first new "dynamic" alert allows you to raise an alert once the number of messages on one (or more topics) is less than a certain number per defined time unit. So, for example, you can configure an alert once the number of messages placed on a topic called "Exception" is less than 15 messages within 5 minutes.
--//__Alerting manual pause__//
--A few releases ago we changed the behavior of alerting in the deployment plan. Now each time when a deployment plan is executed the alerting will be automatically re-enabled when the deployer closes the deployment plan or closes the web browser. The majority of the users are happy with the new behavior, but there are some use cases that you do not want start the alerting immediately. With this release, if alerting has been paused manually, this will not be activated automatically after a release deployment.
++[[image:Main.Images.Release Blog.WebHome@196-release-blog--alert-config-example.png]]
--//__Ordering of graphs in Manage__//
--The graphs in Manage are now ordered according to importance. This means that runtime statistics are sorted by "Process CPU usage" (highest first), queue statistics are sorted by "Messages in queue" (highest first), and HTTP statistics are sorted by "Unsuccessful requests" (highest first).
++The second new "dynamic" alert allows you to raise an alert once one (or more) consumer groups are more than X messages behind on one or more topics. The configuration of this is comparable to what we saw before.
--//__UTC times in Grafana panels__//
--All Grafana panels now show UTC times, which are normally used in eMagiz, instead of local (browser) time zones. This way, it is easier to match graphs with logging events or alerts.
++For more information on the generic way of working surrounding alerting, please check out this [[microlearning>>doc:Main.eMagiz Academy.Microlearnings.Crash Course.Crash Course Platform.crashcourse-platform-manage-alerting-gen3||target="blank"]] and this [[microlearning>>doc:Main.eMagiz Academy.Microlearnings.Novice.Alerting.novice-alerting-dynamic-alerting.WebHome||target="blank"]].
--//__Update flow designer version__//
--The framework used in the flow designer has been updated, improving performance.
++== **3rd generation improvements** ==
--//__Carwash track TLS versions in logging__//
--A new logging feature will be released, enabling us to make better choices in deprecating old encryption standards.
++//__Increased grace period for shutdown__//
++In this release, we have increased a container's grace period to shut down before it is forcefully shut down. This should reduce the chance of unwanted failover behavior within your model.
--//__Moving channels in the flow designer__//
--Moving already attached channels in the flow designer has been made sligthly easier.
++//__Update at once or not__//
++This release fixes and re-introduces the option to execute your actions on the eMagiz per zone or all at once when running in a double-lane Docker setup.
--//__Topic sizes description change__//
--In the change description (and History) when altering the topic size of a topic the new and old value were switched around creating confusion, this has been resolved.
++//__Improved user feedback while executing a deployment plan__//
++This release introduces additional feedback to the user when the deployment plan is executed. This is noticeable when a step cannot be executed properly. The relevant information about why this cannot be executed is shown to the user. This way, they can take this information and act upon it instead of assuming everything went well.
--//__Password change notification__//
--When an account password change request is made, even when this fails, a mail is sent to the account owner to inform the owner about the action.
++//__Improved auto-healing when running in a hybrid situation__//
++In situations where you run in a hybrid situation (i.e., partly next-gen and partly the current generation), we have improved the auto-healing functionality in case an "out of memory" appears on a runtime running in the current generation but on a next-generation architecture.
--//__Password comparison__//
--When changing a password, it is compared to a list of known database breaches for security. A warning is shown when the password corresponds with a password in the database.
++== **Feedback items ** ==
--//__Inactive user alerting__//
--Inactivated users are now removed from all alert settings (included “disabled“ settings) to avoid undesirable notifications.
++//__Make sure the message format can be viewed without "Start editing"__//
++With this release, we have ensured that when you navigate to Design -> System message, you can see the message format (i.e., XML, JSON, or EDI) without entering the "Start Editing" mode.
--//__Alphabetical sorting on user level in HTTP statistics__//
--Variables in the 3rd generation runtimes HTTP statistics detail pages are now sorted case insensitive.
++//__Various styling improvements in the flow testing functionality__//
++Various minor styling improvements have been added to the flow testing functionality to improve the overall user experience. Please check out the release notes for a complete list and more details.
--
--== **Bug Fixes** ==
--
--//__Flow designer styling__//
--The styling of the flow designer's left component panel has been restructured, solving a rare bug which would break the styling of certain functionalities.
--
--//__Partial search for messages__//
--It is now possible to search on messages partially in Manage Monitoring. For example, a search for Uptime can be done by searching for "up" "time" "ptim".
--
--//__Disk usage after cloud template update__//
--In the last cloud template update there was an issue with disk performance. This has been resolved in this release. You can manually upgrade your cloud template, or rely on automatic updates.
--
--//__Error handling migration__//
--If there is no custom error handling, when migrating to Gen3, the error channel to “errorChannel” is only created for the first inbound in an entry flow. This has been fixed by adding a migration step, where we set the error channel of all inbounds in a flow to “errorChannel” if the custom error handling is set to false.
--
  == **Fancy Forum Answers** ==
  As always, this is a gentle reminder to ask questions via the Q&A forum. The Q&A forum is available in the eMagiz iPaaS portal, so we can all benefit from the knowledge within the community. For some inspiration, take a look at these forum answers:
--* [[JSON Web Tokens (JWT authentication)>>https://my.emagiz.com/p/question/172825635703606919||target="blank"]]
--* [[Determining container sizes & reading architecture pages>>https://my.emagiz.com/p/question/172825635703594204||target="blank"]]
--* [[SFTP connectivity - supported algorithms>>https://my.emagiz.com/p/question/172825635703607141||target="blank"]]
--* [[Deploy property release results in missing flows in runtime>>https://my.emagiz.com/p/question/172825635703607354||target="blank"]]
--* [[MessageDeliveryException: Dispatcher has no subscribers>>https://my.emagiz.com/p/question/172825635703619797||target="blank"]]
++* [[Update Core01 without downtime?>>https://my.emagiz.com/p/question/172825635703427783||target="blank"]]
++* [["EDI" mapping?>>https://my.emagiz.com/p/question/172825635703440524||target="blank"]]
--
  == **Key takeaways** ==
--Thanks to all who helped build and those who tested and gave feedback to improve the product. To end this post, here are some key takeaways for you:
++Thanks to all that helped build and those who tested and gave feedback to improve the product. To end this post, here are some key takeaways for you:
  * If you have questions surrounding our Program Increment Planning, please get in touch with [[productmanagement@emagiz.com>>mailto:academy@emagiz.com]]
  * If you have feedback or ideas for us, talk to the Platypus
--* Please inform us of new additions to the store ([[productmanagement@emagiz.com>>mailto:academy@emagiz.com]]) so we can all benefit from these.
++* Please inform us of new additions to the store ([[productmanagement@emagiz.com>>mailto:academy@emagiz.com]]) so we can all benefit from these.
  * Clear your browser cache (Ctrl + Shift + Del)
  * Check out the release notes [here]
  * Start thinking about how the license tracker can aid your development
@@ -108,6 +108,6 @@
  {{info}}
  ~* Indicates a Beta feature. If you would like to get access to this beta feature, please contact [[productmanagement@emagiz.com>>mailto:academy@emagiz.com]]
--~*~* Indicates a next-generation-architecture only feature.
++~*~* Indicates a GEN3-only feature.
  {{/info}})))((({{toc/}}))){{/container}}
  {{/container}}

Changes for page 212 - Failover Fiesta

Summary

Details

Navigation