Virtuozzo Application Platform 5.8.2

This document is preliminary and subject to change.

In this document, you will find all of the new features, enhancements and visible changes included to the PaaS 5.8.2 release.

Container Types in Wizard

Added information about deployment options available on the platform to the topology wizardLearn more

Container Reboot via SSH

Added an SSH command to reboot individual container without the necessity to affect whole environmentLearn more

Tasks Panel Amendments

Tuned and optimized the Tasks panel to improve the user experienceLearn more

Isolation Timeout

Increased the timeout for isolation operations to one minuteLearn more

API Changes

Listed all the changes to the public platform API in the current releaseLearn more

Software Stack Versions

Actualized list of supported OS templates and software stack versionsLearn more

Fixes Compatible with Prior Versions

Bug fixes implemented in the current release and integrated to the previous platform versions through the appropriate patchesLearn more

Bug Fixes

List of fixes applied to the platform starting from the current releaseLearn more

Container Types in Wizard

PaaS is an extremely versatile platform that supports multiple solutions for application development and hosting that will suit even the most demanding clients. However, it may not be obvious (especially for the new customers) which deployment options are available on the platform, so a dedicated list with explanations was added to the topology wizard.

wizard container types

By default, the wizard provides a categorized list of platform-managed stacks (the ones in the Java, PHP, Ruby, etc. tabs) that are already pre-configured and optimized for you in the most beneficial way. Additionally, such managed containers are fully compatible with the platform functionality (scaling, automated deployment, redeployment, SSL, etc.).

In case of switching to the Docker tab, you can review other deployment options available on the platform. Herewith, all the variants listed below are utilizing the basic system container, which makes them compatible with most of the platform-distinguishing features (e.g. vertical and horizontal scaling):

  • Custom Docker Containers - any Docker image (based on the supported OS) deployed into the system container, which provides access to a great variety of solutions. However, the software operability and compatibility with the platform cannot be guaranteed (as the content is managed by respective image maintainers).
  • Docker Engine - a Docker Engine CE deployed into the system container. It provides access to all the Docker native functionality, including deployment, scaling, and management of multiple application containers inside.
  • Kubernetes Cluster - ready-to-go Kubernetes cluster with preconfigured master and worker nodes created based on the system containers. Microservices deployment, scaling, and orchestration are handled by Kubernetes control units, while the platform manages control plane and worker nodes as a whole.

More info

Back to the top

Tasks Panel Amendments

A new Tasks panel, that was introduced in the previous release, obtained additional tuning for better user experience:

  • In order to provide better clarity, the Copy and Send Report buttons were moved to the left part of the panel. Also, they will be displayed only upon hovering over the appropriate Params and Response sections.
  • A new Copy error text button was added to get an error without the necessity to expand the corresponding task.
  • The Params section was extended with a complete list of parameters used in the request (instead of just the main ones). Herewith, all the sensitive data (passwords, tokens, SSH keys, etc.) is automatically masked for security reasons.
  • An internal actionkey parameter was removed from the tasks' response.
  • Automatic sorting of the fields in the Params and Response sections was implemented. Herewith, the result and error (if present) responses are always displayed first.
  • An ability to expand on-going tasks was added to get additional details on the action immediately after the start.

More info

Back to the top

Container Reboot via SSH

In the present 5.8.2 PaaS release, the ability to reboot a server via SSH without additional permissions was implemented. Such an operation may be required in case of some major issue on the container, which cannot be resolved via service restart, e.g. the jem service restart command hangs. For now, it is possible to reboot the whole operating system of the container via the following new commands:

1
reboot

or

1
/usr/sbin/reboot

With such an improvement, it’s no longer necessary to stop-start the whole environment so that only a problematic container will be affected by the reboot.

Back to the top

Isolation Timeout

Network Isolation is an important feature of the platform that allows managing groups of environments separately (by denying unintended connections via the internal network). Usually, all the isolation operations, such as group creation, nodes addition/removal, etc., are quick and need just a few seconds. However, in some cases (high platform load, network issues), these processes can require more time than the 30 seconds reserved by default. Thus, in the current 5.8.2 release, the timeout for isolation operations was increased to one minute.

More info

Back to the top

API Changes

Below, you can find a list of all changes to the public API in the 5.8.2 platform version (compared to the preceding 5.8 ones):

  • added two optional parameters for the stopEvent action to provide notification after the event interruption:
    • type - the notification type based on the specified value: success (the default behavior without notification), info/warning/error (pop-up style as described in the CS documentation)
    • message - a text that should be displayed in the pop-up (if not specified, the default notification is provided)
  • improved the domains parameter description for the AddDomains method
  • the actionkey parameter became optional for the ChangeTopology method

More info

Back to the top

Fixes Compatible with Prior Versions

Below, you can find the fixes that were implemented in PaaS 5.8.2 release and also integrated into previous platform versions by means of the appropriate patches.

PaaS 5.8.2
#Compatible fromDescription
JE-513273.3Installation of the Redmine package fails due to incorrect dependencies
JE-513813.3Installation of the PrestaShop package fails due to the usage of the outdated Apache PHP version
JE-518513.3The WordPress packages should install fake Let’s Encrypt certificates for the default environment domains
JE-521443.3Installation of the Drupal package fails due to the inability to extract from the .tar.gz archive format
JE-522103.3Installation of the Ghost package fails due to the usage of the outdated Node.js version
JE-523533.3Installation of the Jenkins package fails due to the usage of the outdated Tomcat version
JE-523593.3An installation of the Cyclos 4 PRO Cluster package fails due to usage of the incorrect Java version
JE-526103.3Old upstreams are not removed after changing backends in the Traffic Distributor package
JE-403505.3Firewall rules are absent on the application server nodes after the creation of the GlassFish/Payara cluster
JE-474095.3An error occurs while scaling out GlassFish Cluster with the deployed project
JE-400405.4Firewall rules are absent on the GlassFish/Payara cluster nodes if created in the extra layer
JE-436195.4The deployed projects are not available after scaling out the GlassFish/Payara/WildFly instance with the Auto-Clustering enabled
JE-492745.4An email about a DAS node should not be sent upon the creation of the standalone GlassFish instance
JE-528105.5New GlassFish/Payara instances are displayed with the “Stopped” status in the admin panel after the DAS node redeploy
JE-527225.7The “old value for weight not found” error occurs during the Traffic Distributor package installation
JE-521685.7.2Incorrect uid and gid permissions for the default jelastic user on the LiteSpeed ADC stack
JE-521975.7.2The minimum cloudlets requirement for the LLSMP stack should be increased to ensure operability
JE-524375.7.4The root login option is disabled in the sshd_config file of the Debian VPS instance

Back to the top

Software Stack Versions

The software stack provisioning process is independent of the platform release, which allows new software solutions to be delivered as soon as they are ready. Herewith, due to the necessity to adapt and test new stack versions, there is a small delay between software release by its respective upstream maintainer and integration into the platform.

The most accurate and up-to-date list of the certified software stack versions can be found on the dedicated documentation page.

More info

Back to the top

Bug Fixes

In the table below, you can see the list of bug fixes applied to the platform starting from PaaS 5.8.2 release:

PaaS 5.8.2
#Description
JE-44341The password placeholder doesn’t work in cases when node groups are used in the JPS manifest
JE-51305An error occurs when installing the Jenkins package from the Marketplace
JE-51956An error is returned for the successful Install API call
JE-51959If a node group in the load alert is edited, it should be displayed at a different layer immediately
JE-51965Environment status is stuck in the “Creating” state when installing from JPS
JE-51970Incorrect node tags provisions for environments installed from JPS
JE-51981Environment status is not updated for the Source > Environments Nodes list in the Firewall section
JE-51983Node icons in the Source column of the Firewall section are cropped
JE-51985Disabled firewall rules should not be automatically enabled after edit
JE-51997The Destination field is empty when editing existing outbound firewall rules with Environment Node as a destination
JE-52010An incorrect environment domain name is displayed for the collaborators on the reseller platform
JE-52018Text is aligned incorrectly in the “To container(s)” list when exporting folders from a container
JE-52021An additional environment in the “Installing” state is displayed in the dashboard when cloning an environment with a cluster
JE-52022During the Let’s Encrypt add-on installation, the environment status is updated after external IP attachment and not after the whole operation completion
JE-52023A package installation window is stuck on the “Configuring Environment” step if the action was called via API
JE-52029Incorrect hint upon hovering over nodes in the extra layer of the topology wizard
JE-52053Incorrect host IPs adjustment on the NGINX load balancer during the backends removal
JE-52067Long folder names are displayed incorrectly in the Exports section of the configuration file manager
JE-52161Environment status is stuck in the “Creating” state when creating a cluster or multiple environments
JE-52162A special icon for the low disk space on a node is displayed cropped in the dashboard
JE-52164The admin panel option should be displayed on the layer and not node level for the clusterized GlassFish/Payara templates
JE-52179An application server is automatically added when switching from Docker to another programming language during the existing environment topology edit
JE-52202An error occurs when restarting an NGINX load balancer with no backends
JE-52217The long environment name is not cropped if the dashboard was refreshed during creation
JE-52233An error occurs when restarting an NGINX load balancer after removing all of its backends
JE-52266The tag of the DAS node in the GlassFish/Payara cluster should be the same as for workers and not editable
JE-52269The ability to remove the DAS node should be disabled in the GlassFish/Payara cluster
JE-52364One of the IP addresses is stack in the “Attaching” state when IPv4 and IPv6 are added simultaneously
JE-52365The region icon is displayed incorrectly if the environment creation was finished while the Tasks panel was in fullscreen
JE-52405Master node CPU usage is displayed in the Usage pop-up of the layer (instead of the sum of all nodes)
JE-52406Topology wizard cannot be opened for environments with the legacy templates
JE-52450An error occurs when removing all application server nodes for the Payara cluster
JE-52510An error occurs when trying to mount an external server
JE-52515The Activity Log tab works incorrectly when multiple tasks are in progress simultaneously
JE-52554Incorrect collaboration invitation link for new users in the appropriate email
JE-52589Package installation during container redeploy fails due to the incomplete RPM metadata error
JE-52600The /var/tmp folder is absent on the Debian VPS nodes
JE-52665Incorrect “Log In Now” link in the email sent to new users after registration
JE-52666An error occurs while creating environments with clusterization option for nodes via JPS
JE-52709Errors with the SetEnvStatus API method should be correctly handled in the dashboard
JE-52752The Clone Environment button should be locked if the maximum number of environments is reached
JE-52783An error occurs while cloning the environment
JE-52857Environment and Add-Ons window are stuck in the “Refreshing” state if the add-on installation fails
JE-52865Environment status is stuck in the “Installing” state if the dashboard was refreshed during its creation (via Marketplace or Import)
JE-52871Environment status is stuck in the “Applying Settings” state if the topology was adjusted via CLI
JE-52875Topology changes via CLI should be immediately displayed on the dashboard after the operation is completed
JE-52881An error occurs when opening the Add-Ons section for the Tomcat node

Back to the top