See this article for Fixed Bugs and Changes in Vortex Insight.
Version 2.1.3
| Report ID. | Description |
|---|---|
| THM-1322 | Insight connector feature requires the msvcr100.dll on Windows machines. On Windows machines, if the user wants to use the local insight connector, the msvcr100.dll is required. Connecting locally using cmsoap requires this dll. |
| THM-1319 | Vortex Cloud and TCP configuration not displaying DDS entities. Insight no longer gets notified about cafe entities when ddsi.network.udp.enabled=false. (TCP) |
| THM-1218 | QoS Settings tab for publishers and subscribers does not display Presentation policy settings. QoS Settings tab for publishers and subscribers does not display Presentation policy settings if the publisher or subscriber does not have writer or reader children. |
| THM-1214 | Publishers and Subscribers with no child writers or readers are not rendered in the browser tree after a refresh. Initially when a Cafe application starts, the corresponding participant, publisher and subscriber entities are displayed in the domain tree browser. On Refresh (F5), the Cafe publishers and subscribers with no children are not displayed in the tree. |
| THM-1172 | Vortex Lite readers/writers show compatibility when they are incompatible from partition. An application running Vortex Lite creates a publisher with writer with a partition QoS policy set, and another application creates a subscriber with reader on the same topic, but with a different partition QoS policy. In Insight's QoS comparison view, the comparison between the reader and the writer shows green when it should be red. Additionally, the partition table row is not present in the comparison table. |
| THM-1171 | The partition for a given publisher/subscriber is not displayed in the domain tree browser. In quite a few cases, the partition for a subscriber/publisher is displayed as "DEFAULT_PARTITION" (meaning an empty partition name) in the domain browser tree when it is not. The primary cause of this is that the low-level samples collected by the Insight server (which describes the subscriber/publisher entity) would contain inconsistencies in their partition values. Cases where the issue has been observed are Vortex Café/Web as well as Vortex Tester/Tuner subscribers and publishers. |
| THM-1170 | Starting and stopping multiple applications writing to the same topic for which threshold monitoring is enabled causes an exception If a user starts and stops multiple applications writing to the same topic for which threshold monitoring is enabled in Vortex Insight, it is possible for a ConcurrentModificationException to be thrown in the console. The issue has no impact on the monitored values recorded by Vortex Insight in the threshold monitoring graph. However, left-over participants of the disposed applications are not removed from the threshold monitoring tree browser. Workaround: Refreshing the page cleans up the left-over participants from the threshold monitoring tree browser |
| THM-1169 | Screenshots in the PDF guides not showing up at the right places In the VortexInsightUserGuide.pdf and the VortexInsightInstallationGuide.pdf, some screenshots and images show in the wrong sections. Because of this, certain aspects of the documentation might be unclear. Workaround: Use the HTML versions of the user guide and the installation guide. active |
| THM-1167 | Vortex Insight doesn't redirect to the connect page when connection to a stopped Vortex OpenSplice system is lost When a Vortex OpenSplice system that Vortex Insight is connected to shuts down, an explanation pop-up shows up indicating that the connection is lost. However, after the user acknowledges the pop-up, Vortex Insight does not redirect to the connection page. Instead, the browsing page remains active suggesting that the connection is still alive. Workaround: Simply disconnect Vortex Insight using the disconnect button located at the top-right of the page. |
| THM-1166 | Federations are not initially recognized when the default "no_internals" filter is active When connecting Vortex Insight to a shared-memory Vortex OpenSplice deployment with the default "no_internals" filter enabled on the domain tree browser, the "Federation" node is not displayed even if it contains visible unfiltered applications. Instead, these unfiltered applications are shown as if they were single process deployments. Workaround: Simply disabling, then re-enabling the "no_internals" filter forces the federation to be properly displayed with its unfiltered children applications underneath it. |
| THM-1037 | Insight's CPU info gauge indicator does not update and indefinitely displays a value of "zero" if Insight is run using OpenJDK 1.7 If Insight is run using OpenJDK 1.7, it is possible that, when looking at node monitoring statistics, the CPU info gauge indicating CPU load of the viewed node does not update. Workaround: Use Oracle's Java (JRE or JDK 1.7 and up) as OpenJDK is not officially supported. |
| THM-126 | If the Vortex Insight server application crashes, the participant named VortexInsight is not cleaned up from the OSPL DDS system. Currently, if Insight process is abruptly stopped, for example by killing the process (using "taskkill" command on windows or "kill" command on linux), it will not be able to de-register its participants from the OSPL DDS system. This means that if this situation happens multiple times, the OSPL DDS system will experience a build-up of left-over Insight participants in the system. This is not desirable because to remove the left-over participants, the OSPL DDS system has to be restarted. Workaround: To avoid this issue all-together, it is recommended to properly disconnect Insight from the OSPL DDS system before shutting it down. This can easily be done by using the "disconnect" button available at the top-right of Insight main page in your browser. |
| THM-591 | Java version 1.8: Insight crashes when opening browser page. [error] play - Error loading fake key store java.security.cert.CertificateException: Subject class type invalid. It is possible when running Vortex Insight for the first time using Java version 1.8 and opening the browser page that Insight crashes with java.security.cert.CertificateException: Subject class type invalid. Workaround: Trying to restart Insight might fix the issue. Otherwise, use Oracle Java 1.7 or manually generate your SSH key. If all the previous suggestions are unsuccessful, another work-around will be to temporarily use "http" connection instead of "https". For information, see Insight documentation about these features. |
| THM-336 | Configuration/Monitor : Having more than one monitor with the same participant name is disallowed. It is currently not possible to create two or more threshold monitor configurations for two participants of the same name even if they are fundamentally different entities (e.g. they have different ids). Note that for Java applications, this problem does not occur since the participant name always includes the process id of the java application. Workaround: A possible work-around is to provide a unique name to your participant so as to avoid any name clashing. |
| THM-464 | Unable to operate more than one client tabs or browsers using Internet Explorer. If the user opens two clients on Internet Explorer browser to connect to Insight server, only the first client works properly. On the second client, the page loads but does not update nor does it react to user clicks. This is the same for any subsequent IE browser clients started at that point. Workaround: Adjust Internet Options settings. In InternetOptions/Security/Local intranet/Sites : uncheck all the sites. |
| THM-591 | Java version 1.8: Insight crashes when opening browser page. [error] play - Error loading fake key store java.security.cert.CertificateException: Subject class type invalid. It is possible when running Vortex Insight for the first time using Java version 1.8 and opening the browser page that Insight crashes with java.security.cert.CertificateException: Subject class type invalid. Workaround: Trying to restart Insight might fix the issue. Otherwise, use Oracle Java 1.7 or manually generate your SSH key. If all the previous suggestions are unsuccessful, another work-around will be to temporarily use "http" connection instead of "https". For information, see Insight documentation about these features. |
| THM-839 | 2 Insight servers cannot connect to Cloud in TCP mode from the same machine [error] cafe.ddsi.net - Unable to open socket 0.0.0.0/0.0.0.0:7800 ! Try to configure another port using ddsi.discovery.tcp.port system property. In order to connect and interact with a Vortex Cloud deployment, Insight needs to acquire a local port. Currently, that port number is static (portNumber=7800). As such, it is not possible for more than one Insight server running on the same machine to connect to a Cloud deployment. |
| THM-509 | Statistics timeline on smaller web client windows appears on items that are not statistics related. On small-size web clients, the statistics timeline displays incorrectly; e.g. it shows on top of non-statistics data This improper behavior is caused by the fact that the statistics timeline currently has a fixed position relative to the browser window. Workaround: Gradually resize the window to a larger size until the timeline starts displaying properly. |
| THM-705 | Monitoring view: On a narrow browser screen, the screen keeps jumping up to the tree view. On small-size web clients, if the monitoring tree focus in on an entity located at the top of the tree and the user scrolls the page down to view other information, the screen keeps jumping back up to the focused element in the tree; thus preventing the user from browsing other contents. Workaround: Gradually resize the window until all information in the page can be seen without having to scroll away from the focus node in the tree. Alternatively, temporarily click on a node located at a position in the tree that is still visible. |
| THM-751 | Calling Ctrl-C on an insight server monitoring a DDS sometimes causes an exception. Calling Ctrl-C on a running Insight server connected to a DDS domain causes an exception that reads "org.h2.jdbc.JdbcSQLException: Database is already closed". This happens because some internal Insight processes attempt to access the database when it has already been closed upon disconnect. This error has no impact on actual operations and can be ignored. |
| THM-862 | Cloud domain tree freezes when attempting to process large entity updates (> 100 entities). When Insight is connected to a Vortex Cloud domain, if a high number (>100) of cloud entities are created or dispose at once, the cloud browser tree freezes for a while, failing to update the displayed topology on time. During that period, the tree is effectively unresponsive to user actions. |
| THM-859 | Insight-Cloud is not using proper participant names when displaying cloud participants in the tree view. Insight currently displays internally generated generic naming to identify cloud participants. This can make it hard to recognize these entities by just looking at their names. |
| THM-1013 | No way to reliably determine if multiple Cafe participants belong to a single application. If user starts a single Cafe application containing multiple participants and connect Insight to it. Insight represents each participant as a separate Cafe application which is an incorrect representation. This is due to how Insight currently infers topology from data about the Cafe application. |
| THM-1018 | Change rate graph data does not display if potential out-of-sync states between client statistics js and cubism graph engine occur. It is currently possible (although rare) for the change rate graph to not display at all if the client statistics and cubism graph engine get out-of-sync. The client statistics and the cubism graph engine both operates asynchronously and are only bound by both following current time. Workaround: Disconnect and re-connect Insight to the monitored target system. |
| THM-1034 | Can't start Vortex Insight, staying "Loading..." If starting Vortex Insight using OpenJDK and connecting to an OSPL target, it is possible for the domain tree to never load and simply show the "Loading..." message along the spinning animation. Workaround: Make sure you are using Oracle JDK 1.7 and not OpenJDK as the later is not supported |
| THM-1078 | NPE in BuiltinTopicReader (intermittent) Sometimes, a null pointer error occurs in the BuiltinTopicReader. This happens when the reader fails to retrieve the corresponding cm object of a discovered entity while it expects it. The error shows in Insight's console. Workaround: Disconnect and re-connect Insight from the monitored target. |
| THM-1081 | NPE in ThresholdMonitorActor when disposing a monitored participant with a high number of writers (intermittent) When monitoring a participant's activity using the threshold monitor graph, it is possible that an error occurs when that participant becomes disposed if it had a high number of active writers (more than 25). Workaround: Simply navigate to the configuration, make a change to a threshold monitor configuration and save. This forces the threshold monitoring system to restart. |
| THM-1085 | The Insight dds connector requires the msvcr100.dll in order to run. The msvcr100.dll includes the Microsoft Visual C++ Redistributable. This dll is included in the Java JRE bin directory. Workaround: Vortex Insight requires that the Oracle JRE is installed. Ensure that the JAVA_HOME/bin directory is included on the PATH. |
| THM-1100 | ProcessInfo tab does not always show up for non-service processes other nodemon process. Currently, the "Process Info" tab showing process monitoring information for a given application (part of the node monitoring feature) only shows up for the nodeinfo application itself. It does not always show for other applications as it should be the case. This maybe because the PIDs of theses applications are not being registered to NodeInfo service in VortexOpenSplice. |
| THM-1104 | If viewing node monitor statistics, the updates will stop indefinitely if NodeStatsActor crashes This issue happens because the NodeStatActor currently does not recover from crashes. Workaround:Simply disconnect and re-connect Insight to the monitored target system. |
