Issues tables

Symptom: The Reporter system task displays a higher number of failed Helix Dashboard reports than the actual failed reports.

Scope: When running a large number of reports (for example, 600-700), and each report size is greater than 11 MB. 

In BMC Helix Continuous Optimization deployments on AWS, the Remote ETL Engine (REE) cannot be downloaded directly because of AWS file size limits.

Workaround: To install the REE, download the installation files from the BMC Electronic Product Distribution (EPD) website.

Symptom: The Entity cleaner task causes the service optdataorg to fail with the out of memory error. 

Scope: When you have a large number of entities to be deleted. 

Workspace

Open

DR0AV-15209

Capacity Views

Open

DR0AV-16015

Benchmark

Open

DR0AV-16771

The Capacity Agent does not support systems where SELinux is configured in enforcing mode. The service daemon process (bgssd.exe) may fail to run when started through systemd.

Workaround:

1. Change SELinux to permissive mode.
2. Start the service daemon manually: /etc/bgs/SD/bgssd.exe -d /etc/bgs/SD -s.
3. (Advanced – allow systemctl start under enforcing mode):

   1. Run as root:

      • semanage fcontext -a -t usr_t \readlink /usr/adm/best1_<version>`/bgs/bin/bgssd.exe`
      • restorecon -R -v \readlink /usr/adm/best1_<version>`/bgs/bin/bgssd.exe`
   2. Verify security context: ls -Z \readlink /usr/adm/best1_<version>`/bgs/bin/bgssd.exe`
   3. Start the service: systemctl start bgssd

Users assigned to the Capacity Planner role couldn't set metric thresholds because of permission errors, even when they had the necessary permissions.

Extrapolation models sometimes returned the same output values for different systems, regardless of their configuration. 

ETL runs finished with a warning status due to recurring gzip warnings for already compressed .vis files when archive compression was enabled.

When a scheduler task was deleted before migration, the migration process inserted an empty value for the missing scheduler ID, causing data mart creation errors.

Email notifications were not sent for ETL failures, even though diagnostic alerts were generated, which affected timely awareness of processing problems.

Kubernetes  PODs linked to a namespace without a deployment were not shown in the Kubernetes view, leading to an incomplete display of container workloads.

The Kubernetes Pod Recommendation view did not show workloads directly linked to a namespace without a controller, leading to incomplete visibility of overallocated pods.

Tag values in the Add Threshold dialog box were not sorted alphabetically, making it difficult to locate tags in long lists.

Custom scheduling settings for datamart materializer tasks were reset to default when out-of-the-box views were redeployed, causing the loss of user-defined execution times.

Charts failed to load at the monthly resolution when actual data overlapped with externally estimated values imported through CSV, causing analysis views to crash.

Activity dates were incorrectly cleared in the entity lookup screen when multiple providers monitored the same entity, resulting in missing or inconsistent update timestamps.

Systems under domains with multiple root paths displayed only a single breadcrumb trail, resulting in an incomplete navigation context in the domain hierarchy.

ETL

DR0AV-21763

Entity filter options that required a container domain were shown during Optimizer Rule creation, even though Optimizer Rules do not support container domains, leading to errors in rule creation.

The Data Warehouse Status Volume values varied based on the selected time zone, leading to discrepancies in metrics like Processed today, Processed yesterday, and Average daily processed rows when using non-UTC time zones.

When saving a Forecast Saturation event in the Events interface, the classification and description were changed to Detected event instead of Forecasted Saturation because the Detected event option was selected by default and could not be changed during editing.

In the Overallocated Containers page of the Kubernetes View, the recommended CPU and memory limits were lower than the recommended CPU and memory requests for certain containers.

In the Grid – change number of servers what-if event, you could only select one server at a time, which limited modeling changes across multiple systems.

The Time Forecasting Model option remained stuck in a loading state on the Views page when a domain with a large number of child systems was selected from the Golden model results list.

The Time filter setting was ignored for external data in Analysis. The data was displayed for the entire available time range instead of the selected time range.

Creating a forecast model for disks monitored by Dynatrace resulted in a 502 nginx error due to timeouts when handling many entities with duplicate sub-resources.