EA Collector 36.400

LogicMonitor has added the !perf debug command to check the collector performance of CPU, queue, and Datasources. You can use type=cpu, type=queue, and type=ds to get the performance analysis. You can apply the filter dsFilterColumn=timeout on the Timeout column for Datasource type. Additional filter options are nan, queue, dropped, and avg_exec. For example, you can run the following debug commands:

• !perf type=cpu
• !perf type=ds dsFilterColumn=timeout
• !perf type=ds dsFilterColumn=nan ds=*DS_TIMEOUT
• !perf type=ds dsFilterColumn=avg_exec ds=*DS_TIMEOUT h=127.0.0.1

To ensure that configurations are properly applied on custom size collector, LogicMonitor has made the following enhancements:

• For any newly updated or added property in the predefined configuration, based on collector size detection as per the max memory use, the property value will now be reflected in the custom size collector.
• The highest property value between the default value or the customer set value is applied. To apply this new value, you must update the collector to EA Collector 36.400 and then restart it.

For example,
For custom size collectors, if LogicMonitor increases the default value of a predefined property powershell.spse.queue.size from 50 to 60 then after upgrading to EA 36.400 and restarting the collector, the highest value 60 replaces the old value 50. If the default value (60) is less than the value set by the customer (75), then the value set by the customer is used.

In another example, when LogicMonitor introduces a new property, it will be automatically reflected in the custom size collector configuration after upgrading to EA 36.400 and restarting the collector.

When Syslog and SNMP Traps are processed through the agent.conf settings that is, when lmlogs.syslog.enabled=true and lmlogs.snmptrap.enabled=true in the agent.conf settings, the resource attributes will be added to ingested Syslog and SNMP trap based on the following resource mappings. The same is used for resource mappings with the OR operator in the same order with the resource mapping prioritisation applied.

To improve the SNMP Trap LogSource, LogicMonitor has now added varbind value based filter to the SNMP trap LogSource. Also, multiple filters will now be evaluated using the logical OR, instead of logical AND.

Fixed an issue where upgrading from GD Collector 35.001 to GD Collector 36.000 led to CPU and memory spike for the SPSE process.

Fixed an issue where the ESXi devices intermittently failed to collect data and resumed data collection after the collector restarts.

Fixed an issue where as a result of enabling the vault.url.cache.enabled property, multiple vault credentials accessed the same URL multiple times to cache the token. This led to delay in starting collector, monitoring devices, and generated several alerts. To fix the issue, the vault.url.cache.enabled property is now set to True by default.

Fixed an issue related to OID translation for varbinds that referenced to a specific index.