Since we don't see any measure for some of the OOB alerts (For ex - Application process unavailable), i see no way to restrict it to get alerted on only specific agents/agent groups. I am wondering if anyone has achieved this by any other alternative way.
Thank you. This also won't be able to help. I have tried that. Plugin is using agent group filters but dynaTrace engine is not providing information about agent groups for some of the incidents.
Yes, the Extended Email Action Plugin works for the "Application Process Unavailable (unexpected)" incident. It takes information about the agent from the incident's message.
Let me know if you have further questions.
I am specifically talking about the "Application Process Unavaiable" incident not the "Application Process Unavaiable (Unexpected)". Last time when we looked at it, you have observed that dynatrace engine is not providing information about agent groups hence we cannot get agent specific alerts. if anything changed about this please let me know.
There is no OOTB "Application Process Unavailable" incident. There is "Application Process Unavailable (unexpected)" OOTB incident. See highlighted incident in the attached screenshot here. The Extended Email Action Plugin takes agent name and agent host from the message of the incident for the "Application Process Unavailable (unexpected)" incident.
There are other OOTB incidents, e.g. "Corrupted PurePath(s)", etc. where dynaTrace engine does not pass information about source of the incident to the plugin. To check if you are getting information about source of the incident please setup the Extended Email Action Plugin for this type of incidents, trigger it and check if agent-name and agent-host are available.
Thanks Eugene. i was thinking "Application Process Unavaialble" as OOTB alert. i haven't created this. So i was thinking this is OOTB alert. is there a flag in the config which tells us if the incident is OOTB or user created? i don't see any measure also in this. Basically i want to get the agent specific alert for this.
You have mentioned -
" To check if you are getting information about source of the incident please setup the Extended Email Action Plugin for this type of incidents, trigger it and check if agent-name and agent-host are available."
Where do i check this? in the Logs to see if agent-host info is available or not?
The "Application Process Unavailable (unexpected)" is the OOTB incident. There is no flag to indicate if incident is OOTB incident or custom created incident. OOTB incidents are always exist for any system profile. They are listed in the Incident Rules column of the Incident Rules for <system-profile> window (see example here). There is no measure to setup for the "Application Process Unavailable (unexpected)" incident. dynaTrace engine will trigger this incident when connection to a previously connected Application Process/Agent has been lost.
To get an alert for this incident for a specific agent, please setup the Extended Email Action Plugin and put in the Agent filter agent's name (or regular expression for the agent's name). In this case notification will be send only if incident is coming from the specific agent.
And lastly, how to check if the Extended Email Action Plugin is getting source information about the incident (i.e. agent name, agent host, etc.): cut and paste attached BodyHeaderExample.txt file into the body header plugin’s parameter field. Make sure that "HTML Mail Format" parameter is set to "true" (i.e. checked). Notification e-mail will contain list of the maintained by the plugin runtime variables including agent name, agent host, etc.
Please let me know if you have any further questions.
Question that was posted here was resolved back in May through direct communication. To help you I need to get details about issue that you are having. Please post them here or send me direct e-mail.
The Extended Mail Action plugin log is located in the <dT-home>/log/server directory. Name of the latest log file is "com.dynatrace.diagnostics.plugins.extendedmailreport.action.0.0.log".
Please upgrade to the latest version of the plugin if you haven't done it already.
I have a similar case where I want to use the OOTB incidents such as Failure Rate Too High or Response Time Degraded but specify different email distributions for specific applications. The system profile in question has multiple applications in one. These OOTB incidents have no measure to use and aren't able to be duplicated. The client likes these OOTB incidents/measures and the base-lining it uses. Ideally what i'm looking to do is use these OOTB Incidents that utilize the base-lining but propagate alerts based on application, as there are more than one. Also the extended email action plugin does not allow for alert filtering based on application...
Any thoughts at how to approach this?
The Extended Mail Action plugin version 0.9.9.13+ allows to couple source of the incident with the e-mail recipients for multiple filters. If you use agents or agent groups as filters it should allow you to send notification e-mail based on agent name or agent group name. While we do not filter incidents by application name, filtering by source of the incident, i.e. agent name, agent group name, monitor name, etc. should work for you.
There is a known deficiency in the dynaTrace plugin interface design which may prevent plugin from getting source of the incident information. For example, please see the following two posts:
You need to check if incidents you are working with will provide incident's source for the plugin.
Thank you for the reply Eugene.
I do in fact need to either be able to filter alerts on application OR somehow duplicate the OOTB incidents that utilize the smart base-lining feature and create an incident for each application within the single system profile that will alert if it deviates a certain percentage from that baseline. My attempts have not been successful as of yet...
More info regarding this scenario...This is one system profile with multiple applications within. The reasoning for this is because this system profile includes the client's IBM enterprise Portal Application and those applications that sit behind/authenticate through their Enterprise portal. Anytime an incident is triggered for degraded response times of failure rates based on the base-lining for the Web Page Requests BT (the OOTB monitoring dashboard...), it can only be sent to one email distribution, ideally we want to be able to differentiate each application and alert to the proper team. I agree that in normal circumstances the agent/agent group filter would work but in this case all entry points are from the Enterprise Portal Application agents... So with that I need to either filter the email action or create BTs for each Application that will alert based on the baseline.
I followed this case's (https://na5.salesforce.com/5007000000ZVtUw?lt%3Bhttps%3A%2F%2Fna5.salesforce.com%2F5007000000ZVtUw%3...) suggestions but when I try to create an incident with the URI path filter and web request response time filter, it states I need to define a threshold for the web request time measure... I want it to use the baseline...