BTM agent relationship with Hawkular agent
by Gary Brown
Hi
On the team call just now, Heiko raised a question about whether the BTM agent could work with the Hawkular agent. So thought I would start this discussion thread to see what the potential options are.
I currently see two issues:
1) The BTM agent must be configured on the jvm command line as a "-javaagent" to install ByteMan for instrumentation purposes. This is instantiated before the JBoss modules (and therefore subsystems etc) are initialised.
2) The hawkular agent won't necessarily be installed in all monitored servers, and instead remotely monitor some. The BTM agent would need to be installed in all servers where business transactions are executing.
One type of integration that may be possible is in terms of delivering the captured business transaction information to the backend? i.e. the BTM agent locally reports it to the Hawkular agent as a relay?
Thoughts?
Regards
Gary
1 week, 1 day
some new agent features
by John Mazzitelli
Just wanted to blast out a message about some new wildfly agent features.
1) Agent now supports secure communications between itself and any of its managed WildFly servers. I documented how to enable security in the management interface of WildFly here:
https://github.com/hawkular/hawkular.github.io/blob/pages/src/main/jbake/...
2) Agent now supports monitoring JMX resources via Jolokia interface.
The agent now can collect JMX metrics and availability data and can inventory JMX resources in the same manner as it can with DMR resources using the Jolokia interface (Jolokia is a very nice REST interface to JMX). Just as we have <metric-dmr>, <avail-dmr>, <resource-type-dmr>, <remote-dmr>, and related DMR settings in the agent subsystem, we now have <metric-jmx>, <avail-jmx>, <resource-type-jmx>, <remote-jmx> settings as well. So the agent can be given metadata that tells it what to do. As with remote WF servers, the agent also supports secure comm to the remote jolokia JMX server.
There are a few things currently on the TODO for the agent:
1) utilize the new inventory "name" attribute - they now have "name" as a first-class setting - I need to see what the agent has to do to use it
2) utilize the inventory's "operation" metadata - I believe operations are now first class concepts in inventory - I need to figure out how to populate it
3) provide a mechanism for the agent to detect runtime changes to inventory. Right now, after the agent does its initial inventory scan, that's it. The inventory is fixed, all resource configuration is fixed, and all we do is collect metrics. The caveat is that if someone deploys a new app or adds a new JDBC driver, we force a new scan. We need a way to detect new resources in inventory (and old resources removed from inventory) and we need a way to detect changes to resource config. This should work for both DMR and JMX inventory items.
4) Need a way to consume wildfly notifications that other subsystems emit and figure out how to send that to a hawkualr events subsystem.
5) Peter is implementing/has implemented the JDBC driver CRUD and datasource CRUD, gonna see how easy it is to implement reload.
I'm sure I'm forgetting something, but .. that's what I remember.
4 weeks
Corrections and suggestions on Explorer for android client
by Anuj Garg
Hello everyone,
A tree structure of the explorer is approx ready for the android client and
I was curious if i am wrong somewhere.
First of all, All feeds are collected by call to
/inventory/feeds
feed call for resources
/inventory/feeds/{feed}/resources
call for rec resources and metrics
/inventory/feeds/{feed}/resources/{resid}/recursiveChildren
/inventory/{feed}/resources/{resid}/metrics
Am I missing something?
Thanks
Anuj Garg
1 month
some new hawkular openshift agent stuff
by John Mazzitelli
FYI: some new things went into HOSA (that's the Hawkular OpenShift Agent for the uninitiated).
1. The agent now emits its own metrics and can monitor itself. Right now it just emits some basic "go" metrics like memory usage, CPU usage, etc along with one agent-specific one - a counter that counts the number of data points it has collected in its lifetime. We'll add more metrics as we figure out the things people want to see, but we have the infrastructure in place now.
2. The agent is deployed as a daemonset. This means as new nodes are brought online, an agent will go along with it (or so I'm told :)
3. The agent has changed the way it discovers what to monitor - it no longer looks at annotations on pods to determine where the configmaps are for those pods. Instead, it looks up volume declarations to see if there is an agent configmap defined. This was done to be ready for the future when new security constraints will be introduced in OpenShift which would have broken our annotation approach. This approach using volumes should not hit that issue.
NOTE: If you are building the latest agent from master, we added some dependencies so you have to update your dependencies via Glide by using the "make update-deps" target prior to building from source.
1 month, 2 weeks
Inventory API question
by Austin Kuo
I was creating 2 different resource types with the same http client at the
same time.
But one succeed, the other failed with the response 400 and body:
{
"errorMsg" : "The transaction has already been closed"
}
Is it not allowed to do so?
Austin.
2 months, 1 week
Hawkular Metrics 0.13.0 - Release
by Stefan Negrea
Hello Everybody,
I am happy to announce release 0.13.0 of Hawkular Metrics. This release is anchored by Hawkular integration enhancements, under-the-cover refactorings and fixes, and a new bulk data generation tool.
Here is a list of major changes in this release:
1) Data Generation Tool
* A new tool that generates bulk metrics data to be used in performance and load testing.
* It generates directly Cassandra data files (SSTables), which leads to a very fast generation process for large amounts of metrics data.
* For more details: https://github.com/hawkular/hawkular-metrics/tree/master/data-generator (HWKMETRICS-355)
2) Receive Metrics via Hawkular Bus
* When deployed within Hawkular distribution, the project now accepts metrics via the Hawkular Bus; until now only the REST API had support for Metrics insertion.
* Hawkular Metrics currently support publishing of newly inserted metrics to the bus and receiving metrics via the bus.
* For more details: HWKMETRICS-347, HWKMETRICS-352
3) Sorted Stacked Metrics Results
* When requesting stacked metrics aggregation the results are now ordered (HWKMETRICS-353)
4) External Integrations
* Heapster sink now divides storing to multiple calls. This is an improvement over the initial implementation that had one REST API call per metric (HWKMETRICS-290)
* ptrans now works with a Hawkular Metrics instance protected by Hawkular Accounts via Hawkular distribution (HWKMETRICS-342)
* Grafana integration via Influxdb compatible end-point allows connections to a Hawkular Metrics instance protected by Hawkular Accounts via Hawkular distribution (HWKMETRICS-343)
Github Release:
https://github.com/hawkular/hawkular-metrics/releases/tag/0.13.0
JBoss Nexus Maven artifacts:
http://origin-repository.jboss.org/nexus/content/repositories/public/org/...
Jira release tracker:
https://issues.jboss.org/browse/HWKMETRICS/fixforversion/12329530
Hawkular Metrics Clients
* Ruby: https://github.com/hawkular/hawkular-client-ruby
* Python: https://github.com/hawkular/hawkular-client-python
* Go: https://github.com/hawkular/hawkular-client-go
A big "Thank you" goes to John Sanda, Thomas Segismont, Mike Thompson, Matt Wringe, Michael Burman, and Heiko Rupp for their project contributions.
Thank you,
Stefan Negrea
Software Engineer
5 months, 2 weeks
