Led by Ron Clapman.
- Ron sees many customer sites, post-sales, i.e. in production. Which means he sees many different solutions to problems, good and bad.
- Doing so, Ron had an epiphany for DevOps and monitoring of our products.
- The following product mentions are not meant as endorsements of one over the other!
- Splunk is used by most of ForgeRock customers: It's easy to install and get up and running. (e.g. their agent).Very easy and quick to find causes of issues.
- A Splunk handler for the ForgeRock Common Audit Framework (CAUD) is coming!
- One attendee's company is using the ELK stack for troubleshooting.But alerting is important. Monitoring is important. Troubleshooting is important.--> And they are all different, and different tools are being used right now.
- One attendee mentioned:
- for critical events there should be a guarantees, but:
- It’s quite common (and ok) to log data to two destinations (e.g. file system and Splunk/ELK).
- Question is how to then get stored data into Splunk after it came up again (consistency?).
- One question was also about Java specific monitoring: JMX is there, but not many people seem to use it. Useful but not easy to set up. Works fine locally, but hard to do remote.
- Other tools some of customers are using are Datadog and New Relic. Both are cloud based, but agents need to be installed locally.
- Other tools mentioned by attendees are getlin.io, Sensu, Prometheus.