- Elastic releases new minor versions crazy fast, be prepared for things in a lab or two not working as expected
- Warn students about the potential setbacks ahead of time
- Use the setbacks as an opportunity to educate on how to recover using Elastic Search documentation and a Google search
- Do not rush to answer questions, steer to the guidance on what keywords you would have used to find the answer to get the students to self-sufficiency early
- Do not jump into resolving an issue with an installation, point to where to look for the problems: linux logs, _cat end-points, documentation
- The overall idea is not to demo instructor's knowledge, but to instill confidence that students will run into trouble and will find a way to recover on their own
- Launch Ubuntu Server 16.04 LTS on Aws VM for each student
- Instance type I have been using: t2.medium as ElasticSearch by default requires 4GB of RAM
- Configure access to the instances with a key-pair and send students the public and private keys
- Send student both keys ahead of time
- Students may have troubles accessing the VM's with their laptops using, as the last resort I have been using codeanywhere.com for the students to connect to the VM from a remote container
- For the students VM's I configure Elastic IP's and shut down the VM's for the night to reduce the charges
- X-Pack exercise add instructions for JMeter or see another note in this list
- Watcher restart FileBeat
- Extract X-Pack to its own chapter and add Watcher?
- Remove Vlad's aspect of the course?
- Do not add public key and repository to the list in for every product
- Capacity planning - turn into stress test and monitor exercise?
- Split Timelion into its own chapter?
- Add graph chapter