Created 04-15-2016 04:43 PM
I ran into a number of operating system issues with the HDPCA practice environment (Hortonworks HDPCAdministrator_2.3 PracticeExam_v3- ami-01b8066a) in AWS that has me concerned. Are the following issues part of the exam, or part of an issue with the environment?
When adding Node1 to the cluster NTPD would not start, but I also saw that is wasn't running on the NameNode. Then, iptables failed to stop on Node1, threw an error. The manual registration of Node1 succeeds, but host registration fails due to these two issues.
Continuing with the DataNode, NodeManager and Client Services install, the process hangs on the DataNode install, then finally times out because the installation of BASE packages rpcbind, libtirpc and libgsglue failed due to no internet access to download these packages.
I find this whole scenario out of character for an exam based on the published objectives. Is the exam based on Hortonworks HDPCA administration or CentOS administration? Any help or guidance to better understand this exam and its objectives is greatly appreciated.
Thank you
Jim
Created 04-18-2016 06:12 PM
It is a dockerized multi-node environment. Firewall and NTP issues cause warnings during the node registration, which can be safely ignored. Because all nodes are virtualized inside the VM itself using Docker so, you do not need to address Firewall issue. Same for NTP, all nodes are getting time from the host OS, so they will be always in sync. They should not cause a fatal error. Please provide log for the registration-failure issue. So that we could pin-point the actual error.
The issue could be related to AWS->security group. Did you open required ports in AWS->security group so that the VM could download HDP and Ambari repositories from web during the node-registration and HDP installation process.
Can you open all the ports for the time-being and try again.
Created 04-18-2016 06:12 PM
It is a dockerized multi-node environment. Firewall and NTP issues cause warnings during the node registration, which can be safely ignored. Because all nodes are virtualized inside the VM itself using Docker so, you do not need to address Firewall issue. Same for NTP, all nodes are getting time from the host OS, so they will be always in sync. They should not cause a fatal error. Please provide log for the registration-failure issue. So that we could pin-point the actual error.
The issue could be related to AWS->security group. Did you open required ports in AWS->security group so that the VM could download HDP and Ambari repositories from web during the node-registration and HDP installation process.
Can you open all the ports for the time-being and try again.
Created 04-19-2016 01:03 PM
Manish - the AWS security group clue seems to be the answer. I was using a specific security group for our corporation. I started another instance on my personal AWS account and it is responding as expected with focus on the HDPCA tasks. Thank you for that recommendation. Jim
Created 04-19-2016 06:17 PM
Awesome.
Please select this answer as the best answer, so that this thread will be marked as resolved. and if you could vote my answer 🙂
Created 02-13-2017 04:47 PM
Good Day , I got not found the exam practice hdpca . Which is way ?
,
I got not found the exam practice ? Could send the way ?
Created 12-18-2017 09:38 AM
The practice exam AMI is available in Oregon region. Please switch to Oregon region and try adding instance. Follow the practice exam PDF and follow the configuration details.