Member since
01-21-2016
14
Posts
0
Kudos Received
1
Solution
My Accepted Solutions
Title | Views | Posted |
---|---|---|
5162 | 03-14-2016 11:27 AM |
03-21-2017
01:13 AM
Thanks! I will put this in the cluster config and try: ... YARN { yarn_service_config_safety_valve: """ <property> <name>yarn.nodemanager.log-aggregation.roll-monitoring-interval-seconds</name> <value>3600</value> </property> """ } ...
... View more
02-24-2017
11:32 AM
makes sense. thanks a lot!
... View more
03-14-2016
11:27 AM
Thanks for your help! It finally works now! Turns out that "normalizeInstance" has to set to true. It's a bit confusing though as in the comment it says the following and I thought most of them are already done in the preloaded AMI. # Normalization includes: # downloading and installing packages # minimizing swappiness # increasing the maximun number of open files # mounting ephemeral disks # resizing the root partition. I will modify the packer scripts to download additional parcels we need. It will be great if you add support semi-preloeaded AMI so small parcels can be downloaded during provisioning.
... View more