Member since
12-14-2015
12
Posts
11
Kudos Received
2
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
6209 | 03-02-2016 04:01 PM | |
7054 | 02-26-2016 01:22 PM |
05-30-2017
06:45 PM
Hi Matt, I tried exactly what you suggested while I was waiting for your reply. I was able to access the UI without the data flow running. I looked at the System Diagnostics you mentioned earlier. It was "4 times" without data flow running and still "8 times" now after I started data flows. Our cluster has 2-node and each has 16 cores. The "max timer driven thread count" is set to 64 and "max event driven thread count" is set to 12. I've been monitoring "top", cpu usage at this time (busy hour) is about 700%. Good news is after I started nifi in stopped state and manually restarted the data flows, the problem I had this morning has not recurred yet. Heartbeats are generated in reasonable intervals - about 7 seconds. What happened this morning is still a mystry to me, but I am happy now it's working. Thank you so much for all the help!!! Xi
... View more
01-16-2018
07:45 PM
Hey @Montrial Harrell, We are facing this once or twice a week and we use to solve this issue by restarting the hive server. Can you let me know what and where did you increase the limit to clear this issue.
... View more
02-27-2016
05:03 PM
Only apply patches if necessary and instructed by support. In case you don't have a support contract, here's Pivotal instructions to patch Ambari, we don't provide steps due to the reasons above. http://hawq.docs.pivotal.io/docs-hawq/topics/hdp-prerequisites.html Needless to say its at your own risk.
... View more