Member since
12-10-2015
6
Posts
2
Kudos Received
2
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
23804 | 12-10-2015 09:53 AM | |
1277 | 12-10-2015 09:18 AM |
12-10-2015
10:46 AM
1 Kudo
If the sum of data sizes is greater than the amount of memory reserved for the hash tables (see below config param), then it happens. hive.auto.convert.join.noconditionaltask.size=1370MB
... View more
12-10-2015
09:53 AM
1 Kudo
In log, it is clear (retention.bytes -> 1073741824) that below parameter is applied. log.retention.bytes=1073741824
... View more
12-10-2015
09:18 AM
Write custome MBeans to derive custom metrics (like you mentioned) in storm and monitor them from JConsole or use tools Ganglia. https://dropwizard.github.io/metrics/3.1.0/manual... Kafka is JMX compliant, so enable JMX and monitor the same way as above. Please note that these metrics cannot be shown on ambari or ambari views. You may use Ganglia UI. For alerts you can use http://code.google.com/p/ganglia-alert/ All the Best:)
... View more