Pinot stores cluster metadata, including schema defintions, table config, and segment assigment, in Zookeeper. Internally, Zookeeper uses a hierarchical namespace, similar to a file system. The nodes in this hierarchy are called znodes and they can store data and have children.
The default maximum size of znodes is 1MB, which is sufficient for most deployments. However, if you have 100s of thousands of segments, it's possible that this size limit is exceeded. If this happens, you will see an error message similar to the following:
java.io.ioexception: packet len 1250829 is out of range!
There are broadly two ways that we can address this problem.
The first way is to reduce the number of segments, which will reduce the amount of metadata that's stored in the
EXTERNALVIEW znodes, which are the two znodes most likely to exceed 1MB.
The other way is to adjust the maximum size of znodes in Zookeeper.
This is done by configuring the
jute.maxbuffer Java system property, which defines the maximum znode size in bytes.
To read more about this property, see the Zookeeper documentation (opens in a new tab).
A reasonable value to try out would be 4MB and the parameter should be set to the same value on the Zookeeper node as well as clients that are interacting with Zookeeper i.e. all of the Pinot components