Forum

[tajones]

We have a very small Kognitio cluster in our non-prod environment. It's an older version (ver80202rel180830). I'm trying to reconfigure to give the container more RAM but I cannot seem to do this either using the config file or setting the values on the command line. How do I know where these setting are coming from and how can I resize? What I have tried is below:

1) Setting on command line:

Command:
CONTAINER_MEMSIZE=40960 kodoop cluster kognitiodev reconfigure

Result:
Kognitio Analytical Platform software for Hadoop ver80202rel180830.
(c)Copyright Kognitio Ltd 2001-2018.

Current cluster configuration
=================================================================
Cluster configuration for kognitiodev
Containers: 1
Container memsize: 16384 Mb
Container vcores: 4

Internal storage limit: 100 Gb per store
Internal store count: 1

External gateway port: 6550
Kognitio server version: ver80202rel180830

Cluster will use 16 Gb of ram.
Cluster will use up to 100 Gb of HDFS storage for internal data.

Data networks: all
Management networks: all
Edge to cluster networks: all

Use nodes from queue: <<default>>
Stats reporting: Disabled
=================================================================
Adjustment of internal store limits is not currently supported
Add or remove stores instead

2) Using the config file:
Command:
kodoop cluster kognitiodev reconfigure /home/kodoop/kodoop/config/kognitiodev.sh

Within kognitiodev.sh the following is set:
CONTAINER_COUNT=1 ## number of containers to allocate for the cluster
CONTAINER_MEMSIZE=40960 ## memory size in Mb of each container
CONTAINER_VCORES=2 ## number of vcores each container will use

INTERNAL_STORAGE_STORE_COUNT=2 ## number of internal datastores to create in HDFS and
## use for table storage (defaults to 1 per container)
INTERNAL_STORAGE_STORE_LIMIT=4 ## max size in Gb for each datastore

GATEWAY_PORT=6550 ## the ODBC port number on edge nodes to listen on

NETWORK_SPEED=1000 ## network speed to configure for. 1000 is 1G ethernet. If you have
## 10G or above then using 10000 here will improve performance

Result:
Kognitio Analytical Platform software for Hadoop ver80202rel180830.
(c)Copyright Kognitio Ltd 2001-2018.

Current cluster configuration
=================================================================
Cluster configuration for kognitiodev
Containers: 1
Container memsize: 16384 Mb
Container vcores: 4

Internal storage limit: 100 Gb per store
Internal store count: 1

External gateway port: 6550
Kognitio server version: ver80202rel180830

Cluster will use 16 Gb of ram.
Cluster will use up to 100 Gb of HDFS storage for internal data.

Data networks: all
Management networks: all
Edge to cluster networks: all

Use nodes from queue: <<default>>
Stats reporting: Disabled
=================================================================
Adjustment of internal store limits is not currently supported
Add or remove stores instead

[michaeld]

Hi,

The error is 'Adjustment of internal store limits is not currently supported, add or remove stores instead'. Your current configuration is:

Internal storage limit: 100 Gb per store
Internal store count: 1

And in kognitiodev.sh you have:

INTERNAL_STORAGE_STORE_COUNT=2 ## number of internal datastores to create in HDFS and
## use for table storage (defaults to 1 per container)
INTERNAL_STORAGE_STORE_LIMIT=4 ## max size in Gb for each datastore

So it's attempting to change INTERNAL_STORAGE_STORE_COUNT, which is fine, and INTERNAL_STORAGE_STORE_LIMIT, which is not supported. If you edit kognitiodev.sh to set INTERNAL_STORAGE_STORE_LIMIT back to 100, or remove the line, it should work. As the error suggests, the size of each internal store (virtual disk) cannot be changed, but you can add or remove stores.

Regards,
Mike

[tajones]

Thank you, that worked. I guess since I was not attempting to modify those settings I didn't even realize they were different. I thought that message was more of an informational message vs an error message. Thanks!