Hello.
I assume that you using client nodes or monitor network traffic between server
nodes.
In that case, there are metrics for network messages:
`communication.tcp.{node_consistent_id}.sentMessagesToNode` - count of
communication messages sent from local node to {node_consistent_id}
`communication.tcp.sentBytes` - count of bytes sent via communication by the
local node.
`communication.tcp.receivedBytes` - count of bytes received via communication
by the local node.
There are metrics for each message type sent by Ignite node:
`communication.tcp.sentMessagesByType.{type_id}`
`communication.tcp.receivedMessagesByType.{type_id}`
Where {type_id} is a internal Ignite message type id.
Please, take a look at this metrics to:
1. Make sure that Ignite nodes relates to traffic increase.
2. To know which message type makes most traffic for your deployment.
> 12 июля 2023 г., в 11:02, kimec.ethome.sk <[email protected]> написал(а):
>
> Any ideas?
>
> Kamil
>
> On 2023-07-10 16:39, kimec.ethome.sk wrote:
>> Hi Stephen,
>> nothing scientific, just network transfer rates between cluster nodes.
>> We upgraded Ignite nodes and nothing else. Cache configurations are
>> same as before and no OS tuning was changed after the upgrade. Yet, we
>> see network traffic increase between server nodes in our Ignite
>> cluster.
>> Kamil
>> On 2023-07-10 14:54, Stephen Darlington wrote:
>>> How are you defining “chatty”?
>>>> On 10 Jul 2023, at 13:33, kimec.ethome.sk <[email protected]> wrote:
>>>> Greetings,
>>>> we have recently upgraded Ignite server nodes from 2.8 to 2.14 and we see
>>>> a ten fold increase in cluster chattiness.
>>>> Since 2.8 was quite old, I assume there may have been some announcement
>>>> about protocol changes but I could not find any info on my own.
>>>> Is this the expected behavior?
>>>> Kind regards,
>>>> Kamil
