[
https://issues.apache.org/jira/browse/IGNITE-15102?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ivan Daschinsky updated IGNITE-15102:
-------------------------------------
Description:
I suggest to add monitoring capabilities to {{pyignite}} similar to
[pymongo's|https://pymongo.readthedocs.io/en/stable/api/pymongo/event_loggers.html#module-pymongo.event_loggers]
Suggested api:
{code:python}
from pyignite.monitoring import OpLogger, ConnectionLogger, TopologyLogger
from pyignite import Client
client = Client(event_listeners=[OpLogger, ConnectionLogger, TopologyLogger])
with client.connect(...):
......
{code}
I suggests to add listeners to:
# *Connection events* connect or disconnect to specific ignite server,
connection errors
# *Topology events* when partition awareness is enabled, log new topology
versions
# *Operations events* start,success or failure with request_id, server
(address, port and uuid), operation_id, error string if presents
This approach can implement custom metrics, tracing and other useful
client-side stuff in order to make client more observable
was:
I suggest to add monitoring capabilities to {{pyignite}} similar to
[pymongo's|https://pymongo.readthedocs.io/en/stable/api/pymongo/event_loggers.html#module-pymongo.event_loggers]
Suggested api:
{code:python}
from pyignite.monitoring import OpLogger, ConnectionLogger, TopologyLogger
from pyignite import Client
client = Client(event_listeners=[OpLogger, ConnectionLogger, TopologyLogger])
with client.connect(...):
......
{code}
I suggests to add listeners to:
# *Connection events* connect or disconnect to specific ignite server,
connection errors
# *Topology events* when partition awareness is enabled, log new topology
versions
# *Operations events* start,success or failure with request_id, server
(address, port and uuid), operation_id, error string if presents
This approach can implement custom metrics, tracing and other useful
client-side stuff in order to make client's more observable
> Implement monitoring of various pyignite's events
> -------------------------------------------------
>
> Key: IGNITE-15102
> URL: https://issues.apache.org/jira/browse/IGNITE-15102
> Project: Ignite
> Issue Type: Improvement
> Components: python, thin client
> Reporter: Ivan Daschinsky
> Assignee: Ivan Daschinsky
> Priority: Major
> Labels: python, thin
>
> I suggest to add monitoring capabilities to {{pyignite}} similar to
> [pymongo's|https://pymongo.readthedocs.io/en/stable/api/pymongo/event_loggers.html#module-pymongo.event_loggers]
> Suggested api:
> {code:python}
> from pyignite.monitoring import OpLogger, ConnectionLogger, TopologyLogger
> from pyignite import Client
> client = Client(event_listeners=[OpLogger, ConnectionLogger, TopologyLogger])
> with client.connect(...):
> ......
> {code}
> I suggests to add listeners to:
> # *Connection events* connect or disconnect to specific ignite server,
> connection errors
> # *Topology events* when partition awareness is enabled, log new topology
> versions
> # *Operations events* start,success or failure with request_id, server
> (address, port and uuid), operation_id, error string if presents
> This approach can implement custom metrics, tracing and other useful
> client-side stuff in order to make client more observable
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
