Forum Discussion
SyadSulthan
Nimbostratus
NimbostratusPool Member went Down but server is up
Today i faced an issue.. i have a virtual service in which inly one member is configured. Suddenly we couldn't able to access the web service, when i check the status the pool member went to inactive but i can able to ping the server. After few minutes Pool member came up and services started working without any troubleshot.
When i check the logs it shows pool member/common/ip monitor status down.
Do anyone have idea on this and how can i avoid such issue in future.
Thanks,
niels_andersenHow is your monitor set up? It sounds like normal behaviour to me. There is usually a small delay with the monitor check.
if your monitor is on tcp up, then perhaps the node was patched or updated, it bounced and came back. If using http, then the http 200 ok state may not have been available for the monitor url
- zamroni777
Nacreous
ping is only L3 test and it doesnt reflect the status of the server's application layer
if the webserver software/service is down, ping monitor is still successful.ideally, the pool monitor must be application layer monitor, e.g. http monitor to the path that truly reflect server's application state.
- niels_andersen
indeed, and why I asked how the monitor was set up
SyadSulthanThe up interval for the monitoring has been set to disabled from the beginning. Also, i am not sure if the pool member has been set to application layer monitoring or not.
