The discussion revolved around the caching of data from the agents to the monitor during network outages. It was confirmed that the monitor caches data for 24 hours in C:\Windows\ServiceProfiles\NetworkService\AppData\Roaming\ControlUp\CacheActivity and uploads it once the connection is restored. It was also mentioned that the agent does not have caching capabilities, only the monitor does. The potential impact of agent-level caching on user experience in VDI/RDSH environments was also brought up. This topic is not specific to v9 of ControlUp.
Read the entire ‘Understanding Data Caching in ControlUp’ thread below:
Looking for documentation on agent comms to the monitor when the network is down. How long will the endpoint cache data and send when the link comes up?
I assume this has to do with the reverse agent communications in v9? Not sure about if we even cache it for outbound (don’t believe so) but for inbound (old method) it’s not cached at all.
Inbound is cached as part of the cached config in the service profile.
But a great question that I have no idea about…
I was not asking specifically about v9.
Just wanted to know how much data or how many hours\days of data can be cached while link is down
Inbound is 24 hours. Should be the same for outbound on agents that are already connected to a monitor. If an outbound agent is restarting or comes up new during an outage, that I’m not sure about.
@member Can you confirm the outbound agents behavior during an internet outage?
As for more detail, the monitors cache data in C:\Windows\ServiceProfiles\NetworkService\AppData\Roaming\ControlUp\CacheActivity and then upload. If the upload fails it will keep the cache and upload when the connection is restored.
Which network are we talking about?
Monitor CU cloud? Or monitor agent?
monitoragent
Because the discussion above is about monitor cloud
Well caching in general. In this case, not v9 specific.
My point is if the monitor can collect the data but not hit the cloud, its cached.
The unknown is v9 outbound and if the agent caches the monitor to connect to. Thats the only unknown in an "offline" scenario.
Yeah but if the monitor agent can’t talk, the agent doesn’t cache anything
so there is or is not agent level caching?
ok
The monitor caches. If the monitor can connect to the agent but doesn’t have internet it caches the agent data there.
There is no caching agent side, only monitor side.
ty!
In the VDI/RDSH world, caching on the agent is not a good thing since every single machine is going to write to the shared storage and that can impact the user experience. I’ve seen this during POC against other solutions, I can name EdgeSight (and its firebird cache!) as it’s not a product anymore…
Continue reading and comment on the thread ‘Understanding Data Caching in ControlUp’. Not a member? Join Here!
Categories: All Archives, ControlUp Real-Time DX