[Spacewalk-list] Need to know about spacewalk internal architecture and why spacewalk is taking time to deploy scheduled actions

David Rock david at graniteweb.com
Tue Apr 2 16:07:49 UTC 2019


> On Apr 2, 2019, at 09:56, James Hibberd <jhibberd at proact.co.uk> wrote:
> 
> For Question 3 – I think this might be what you are looking for
>  
> On the client side you can change the check in times – Standard config is 240 minutes , so every 2 hours the client will check into the server.
>  
> On the clients change the value inside
>  
> /etc/syscon/rhn/rhnsd
>  
> To something like 15 minutes, I have my clients checking in every 40 minutes instead of the 240 minutes

It won’t do anything lower than 60 minutes, regardless what’s set.  I forget exactly where it’s documented, but it is documented.

The important thing to understand is that it’s based on the clients checking in at regular intervals (whatever you end up setting it to), so the lag can be up to the interval you set on the individual clients.  When you set an action to be run on the spacewalk end, it’s just setting a marker so that when the client checks in, it sees if there is something to do.  “forcing” it to run by using rhn_check is just manually making it check in before it’s scheduled time, which is why that appears to make it go faster.


— 
David Rock
david at graniteweb.com








More information about the Spacewalk-list mailing list