Ladislav Slezak write:
> 
> Hi all,
> 
> I have added a new config file for the WebYaST control panel.
> It's stored in /etc/webyast/control_panel.yml file.

It could be problem if we want separate ws from ui, as this directory is used 
for ws now so it could be confusing if you modify ui behavior in same directory 
as ws behavior.

> 
> This change allowed me to fix bugs:
> Bug 558027 - Regularly update patch status
> Bug 558028 - Health status needs to be updated regularly
> Bug 580025 - It should be possible to disable "update status" and "system 
> status" in
> the control center
> 
> So the overall statuses are automatically refreshed now or they can be 
> completely
> turned off (that can be useful during development or testing, you don't need 
> to wait
> for checking updates and system status).
> 
> 
> Here I'd like to discuss some issues I found:
> 
> 
> The config is stored in /etc/webyast/control_panel.yml and there are some 
> other files
> in that directory, but I found that there is /etc/webyast/vendor/logs.yml 
> config file.
> 
> So my question is what is /etc/webyast/vendor directory for? Is there any 
> difference
> between /etc/webyast/vendor*.yml and /etc/webyast/*.yml files? Which one 
> should I use
> in this case?
> 
> 
> Another problem I see is in the defaults:
> 
> 1) The requested default for patch status refresh is 24 hours. I agree that 
> this
> refresh timeout should be rather longer, but this is too long - the problem 
> is that
> 24 hours refresh time out collides with session time out.
> 
> After 24 hours the session will be invalid and no status will be displayed 
> and thus
> the feature with this particular refresh value is useless.
> 
> So I'd like to change it to something like 6 or 8 hours.
I agree that sounds reasonable.
> 
> 
> 2) A similar problem is with the system status time out (but the other way 
> round in
> this case). The requested default value is 1 minute which I find too low. 
> Evaluating
> system status is a quite expensive operation and takes a long time.
> 
> Doing that every minute creates extra load on the target machine. More over 
> if users
> really need such an immediate warning I guess they will very likely use other 
> means
> (like email, SMS message, custom scripts, etc.) without WebYaST involved.
> 
> So here I propose to use default time out value of 5 or 10 minutes.

It is too long for me. 1 minute is good as it can contain problems which you 
want see ASAP. If it is too big load then we should more optimalize it (like 
simplier question on backend - same solution use permission which return 
description for permission only on request).

Josef 
-- 
Josef Reidinger
YaST team
maintainer of perl-Bootloader, YaST2-Repair, parts of webyast
-- 
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to