'idp.service.metadata.checkInterval' on specific time

[Zico]

On Mon, Mar 18, 2019 at 8:24 AM Cantor, Scott <cantor.2 at osu.edu> wrote:

> On 3/18/19, 9:05 AM, "users on behalf of Zico" <
> users-bounces at shibboleth.net on behalf of mailzico at gmail.com> wrote:
>
> > Sorry.. 'drastic under-resourcing' means?
>
> A couple of gigs of heap and maybe 4G of RAM is enough to run a high
> volume IdP if you had to.
>
> > Generally these servers are running with 16GB physical memory; allocated
> max 6.5G for shibboleth only...
>
> That's far more than necessary and you clearly have some other problem.
> Perhaps you're giving it too much memory and causing GC issues. I handle
> 250,000+ logins a day with 2.5G on a 16GB machine, and there are zero
> issues with metadata reload of the InCommon aggregate.
>

Hmm..... that makes sense. Ok, I'll give a shot with 2.5G. I do have little
suspicion on this GC issue though; can't disagree on that.
I will share my result with you.



>
> -- Scott
>
>
> --
> For Consortium Member technical support, see
> https://wiki.shibboleth.net/confluence/x/coFAAg
> To unsubscribe from this list send an email to
> users-unsubscribe at shibboleth.net
>


-- 
Best,
Zico
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://shibboleth.net/pipermail/users/attachments/20190318/037023d6/attachment.html>