Upped memory guidance

[Janemarie Duh]

Thanks.

We just experienced our second production service incident since the
InCommon production aggregate was synced with the preview aggregate.

On Monday the 15th (upon refreshing the metadata, and then again this
evening, our IdP (v2.3.8 on RHEL 5.5) threw an out of memory error and
stopped accepting requests.[1] Restarting the container (Tomcat 6)
restored service. Java heap space is set to 1024K.

Once we consult with our sys admin tomorrow to make sure the server has
enough resources to handle 2G (or even 1.5) of heap space, we'll
increase and hopefully avoid another incident.

     Janemarie

[1]
Feb 18, 2016 5:30:57 PM
org.jasig.cas.client.proxy.ProxyGrantingTicketStorageImpl retrieve
INFO: No Proxy Ticket found for [].
Exception in thread "http-443-Acceptor-0" java.lang.OutOfMemoryError:
Java heap space
Exception in thread "Timer-0" java.lang.IllegalArgumentException:
Negative delay.
at java.util.Timer.schedule(Timer.java:191)
at

org.opensaml.saml2.metadata.provider.AbstractReloadingMetadataProvider.refresh(AbstractReloadingMetadataProvider.java:270)
at
org.opensaml.saml2.metadata.provider.AbstractReloadingMetadataProvider$RefreshMetadataTask.run(AbstractReloadingMetadataProvider.java:508)
at java.util.TimerThread.mainLoop(Timer.java:534)
at java.util.TimerThread.run(Timer.java:484)


On 2/18/16 3:38 PM, Cantor, Scott wrote:
>> No, I don't remember where the guidance was for V2 but that should be
>> changed also.
> 
> Found it, changed, thanks for the reminder.
> 
> -- Scott
> 

-- 
Janemarie Duh
Lafayette College
Identity Management Systems Architect
Information Technology Services