Metadata expiry and computing a new expiration time.

[Simon Fraser]

On Sat, Sep 12, 2015 at 04:43:37PM +0100, Rod Widdowson wrote:

> > > I'm a little confused, sorry - I'm not getting any memory or heap
> related
> > errors, the only thing in the logs is that it has successfully retrieved
> and
> > unmarshalled the federation metadata, and updated the expiry time, when it
> > hasn't. Is that heap related?   If so I can experiment, 
> 
> There is anecdotal evidence (no more) that the http client does not function
> well in memory short environments.  All it would take would be some code to
> accidently swallow the exception silently and hang instead.  It could even be
> the native code below it running into problems.  

Ok.  Thank you for the explanation. I will request more memory for that virtual
machine, so that I can increase Tomcat's allowance - it's currently using 600Gb
resident.

Thanks,
Simon.



-- 
 The Wellcome Trust Sanger Institute is operated by Genome Research 
 Limited, a charity registered in England with number 1021457 and a 
 company registered in England with number 2742969, whose registered 
 office is 215 Euston Road, London, NW1 2BE.