[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MirageOS-devel] Restart the mirage.io unikernel



Yup, just noticed too. Thanks for taking a look. 

Will have to look into this and figure out what's up. I think Magnus and Thomas are aware of retransmissions. 

It look that I've been bitten by partial applications ... the dispatch function were memoizing *all* the web-pages on every requests, ie. parsing all the templates, converting mardown to HTML, creating the atom feed, etc... for every element loaded (the page itself, the JS scripts, the CSS, the images, etc). This stupid mistake was introduced when I've refactored the website last week and is now fixed by [1]. I'm really surprised (and almost quite happy) to see how "well" the server responded to the big CPU load.

The fix is now live and the website starts to be much more responsive again, and the number of page served per second[2] is increasing steadily.

Best,
Thomas



Best wishes,
Amir

--
Sent via mobile

On 29 Jun 2015, at 19:28, David Scott <scott.dj@xxxxxxxxx> wrote:

Maybe someone beat me to it, but it seems to be responding now.

On Mon, Jun 29, 2015 at 7:24 PM, Amir Chaudhry <amc79@xxxxxxxxx> wrote:
Hi folks,

Could someone who has access, please restart the mirage.io unikernel?

Best wishes,
Amir

--
Sent via mobile

_______________________________________________
MirageOS-devel mailing list
MirageOS-devel@xxxxxxxxxxxxxxxxxxxx
http://lists.xenproject.org/cgi-bin/mailman/listinfo/mirageos-devel



--
Dave Scott
_______________________________________________
MirageOS-devel mailing list
MirageOS-devel@xxxxxxxxxxxxxxxxxxxx
http://lists.xenproject.org/cgi-bin/mailman/listinfo/mirageos-devel

_______________________________________________
MirageOS-devel mailing list
MirageOS-devel@xxxxxxxxxxxxxxxxxxxx
http://lists.xenproject.org/cgi-bin/mailman/listinfo/mirageos-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.