[Seaside] Re: REST and Seaside

[Nevin Pratt]

>
>
>The stats may not be accurate because of caching of static
>pages by the web browsers and by the ISPs. Once a user has
>browsed your site, then many of the images/pages will already
>by cached. When repeat customers come back, they probably
>hit your Seaside box much more heavily than your static
>page serving stats are indicating.
>

Yea, that makes sense.

>
>Congrats on the success of your site, and I hope Seaside
>continues to work for you. BTW, do you have any stats on
>how hard Postgres is being hit.
>
>--yanni
>  
>

I don't have PostgreSQL stats, but I know PostgreSQL is not hit hard.  
Most everything is kept in the image-- there is *no* PostgreSQL access 
for read-only viewing anywhere on the site, and only very minimal 
PostgreSQL access when an order is placed.  All of the product 
information is kept in the image at all times.  A "knee-jerk" thought is 
to think keeping all the product information in the image at all times 
is the cause of the image bloat, but it's not.  To illustrate, when the 
image is pushing a gigabyte in size, if I "Clear Caches" (from the 
Seaside Admin page) the image size suddenly drops to about 25 megabytes.

Typical image yo-yo size used to range from a low of about 50 megabytes 
(during low session activity) to a high of around 250 megabytes (during 
high session activity), and it would yo-yo up and down all the time as 
the site was hit.  I was used to seeing this, as it *always* yo-yo'd up 
and down like this, even from the beginning of the creation of the 
site.  This was normal (and expected) activity.

Then I saw the yo-yo high of 250 MB start climbing-- 300 MB, then 350 MB 
or so after a couple of months, then 500 MB, etc.  Now, after several 
months more of this, I'm seeing it yo-yo up to a gig or so.

And at any time, as I mentioned above, and regardless of the image size, 
if I "Clear Caches" (from the Seaside Admin page) the image size 
suddenly drops to about 25 megabytes.

So, it's like I've said-- it looks like it is all the state info that 
Seaside hangs on to, combined with a session timeout of 6000 seconds, 
combined with raw traffic volume, all combining to create the current 
problem.

Nevin