Edwin

Don't know

__________________

LissaKay

This is still unresolved. I am beyond frustrated ... somebody needs to fix this!

The latest from Google support:
Surely someone somewhere knows how to resolve this!

__________________
LissaKay.com is on SH130

MarkRH

My WordPress feed seems to be updating fine with Google Reader, looked at my most recent visitors to it and saw:
http://www.markheadrick.com/images/s...eedfetcher.gif

Might check your latest visitors list and see how many bytes are being transmitted out with a code of 200.

Kind of reminds me of those cases where a burned CD will not play in Player A but will play in Player B. But Player A will play other burned CDs. Which is broken? The CD or the Player?

I notice your using Expression Engine.. unfortunately, I've no experience using it. If I get bored (heh..) I might make a test site with it. Uhm.. no promises there though.

__________________

LissaKay

According to my latest visitors log, the server is returning a 304 response to Google Feedfetcher, even though there are several new posts:

*
/index.php/weblog/rss_2.0/
Http Code: 304 Date: Nov 14 16:37:37 Http Version: HTTP/1.1 Size in Bytes: -
Referer: -
Agent: Feedfetcher-Google; (+http://www.google.com/feedfetcher.html; 13 subscribers; feed-id=786644724791516834)


All other crawlers/bots are receiving the correct response, either 304 or 200 with an appropriate file size.

=================

PLEASE! There has to be someone that knows what is going on here!

I changed NOTHING in my feed when it all of a sudden stopped working. It had been working fine in all readers before. Now ONLY Google Reader is not updating.

Surpass support has washed their hands of this, and is refusing to look further into the matter, such as the latest response I got after harassing Google endlessly about this. I am seriously considering having them move me to a different server! Or just cancelling all of my accounts and going elsewhere!

__________________
LissaKay.com is on SH130

Edwin

Could it be the server? I'm on the same server as you, but I've got no clue on how to test this.

__________________

LissaKay

Edwin,

I'm already using your feed to try to prove to the 'tards at Google that it isn't the server that is the problem ... or EE, or anything on this end. It's them and they just won't admit it

Thanks!

__________________
LissaKay.com is on SH130

Edwin

You welcome

Now I should make sure I post some good things

__________________

Edwin

Seems like my feed is getting through, but your last is from November 8.

So strange. To me (as newbie in this kind of cases), it looks like it's something with Google.

__________________

LissaKay

The latest from Google, which is still being ignored by the help desk ...

I keep forwarding this stuff, but they still fail to do the needful thing

====================

First of all, you can find some background information on Google's
crawl caching proxy here:
http://www.mattcutts.com/blog/crawl-caching-proxy/

You can see there how the system is set up: Google's various crawl
services (web search, blog search, etc.) all share a common cache of
crawled content, to minimize the load we put on people's sites.
However, these services do need to verify that the cached copy of a
page is still current. They do that using "if-modified-since"
requests, which are more light-weight than full content requests. In
the case of your feed, this is what happens:

1. Google Reader checks the crawl cache and finds a recent copy of
your feed.
2. Wanting to make sure it's a current copy, Reader sends your server
an if-modified-since request. (Assume for now that your blog hasn't
been updated, since we crawl more often than you post.)
3. Instead of a 304 ("Not Modified") response, your server sends back
a 200 ("OK") with no content.
4. Reader sees the 200 and assumes this must be the new, up-to-date
version of your feed.
5. Reader therefore ignores the cached copy, and "updates" your feed
with the version it got. Unfortunately, that version was empty, so
there's no change to make.

Some important points about this:
* If we had gotten a 304 in step 3 above, Reader would have used the
cached copy, which probably would have had your recent posts in it.
* If the timing works out such that your blog has indeed been updated
since the last time it was crawled and cached by one of our services,
then the 200 would come back with content and we'd update it in Google
Reader. This explains why you do occasionally see updates. They're
infrequent, though, because there's a relatively narrow time window
relative to our crawls during which you can post and have your server
return a correct response.
* Your feed updates fine in BlogLines and other services because they
presumably don't have this same system set up, with the cached copies
and the "if-modified-since" requests. (If they're only running a
single service, there's less of a need to set up a more complex system
like this.)
* The curl command, referenced several times before, will return
different results depending on the date used. If put far enough in the
past, it will return your content, since you've updated since then.
However, our requests generally have a very recent date/time, so it's
more common for there to be no updates.

To sum it all up, it really is a matter of it all coming down to the
200 vs. 304 responses. If you can get your server to consistently sent
304 responses for unmodified content, then the issue will most likely
go away. For reference, the spec for using 304 response codes can be
found here:
http://www.w3.org/Protocols/rfc2616/rfc2616-sec10.html

If you can get this change made and still verify that there are
problems, we'll be happy to look into this again. However, as things
stand now, there is nothing further we can do from our end. I'm sorry
for all the back-and-forth about it, and I know it's a confusing
issue, but hopefully this (admittedly long) explanation helps.

Thanks,
Graham

__________________
LissaKay.com is on SH130