My site (www.singulair.com) has many truncated response errors, and
this has been going on for months now. I'm trying to determine why
google regards so many responses as truncated, and I'm not getting too
far.
Taking the previous thought one step further, long-running javascript
-- can inline script and body.load event handlers do something to
cause "truncated response" errors?
It may be that google uses more than the initial request's response
text when crawling a page... perhaps it downloads and uses files
referenced in that initial response (EG *.css, *.js, *.swf, *.gif
files, etc...) Some of the referenced files may result in 404s. Would
that cause "truncated response" errors?
Another potential cause is an incorrect "content-length" header, but
I've been told that this should not cause an "unreachable URL" error
(thanks, JohnMU!). Here's a related question: What does google use to
determine if a response is truncated: can this incorrect content-
length header affect the download of referenced files, therefore
marking a response as truncated? Can the fact that the server
promised 27853 bytes but only responded with 27849 bytes cause a
"truncated response" error?
And finally, if I have no luck in determining the cause through
gracious help from you (the reader of this posting), can anyone
provide me with a strategy to determine why google lists so many
"truncated response" errors?
Not sure whats kind of truncated responses you are having - but the
first thing I noticed was that the main domain spits out a 200
response and has a meta redirect to this long unnecessary landing
page... http://www.singulair.com/montelukast_sodium/singulair/consumer/index.jsp..
why would you want to do that?
Maybe using meta redirects has something to do with that? I'm not
sure.
Sorry I can't be of more help. My first guess would be to make sure
that your code is clean, lean and validated (http://validator.w3.org).
That should point out any serious issues. Next I'd externalize
Javascript and CSS.
On looking at the source code of the messy home page above, it
actually included my referrer info?
/*Added setReferrer function to set referrer value to DCS.dcsref
parameter 5/9/2006 */
function setReferrer(){
> My site (www.singulair.com) has many truncated response errors, and
> this has been going on for months now. I'm trying to determine why
> google regards so many responses as truncated, and I'm not getting too
> far.
> Taking the previous thought one step further, long-running javascript
> -- can inline script and body.load event handlers do something to
> cause "truncated response" errors?
> It may be that google uses more than the initial request's response
> text when crawling a page... perhaps it downloads and uses files
> referenced in that initial response (EG *.css, *.js, *.swf, *.gif
> files, etc...) Some of the referenced files may result in 404s. Would
> that cause "truncated response" errors?
> Another potential cause is an incorrect "content-length" header, but
> I've been told that this should not cause an "unreachable URL" error
> (thanks, JohnMU!). Here's a related question: What does google use to
> determine if a response is truncated: can this incorrect content-
> length header affect the download of referenced files, therefore
> marking a response as truncated? Can the fact that the server
> promised 27853 bytes but only responded with 27849 bytes cause a
> "truncated response" error?
> And finally, if I have no luck in determining the cause through
> gracious help from you (the reader of this posting), can anyone
> provide me with a strategy to determine why google lists so many
> "truncated response" errors?
> Thank you for your time, it is appreciated.
> -dave
> has a meta redirect to this long unnecessary landing page...
> why would you want to do that?
It's not the best design, especially from an SEO stance... it's what I
inherited. :(
> On looking at the source code of the messy home page above, it
> actually included my referrer info?
> Are you generating unique source code for every visitor?
Yes, you can do this with jsp pages. The artifact you noticed is used
for WebTrends tracking purposes. NOTE: only the metric-tracking
javascript is unique per page.
I believe I found my answer... approximately 3 months ago a new
revision of the site was deployed; this new revision had a redundant
folder removed. All of the UnreachableURL errors are a result of the
googlebot attempting to refresh its index of pages located in that
(since removed) subfolder. There are 301 redirects to ensure the
requests end up at the correct location, and I don't know why Google
is classifying these pages as having truncated responses... but at
this point, I don't really care. All of my Current and existing pages
are being crawled and indexed. All is well. Thanks again!