Hi, John:

Sorry for another question so soon.

I now have two datasets as follows:

Kunena topics & categories
JReviews using our content

I've been able to precache and export the JReviews content fine. And the precache works fine for the Kunena dataset.

However, I'm running into an error message when I try to export the Kunena dataset.

QUOTE:
0 : Unable to parse XML string

If I simply try the "Link to XML Sitemap" it generates quite a few lines of the sitemap but then it stops and displays the following error message in my browser:

QUOTE:
This page contains the following errors:

error on line 45879 at column 74: PCDATA invalid Char value 27
Below is a rendering of the page up to the first error.

I assume an ASCII character is messing things up.

Any ideas on how to fix that?

Thanks. -- Rik

UPDATE: I deleted the last Kunena topic that was listed via the "Link to XML Sitemap" but that didn't change anything. I assume the problem point is immediately after that. Also, in retrospect, although the precaching seemed to work fine, it only showed about 50K lines and we have 362,000 messages, 56,000 topics, 335 categories. So I'm guessing that it aborted at the error point indicated above.

Hi Rik,

looking at your problem it could be related to a scalability issue. Maybe having so much links the process exists before completing leaving the XML broken.
Also if your links exceed 50,000, you should activate the splitting feature before exporting it.

You should:
1. Enable splitting for example 30,000 links for each file
2. Precache the Kunena topics, etc
3. Export the splitted sitemap

If you have still the problem, so maybe there could be really some invalid characters somewhere, but i should look at it deeply to tell you more.

Best,
John

Hi, John:

I followed your suggestions. Also, I separated the Kunena categories and topics so that that two separate datasets/sitemaps could be created. I can build the Kunena categories sitemap fine but the Kunena topics are the problem and they abort at the same point.

BTW, I have already built split sitemaps several times for our JReviews content of 664,000+ listings with the precaching limit set as high as 50,000. That uses less than 10% of our server memory -- although it uses a lot of CPU. So memory issues don't seem to be the problem.

So what to try next?

Thanks. -- Rik

Hi Rik,

good to hear that you was able to generate a splitted sitemap for more than 600.000 links. As you may know dealing with so large sitemaps is a great challenge.

About the problem with Kunena topics, at this point i think that you was right when saying that an ASCII character is messing things up.
It could be the case that one or more posts link contain some invalid characters breaking the XML format.

My only idea would be to try to identify those posts so eventually they could be excluded. If you look at the screenshot below you see that data sources can be configured to include/exclude topics or categories by IDs.

John