4

I have an item (in a bucket) that I cannot find in the Coveo indexes. I can find the item in my master and web databases no problem, but when I go to the CES 7 admin panel (this is a free on-prem edition) and search my indexes for the item name, I get no results.

I made a small change to the item and saved/published, and in the indexing manager I see a change to the time, but I don't see a status change when I go back to the index status page in the Index - Sources and Collections window - it says the last operation was "rebuild" source over a week ago. I would think this would show the incremental change I made.

I also confirmed the published date/unpublished date aren't preventing it from being published. I don't see anything in the log that screams "problem" either, so I'm now lost in the woods. Does anyone have a gumdrop trail to help? :)

| improve this question | |
5

If this is version 4, make sure that the item is within the scope of your crawler: https://developers.coveo.com/display/SitecoreV4/Changing+the+Crawling+Root+of+an+Index The crawler scope by default will only include Media Library >> Files and Sitecore >> Content. If your bucket is outside of this path, you need to change the crawler's scope.

Also, if you are using Inbound Filtering in the indexing pipeline, make sure it is not removing your item. Sitecore Inbound filtering will affect all indexes, including Coveo.

Finally, since this is an On-Premises index, keep an eye on the RabbitMQ queues when rebuilding: https://developers.coveo.com/display/SitecoreV4/Accessing+the+RabbitMQ+Management+Console+-+On-Premises

Once you are logged in, navigate to the Queue tab to see the status of your queues. If you see remaining documents in your queues, then you have a problem. You can get the messages by using the GetMessage section of that tab. The Get Message will do a "pop" action in the queue, which will allow you to see the message, but will also remove it.

And also keep an eye in the CES Console when rebuilding, the error might not show on the Sitecore side, but could show in Coveo: https://onlinehelp.coveo.com/en/ces/7.0/administrator/using_the_ces_console.htm

For larger items, messages going through the queue is a simple reference to the Sitecore item, which means that Coveo will need to ping Sitecore back. It is possible that Coveo could be unable to reach certain items, the console will tell you.

| improve this answer | |
  • This is the right track - I found the Rabbit MQ was not started. I'd originally set up indexes individually for CM and CD before getting the farm setup correct for proper operation, so I dumped the CM/CD indexes. Now I'm wondering if I screwed something up...when I rebuild the coveo_master_index in Sitecore, I'm not seeing anything happen in the CES console as described here: developers.coveo.com/display/public/SitecoreV4/…. There's no rebuild error, and I still get results, but the CES admin doesn't say it's rebuilt. – Ken McAndrew Nov 9 '16 at 18:01
  • 1
    Okay, I got a little extra help to finish this off. So after turning Rabbit MQ back on, the linkage between Rabbit and Coveo wasn't reinitialized. I had to go into my indexes in the CES admin tool and use the "disable live monitoring" button, the re-enable it. This is in this guide: source.coveo.com/2014/09/23/coveo-for-sitecore-troubleshooting - credit to Jeff at Coveo for pointing me there on Slack. – Ken McAndrew Nov 10 '16 at 17:01
  • 1
    I wrote this blog! :) – Simon Langevin Nov 10 '16 at 19:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.