Pressbooks Clone Fails-Event Stream Connection Error (Reload)- Missing Parts/Chapters on clones

Good morning. We are a self-hosted Pressbooks system that has just gone live. When we clone books with many H5P activities, such as the one listed below, we are missing parts from the books. All I have is chapter 1/Part 1 from my clone yesterday: American Government (2e – Second Edition) – Simple Book Publishing -set to public-
We have noticed this on multiple books. Every time we use the clone tool we get the error, Event Stream Connection Error (Reload). The book does show up in our list of books but as stated above they are missing parts.

We are also seeing this error on logs: Failed to download the requested H5P, cURL error 28: Operation timed out after 30001 milliseconds with 0 bytes received

We have enabled Content Types/ Enable LRS dependent content types on the H5P settings for the book immediately upon clone and waited up to 8 hours, one of our book clones worked and has the H5P activities: 1.3 be ب ، Jim ج , and sin س Series – Basic Urdu
-set to public-

[American Government (2e - Second Edition)]

CC BY | [154 H5P activities]

Over the weekend, the American Gov’t content all came over witht he H5P activities. Chapter 1 Review – American Government (2e – Second Edition)

We still are getting the error listed previously and it seems to be taking at least 48 hours for the clone to complete. Does the Failed to download the requested H5P, cURL error 28: Operation timed out after 30001 milliseconds with 0 bytes received seem to be the culprit? If so how would we get the error off our system and improve the speed?

hi @jhill the duration of a clone will usually be determined by the size & complexity of the book you’re cloning and the server throughput/capacity of the two networks interacting in the cloning routine. If you’re finding that cloning operations are taking that long to complete, you may want to consider improving the server specs for wherever you’re hosting your Pressbooks network. There are several open source network admins in this forum that may be able to advise you on how they’ve done this. If that’s beyond your capabilities, we support our work on this project by providing enterprise SaaS hosting for more than 150 educational institutions. My colleague John McLeod would be happy to tell you more about that option if that’s of any interest (Pressbooks plans & pricing | Pressbooks).

1 Like

Steel, Thank you. We will explore from that side of things. We are trying to go the self hosted route for now. The error was one of the things we found odd, that the clone fails but ends up working.

1 Like

Ah, I see. In this case, the clone operation isn’t failing. The error message simply indicates that the eventstream that’s providing the real time update on the cloning progress/status has experienced a connection error. This usually happens because it’s hit a timeout limit of some kind. In most cases, the clone or export process you’ve initiated is still running and will likely run until completion, but you won’t be able to see real time browser updates because the eventstream has crashed.

1 Like

I notice that the example you use is cloning from Lumen Learning’s Candela/Pressbooks.

Lumen’s implementation of Pressbooks relies heavily on their custom theme, Bombadil, which was based on Pressbooks book 1.0. Because they relied so heavily on that theme to work with their other applications, they stopped updating Pressbooks to the latest version in 2018. They still do updates and patches of course, but really its an older version.

Especially when cloning those large OpenStax or large media rich books, sometimes I do not use the clone tool, but instead create the book and use the import tool. Using the import tool let’s me do one chapter or part at a time, instead of trying to pull it all at once. I find that if I chunk it in that way, I get fewer errors.

The other reason I built the book first and then import the content in is because I want to preserve the attributions that Lumen puts at the bottom of every page. Those expandable attributions are created using a plugin they created, Candela Citations. When you do the copy, whether it is through the clone tool or the import tool, Candela Citations must be activated first before you move the content to your server or the information just kind of gets lost. Since I don’t have it network activated on my server, I have to create the book first, and then activate the plugin.

2 Likes