end-user gets stuck, corrupt suspend data
Nov 08, 2021
This happens often enough I'd like to know if others have encountered it and what a better solution is.
This is a random occurrence for the occasional user, for an occasional course. For almost all users this does not happen for a given course. For this question you can assume latest browser is being used; it always is when this happens.
Publishing Storyline 360 courses as SCORM packages and end-users access in LMS.
Here's a scenario:
The end-user gets to a certain point in a course, that all other users with the exact same environment get to and get past just fine. Let's call it screen 7 of 10 and it is a standard Storyline multiple choice question with a Submit button. We never know how the user ended up in this state but the user launches the course, resumes to screen 7 of 10, an answer is selected, no other answer can be selected, and there is no Submit button. The users cannot proceed because the user can't submit an answer--required to move to the next screen.
The suspend data has been corrupted somehow. The only way I can ever solve the problem is to launch the course myself with a different user, get to screen 7 of 10, do not select an answer nor submit an answer, and exit the course. I then copy the suspend data I generated, and overwrite the end-user's suspend data (in the database). Now the end-user launches the course and resumes to screen 7 of 10 and can select an answer, click Submit, and go to the next screen. If I copy the end-user's previous suspend data and overwrite the suspend data for my user, I have the exact same problem with my user the end-user had: I cannot change answer selection, I cannot click Subit; I cannot move forward in the course. The problem seems to be the suspend data got corrupted.
This is a painful way to fix this occasional problem. It is unacceptable to tell the user "Sorry, you'll have to start over" after deleting the corrupt suspend data. As we've all known for many years, there is no way to edit the unintelligible Articulate suspend data and fix it. I have to march through the course as a different user to generate uncorrupted suspend data so I can give it to the end-user so the end-user can continue without having to start over. All of our courses have locked navigation so this is always very time consuming to get to the same place the user got stuck.
Does Storyline or anyone have a suggestion for an easier way to fix this problem when it happens? (Yet another plea for Storyline to offer a suspend data decrypter so we can solve such problems!)
Thanks
10 Replies
Wow -- thanks for thinking up this solution, which is a lot better than making the user retake the course!
I have occasionally and very upsettingly have this happen to users in my client's LMS, with no solution other than to make them retake the course. The only surefire way I have found to avoid it, which was given on a related thread, was never to allow slides to have "Resumed Saved State" or "Automatically Decide" in the Slide Properties/When Revisiting option -- they should always be set to "Reset to Initial State".
I don't think, however, I've applied this to Quiz questions because I haven't encountered it with them. I'm afraid it would mess up the Review Quiz feature if that were being used.
It's also possible that this issue only happens in the "Classic" player and not the Modern, but I'm not sure...
What version/Edition of SCORM are you using, what is the LMS you are using, and have you tested this on SCORM Cloud to see if you can replicate the problem?
SCORM 1.2 but again it works just fine for hundreds of other users in the same environment; attempting to replicate in SCORM Cloud would be impractical and inconclusive. But thanks for asking.
In my case it is also SCORM 1.2; Cornerstone is the LMS; it is only an occasional user that has the problem. However, the client has sent me the "bad" suspend data, so I suppose I could input it into the SCORM Cloud with the same course and see what happens there, to rule out the LMS.
Would the suspend data be the same string if using SCORM 2004 instead?
Suspend data is limited to 4,096 characters with SCORM 1.2
Suspend data is 64,000 characters with SCORM 2004, Ed. 3 & 4.
My client has enabled extending the suspend data limit in their LMS, and the length of the "bad" suspend data string is well below it. It's true that always choosing "Reset to Initial State" with slides reduces the length of the string, so maybe that helps avoid problems.
Was hoping to hear from Articulate. Maybe they'll see this eventually and offer something. Suspend data length and SCORM 1.2 is not the issue; it is way, way under 4000 characters. I tried to figure out a way to inject the bad suspend data into a cloud.scorm.com session with the same course but I just couldn't do it; it is beyond my browser-developer-tools skills; I know how to do it with my LMS ;-) Add another plea to the pile: Articulate, please give us tools to work with the suspend data. Add it to the pile of the thousands of requests you've already gotten :-)
I'm not sure what you expect Articulate to be able to offer at this point. As you stated, "this is a random occurrence for the occasional user, for an occasional course. For almost all users this does not happen for a given course." I helped Macromedia isolate an obscure stack overflow bug in Authorware which had plagued them through 3 versions, over as many years, because the problem would only appear under very specific circumstances. It was only because I was unwittingly pushing the tool in that direction were they finally able to see the bug manifest itself reliably. In other words, blind luck.
You have likely eliminated the suspend data length and SCORM 1.2 as culprits. The fact that you have back-end access to your LMS' database introduces the possibility of different culprits, including your own internal network; how users are credentialed; how sessions are created, managed, and destroyed; and communications integrity between the LMS application, the course, and the database. Does IT employ network packet-sniffing/traffic-logging tools which could be impacting the data stream? A friend introduced me to the Charles Web Debugging Proxy, which can provide reams of useful data. There are other packet analyzers including Fiddler and Wireshark which could also contribute useful information.
We haven't even touched on how your course is developed, what Storyline functions and features you either are or are not using (and what you're replacing those with), file-naming conventions, the use of JavaScript, or how the output is stored on the LMS. We also haven't touched on your affected users' configuration. Are they on-site or remote? What OS? Desktop, laptop, tablet, or mobile? What other applications are they running when a course locks up? Any VMs in that mix? Are there cloud-based apps (like Teams, Citrix, Box, OneDrive, etc.) competing for network bandwidth?
Have you been fortunate enough to have the debugger running when this issue manifests? What do you see in that file?
We have encountered the same issue. We are using SCORM 2004 (3rd) so it is not a suspend data limitiation. For some users the suspend data gets corrupt for some reason. The training has a knowledge test in the end. It seems like the problem usually occurs when the user exits the knowledge test in the middle and then resumes. We have tried to force the test to restart using variables, but this does not work as it seems like the SCORM is "locked". Have someone found any fix for this issue? It here a way to set "Prompt to resume" only for certain slides?
Hello Jonas,
If you're encountering LMS related issues, here are a few troubleshooting steps that you can try:
For additional assistance, you can reach out to our support team through a support case, and someone from the team will be in touch shortly.