The entire process for Workspace upgrade has been a challenging in my experience. OPM is scheduled for 2 version updates every year, and with each upgrade, users are required to perform a Workspace upgrade, and verify all customization are functioning with existing DGN files. By that logic, Admins would need to spend all year testing if OPM functions as expected after each upgrade.
OPPA has proven to be lacking in its ability to automatically perform a smooth workspace upgrade of 100 percent of user configurations.
We came across several issues upgrading from U8 to U10. I have tested the workspace upgrade using OPPA U11, and still see that there is a portion of customization being missed.
Let's do some math:
Average Piping Component Classes = 50
Average Properties of Piping Component Classes = 20
Average Custom Attributes of Piping Component Classes = 3
Total Items that need to be reviewed during an upgrade = 3000 classes, props, custom attributes need to be reviewed to validate if all customizations have been retained after a workspace upgrade.
Is this a reasonable expectation?
How would you feel if after a Windows update you lost about 5% of your customizations and files, and was expected to comb through every single function to find out what was missed?
It has proven difficult to resolve open product defects if the solution only comes in newer versions of OPM. Yet, the workspace upgrade process is dependent on a complex process.
It would benefit all OPM users if the upgrade process was simplified, and made more robust, so users aren't expected to perform a full product evaluation to verify if all previous customizations haven't been missed or are causing issues in newer versions. I would like to see this improved.
| How frequently will feature be used? | Once a month or longer |
| How much time will you save? | 100000 |
| Your Industry | Oil & Gas |
I agree with you 100%, upgrading is a very time consuming process and prone to mistakes which sometimes takes a while to show its ugly head.
This could be so much simpler if the Schemas are more open to user customisation. Imagine the user does all of his customisation in a set of schema files that do not affect the Bentley schema files, in other words, the child overrides the parent. When a new release comes out you only copy across your schema changes, actual files no comparisons. You then have to deal with things Bentley might have deleted, which does happen but in all my years (since the early 90's) I have not seen that happen all that often.
The comparison tool has never really helped me too much either, I tend to document my changes and then just redo them when a new version comes along, that way at least you can see what is going on. It might take some time, but I have found it much better than trying to work through those thousands of red, green and blue marks which can be a challenge to interpret.
Hoping that there is an easier solution on the cards soon, but again I have been at this long enough to know that it will take a long long time to see it happen, if ever.
This looks huge as part of improvement!
It is always been pain for moving from one release to another. Doing workspace migration is not that much huge concern but when things does not work as expected, this creates a huge issue.
Spent good amount in workspace upgrade and with upgraded workspace observed that dgn is not behaving as expected.
Example:
Properties in U8 shows fine without any issues to sheet drawing. Based on revision tag is modified and now with the upgraded workspace it is not showing the updated tag, it is huge task if it requires to redo the entire sheet cutting.