nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

diff&merge unresponsive

medini_1036016653

Dear ANSYS medini support, currently I'm facing an issue trying to port back work I made in a secondary project using diff&merge. The dialog is hanging without progress (I waited for 15mins+), CPU utilization seems to be one core 100%. The debug window does not show any errors: I tried returning to all standard parameters (regarding garbage collection, max. allowed RAM, debug option), but that also doesn't help. I also tried 2-way and 3-way diff&merge. Could this be related to the error message I received while working on a script? (see https://medini.freshdesk.com/support/discussions/topics/1000086044) I'm dealing with merging back work from the project that exhibited this error message yesterday. Is it possible that the d&m algorithm is stuck at this now "invalid" relation? Thank you and best regards Peter

Find more posts tagged with

medini

medini legacy

Accepted answers

All comments

medini_1036016653

I manually took over the modified script and reran it on the source project. That one still behaves fine, so that hints at my assumption that the improper deletion of relations (see linked thread) led to an inconsistent model. Obviously, this should not happen, as I might have a lot of other unmerged work in that second project. Exporting, deleting, importing the problematic project did not help and thus the issue should be reproducible at ANSYS. Are you interested in an export of the problematic project (NDA restrictions apply)? Best regards Peter

medini_1024120963

Hi Peter, do you use the Eclipse internal SVN for version control? It might happen that the issue you talk about have been caused there. NDA applies. Can you send it therefore through the support channel? This needs to be investigated from our 2nd layer Support, I guess. BR Joël

medini_1036016653

Dear Joel, I stopped using the Eclipse internal SVN for version control as it does not handle 3-way diff&merge. I'll contact support via email. Best regards Peter

medini_1036016653

I've sent the email: Dear support, as instructed by Joel Heinemann, I contact you directly. I uploaded the problematic project to the page https://downloads.medini.eu/products/main/downloads?folder=myDocuments of user ID 606797 with file name scriptFTA-AET_v18_NO_DIFF-MERGE.mprx (NDA restrictions apply. The project contains sensitive information.) Please read the link ZF support thread for additional information. Opening the mprx file in v2019R1.1 works just fine. Just to prevent questions: we are currently stuck at 2019R1.1 due to compatibility considerations (project is worked on by multiple parties.) Thank you and best regards Peter

medini_1024169437

Hi Peter, without looking at the uploaded model (Joel can handle) I would like to mention an option in the diff/merge which is by default enabled but may slow down the comparison dramatically: Please disable that option and try again. Computing all differences also for "derived" attributes might be very slow. Hope that helps Jan

medini_1036016653

Dear Jan, thank you for the suggestion. The same suggestion was given by Michael Soden to me at a face-to-face workshpo at ZF Schweinfurt, and consequently I already have this setting unchecked. Thus, this seems to not be the cause. My current assumption is that the diff&merge is stuck in some kind of infinite loop. Best regards Peter

medini_1024169437

Ok, <?WORD FLAGGED?>. Joel should pass this to our testing team to reproduce than. Have a nice WE Jan

medini_1036016653

Dear Nino, as discussed, this is priority #2 (#1 is FTA cut-set analysis performance). Best regards Peter

medini_1036016653

This ticket is still relevant with release v2020R2.