Yesterday, I was on PBI Desktop Version: 2.67.5404.581 64-bit (March 2019). Received an update notice at launch today. Download page and "what's new" info looks the same as x.581 version, but after download and install, I now show Version: 2.67.5404.801 64-bit (March 2019).
.581 had at least 4 bugs that I dealt with:
SO - I am hopefull that version x.801 contains fixes for some or all of these frustrating bugs....but how the heck can I tell? I was unable to locate any sort of release notes or documentation explaining the diference between .581 and .801. The UTC bug is a pain. Is the expectation that I need to test MSFT bug fixes? I might be ok with that if I could get hint about what is in the release.
Hi @HAL9000 ,
I didn't find any blog that lists all fixed bugs in a new release. On this support page, Microsoft will list some known issues, you could keep an eye on it.
For the first three bugs you mentioned, I was not able to reproduce on my side. Do these problems persist now? If yes, you could report an issue.
For the forth one, you should be aware that the time zone on Power BI service is effectively always set to UTC. To get the correct local time, we can add time difference in your DAX formula, or use Power Query to get the local time zone datetime.
Thanks for the reply. Still validating the latest release to see if the PBI desktop failures persist, that's why I was hoping there was some sort of resolved issues list. Apparently not.
On the time zone issue, the second link you provided throws an access denied error:
We already calculated the time zone we need to appear on the published report - reproduced error by creating a simple pbix file with 2 cards showing their own measures. One measure is UTCnow(), the other subtracting the difference from same. When published, both default to UTC in PBI service.
I'll look into using some of the exact DAX statements provided in the Radacad link to see if that resolves. As it stands today, it is just another strange PBI issue that doesn't seem logical or straightforward. Although a frustrating bug, I was hopefull that the "UTC or nothing" approach was an indicator of some actual localization coming to PowerBI. I guess it was just easier for them to deploy in the current state and let users figure it out on their own.