After upgrading the Personal gateway to the june version blank date fields gets imported as 1899 dates rendering my reports broken (since I filter on e.g deleted_at = blank).
Swithced back to older version (Apr release) returns import to normal behaviour.
This is an on premises scheduled refresh from an AWS RDS ProstgreSQL DB.
Any idea on how to correct this?
We have created an iframe using "Publish to Web" feature and tried to embed it in SharePoint. This is working fine with chrome/edge browser, not in IE. This issue has started on last month ( May 2017).
I though this was fixed in the last release and seemed to stopped happening after I downloaded the newest version. Now it is back and is a complete pain to deal with. Please fix this issue.
I'm a long-term user of Power BI and would expect to be eligible for the Extended Pro Trial (I've received the notification e-mail about the extended trial). Today, when I try to share a Power BI dashboard I receive a message telling me that I must be a pro user to do that, the message invites me to 'Upgrade Now'. When I click 'Upgrade Now' I'm shown a page of options, and select the middle one to 'Share and Collaborate' which it says I can try for free - I'm guessing this is the extended pro trial. Selecting this option leads me to another message that this says I don't need to sign up for pro because I already have a licence. I'm then trapped in a cycle whereby I can't share my dashboard because I apparently don't have a pro licence but then when I try to upgrade it won't let me because it says I've got a pro licence already. Looking at my stoarage settings I've got 10 GB assigned which suggests pro but at the same time I'm described as a 'Basic User'. I'm completely confused and haven't been able to find anywhere I can contact Microsoft for help. Can anyone help me? Thanks
I have a requirement to emebed Power Bi report in a Html file hosted in SharePoint Online site. The content was displayed properly in all browsers initially.Suddenly we have observed issue in IE browser and it always dispalys blank.The other browsers work as expected,It would be great if any one helps to sort it out.
Image link: http://imgur.com/HqCky3f
I have pasted a link to the image above, I could not embed it for some reason.
I am pretty new to PowerBI, so be nice. I have loaded some geographical data against the UK using a postcode file on a map.
I get a yellow warning error bar at the bottom of the ESRI/ ARCGIS map that tells me: "Failed to Locate 23 Features".
I cannot work out how to interrogate, review or fix these 'features'.
It would be really helpful if error reporting was able to be scrutinised a lot more clearly for a novice like myself.
Any comments and support would be welcome.
This morning, when I started Power BI Gateway - Personal, it failed with the error, "GWConf has stopped working A problem caused the program to stop working correctly. Windows will close the program and notify you if a solution is available." I uninstalled and re-installed the application but am still getting the same error. The Personal gateway as worked flawlessly for weeks prior to today. I'm not aware of any changes being made to my PC but I'm using a company computer and sometimes our IT department pushes out updates without prior notification. Where do I begin looking for the cause of the failure? (I'm just a Power BI user, not a developer.)
We have currently a project, that should deliver several PBIX-Files to different customers.
Every PBIX-File should access the same SQL Azure Database, but should use different SQL Login-Credentials (since not every customer should be able to access all database objects). If we now wanna test this on our local environment, and create these PBIX-Files with different SQL Logins, it always overwrites the SQL Logins of the global Datasource. This is indeed annoying, but can be worked around by just entering the logins again and again before publishing.
The real problem shows up when we publish these PBIX-Files to Power BI Web and try a scheduled refresh of multiple datasets:
Since the SQL Login gets always overwritten as soon as you publish a new PBIX-File, it kind of uses a random SQL Login, so we cannot guarantee to provide the correct data to the original files.
Our question now is if it's really fully intended to always overwrite the logins of a global Datasource, even in different, shared workspaces or will there be any solution in the future to store the logins with the files or provide a section to manage the Datasources properly (maybe analog the SSRS-Datasources)?
Not sure if this is related to the data refresh issue, but starting yesterday I have two different external customers (both in California) that view different dashboards. They are able to get to the dashboard frames, however get an error stating that "something went wrong.. unable to load the data model".
Anyone else experienceing this with external users?
Our data is on prem, we use the personal gateway, and the external customers are using the link provided in the original MSFT email I shared with them.
I've been working on a data set for a few weeks that maps the top 10 records by zip code in the Map vizualization. This data set was working up until yesterday (2/13/2017). Today, 2/14, the power bi embedded version no longer maps the zip code data.
I downloaded the Feb 2017 power bi desktop update, and now I have the same problem in the desktop version. I went back to the January update - but the mapping there is also not working.
Did something break in Power BI? Or bing maps?
Please help, as this problem will completely derail a customer initiative.
I have a report with a couple of slicers and the Matrix Preview visual which work fine together in the Desktop client, but the filtering gets "stuck" when viewing/interacting with the report in the Power BI Service. The slicer will often filter the data upon first selection, but when the selection is unchecked, the data in the Matrix Preview visual does not return to its unfiltered state. When the "eraser" button in the header of the slicer is clicked, this DOES sometimes prompt the data to revert to its unfiltered state.
Similarly, sometimes when a selection is checked in the slicer, it does not filter the data at all. Clicking the "eraser" button and then re-attempting to use the slicer sometimes fixes the issue and enables the filtering of data (but the filtering remains stuck and won't "unfilter" until the eraser on the slicer is clicked).
Here are some snapshots:
2) Slicer Selected (filtering as expected in this case, although this is not consistent either):
3) Slicer Unselected (still holding filter from when slicer was selected):
I use custom visuals such as Smart Filter (SQLBI) along with Table Sorter.
Until now everything was working properly, for months.
I now encounter rather strange effects.
Here is what has happened and how I proceeded (as well in PBI Desktop as in PBI Service).
STEP 1 - First filtering with Smart Filter. Results in a Table Sorter visual. Correct.
STEP 2 - Second filtering (different as previous). Results in the same visual. Correct.
STEP 3 - Third filtering : the same as in step 1. Same visual: results partial, i.e. missing rows in the resulting table.
In step 3, when - in Desktop - I switch from the "Table Sorter" visual to the "Table" visual (or "Matrix" visual), results are correct again. And when I come back to the "Table Sorter" visual : same missing rows situation !
Can someone help?
I have several dashboards that I have implemented the new Matrix preview. The funtionality provided by the preview fits the usecase very well. However, I noticed that when I have one or more matrix preview visuals with crosshighlighting enabled, filtering on one visual will filter the preview matrix initially, but subsequent filters sporadically update the matrix or don't at all. Hitting the 'Refresh' button corrects the issue. I had not encountered this issue with the preview until recently. The only solution is to revert back to a table or matrix but I loose the cross-highlight/cross-filtering ability of the new matrix. Can you see if you can address this issue as soon as possible. Thanks.
I noticed today that ALL of my attribute slicers are now set to 'highlight' other visuals rather than 'filter' them. I didn't make this change. They have been set to filter for 2 months. Perhaps this was triggered by the January update? Lots of rework involved to set them up to filter again.
Matrix Preview.. If we set the word wrap on values for a given column. The column word waps should be maintained when the column series changes.
If multiple years are selected
The columns in 2016, should have maintained the 2017 column settings. The totals should behave the same way too.
I have been awaiting this feature for a while now so I was relieved to see this implemented in the latest update.
However, I think I have found a bug. The word wrap resets when sorting on a particular column header. Please can this issue be investigated & rectified.
Also, could an equal column width setting be implemeted to satisfy my OCD requirements
I have dashboard that links to a report with 5 tabs. I have shared this dashboard with users both within my orginazation and outside of it. I have been having issues with users (both internal and external) not seeing all 5 tabs. They are only seeing 2 of the tabs.
I have tried removing their access, and then regranting access and sending them a new link, but that does not solve the problem. They have logged out and logged back in but that does not solve the problem. They have tried accessing the dashboard from the share dashboard URL, the URL in the invite email, and by logging in direct to app.powerbi.com and then navigating to their dashboards, but these do not solve the problem.
Yes, I read the article on going to the support site and submitting a bug report. I'm not a Pro user, so I don't get free support and I'm not about to charge my credit card to determine this is a bug and then have the charge refunded.
The Assume Referential Integrity option in the model does not work. I have 2 tables OrderHeader and Payment. They have a foreign key defined as OrderHeader.OrderID = Payment.OrderID. This foreign key is active and enforcing the relationship, so it is impossible to have a value in Payment.OrderID that does not exist in the OrderHeader table. However, when I select Assume Referential Integrity, I get an error back that there are rows in the Payment table that do not exist in the OrderHeader table. This is false.
The code that is being used is invalid to perform this type of check. It ran the following 3 queries:
(select [OrderID] as [OrderID],
--additional columns deleted
from [dbo.][OrderHeader] as [$Table])
(select [PaymentID] as [PaymentID],
[OrderID] as [OrderID],
--additional columns deleted
from [dbo].[Payment] as [$Table])
(select [OrderID] as [OrderID],
--additional rows deleted
from [OnlineSales].[OrderHeader] as [$Table])
It appears that instead of checking to see if there are actually values in the Payment table that do not exist in the OrderHeader table, it is instead doing a pair of counts, comparing the values, and if they do not match, saying that there is no enforced referential integrity. This is invalid. This set of queries is saying that there MUST be a payment for EVERY value of OrderID. It does NOT say, there must be an OrderHeader.OrderID value for every Payment.OrderID.
We have production implementation of Power BI, which contains various charts that have an interaction of slice set. Since the latest Power BI release, when clicking on a part of the bar graph in one chart all the bars/parts of the entire sibling chart becomes filled with the same value.
Notice that the 17K red portion of the chart on the left for 2017 = 43% of the total. Clicking on the part of the bar yields:
I would have thought that the red 43% portion of the 2017 bar would be highlighted with no changes to the other bars/parts.
Going back and using the January, 2017 release I get this:
OBviously this interaction is not very valuable and I could just turn off interaction for these charts but there are other charts on the same report that are also not slicing correctly.
We are using Compatibility level 1200 for the tabular model, so SSAS 2016. This is just a numeric attribute and the aggregation we are using is Median.