search

lzim / teampsd

Team PSD is using GitHub, R and RMarkdown as part of our free and open science workflow.
GNU General Public License v3.0
15 stars 23 forks source link

ASAP Bug: mtl.how/data 626 Facility File sta6a list in vizDiag graph not from 626 #1639

Closed lijenn closed 3 years ago

lijenn commented 3 years ago

Went to facility file for 626, opened the file, reviewed data and did not correspond to 626.

Lindsey will post screenshot below in a comment.

anthonycpichardo commented 3 years ago

Acknowledged.

Will take a look now and resolve.

lzim commented 3 years ago

@anazariz & @anthonycpichardo

data_facility_file_wrong
  1. We identified sites from New York (not Tennessee) in this list: 626_sta6a_error

Please provide an estimated timeframe to resolve ASAP.

anthonycpichardo commented 3 years ago

Sites where the vizdiag tab has incorrect data 666 519 589 657

For 589 & 657, the divisionname list is the same as the 626 site above.

lijenn commented 3 years ago

Thanks for checking the scope of the issue @anthonycpichardo!

To summarize our andon alert/Fire drill practice as a team from team temp:

  1. Scope - impact on current wave sites. Scope of our active sites that may have been impacted by the andon alert and alert HQ within 30 minutes.
  2. Alert - on platforms. Alert the andon message to partners and on platform (mtl.how/data in this case) - Every team member should know how to send out or set up andon message alert.
  3. Communicate - until resolved. Communicate with sites and partners until issue is resolved.

@lzim @staceypark @anazariz @dkngenda

Should HQ communicate to the mapped QIIC and site leadership? A message like: Your site has been impacted by a bug causing incorrect facility data. We are not sure what went wrong, but are actively working on fixing this bug. We will keep you updated, thank you for your patience.

anazariz commented 3 years ago

@lzim @staceypark @lijenn @anthonycpichardo

I completed the fix. The file to be tested is called SplashMothership-debug.xlsb and it's in the quant's splash_pages folder.
I recommend that 3 of us perform independent testing. We need to test the facilities that had the issues (all 4 vis sheets and all 4 data sheets) as well as 4 good facilities (select on random) as a from of regression testing (making sure that nothing good is now broken).

lzim commented 3 years ago

@anazariz - great!

Do you mean that you have a "fix" that you have already propagated globally? Or, a fix that applies to our 3 top priority sites 626, 657 & 589?

@anthonycpichardo - Let's update the Alert message once we test. @staceypark and I are testing now. Are you able to too, Anthony?

lzim commented 3 years ago

@lzim is reviewing 626, 657 & 589 this morning for leadership coplanning.

@anthonycpichardo & @anazariz will work to test and implement the global fix today.

lzim commented 3 years ago

FYI @anazariz @anthonycpichardo @staceypark @lijenn @dkngenda

How is testing going? I'd like to briefly huddle about the issues and wonder if we should meet earlier ~10:45AM than 11AM quant_operations to do so??

I was just reviewing the 666_facility.xlsx file with leadership in SLC and reproduced the error, it has 646 data in the viz tabs.

I noted several things that I want to review before propagating anything, which are: 1. vizEnc, vizHF and vizMeas tabs do NOT have the "divisionname" grey butto:n on their graphs (like they should and like vizDiag does). Why?

  1. I note that I don't see a "Control" tab, which I was looking for when this error first was observed, because that would help me see what facility the data was populated from. What have decided to do about that? I note that Ash describes fixes about deleting "hidden" elements, so I want to check on that.
  2. I want to make sure the scope of our problem(s) have been carefully checked (e.g., across the "locationname" data).
  3. Why are there grey buttons in the viz graphs that are "inert" (they don't do anything) like "Sum of PTSD." These are not familiar buttons to me in terms of using them.
  4. I also want to cross-check this pre-propagation with VVC #1638: https://github.com/lzim/teampsd/issues/1638
lzim commented 3 years ago

@anazariz

facility.xlsb file - WILL DO Friday 1/15/2021

WILL WAIT on this until next week. detail_facilitate: (for MTL Manual - Facilitator section).

document_team (for MTL Manual - Technical/Admin support section).

lzim commented 3 years ago

This is still an ASAP bug and we are propagating January next week with several facilities leadership coplanning meetings underway.

Errors I'm getting working with the 657 facility file - The file has errors upon opening that crash Excel. I have another Excel file open here, and you can see the errors referencing the 657 file. 657_errors

When I click on "Open" next to the red repair message I get this error message: image

anazariz commented 3 years ago

@lzim I am working on it. This is not code related otherwise it would be happening with all the files. This is related to the file size and the platform not being able to handle these volumes. I will refresh this one file.

lzim commented 3 years ago

@anazariz Thanks will you update when the refresh is complete?

anazariz commented 3 years ago

@lzim I refreshed 657 and verified that I could open it.

lzim commented 3 years ago

Thx! @anazariz testing now

lzim commented 3 years ago

Scope - #1638

Fresno 2/3 is Dr. Norlander is "super user" might test the VVC views and fiscal year.

Facility

Team

anazariz commented 3 years ago
  1. Data UI Fiscal Year and VVC coding was completed. I'm waiting to get the test results back and OK'ed from Jenn
  2. Facility VVC has two components. Location: splash_pages\Dev 1) change to the monthly script splashSourceIncrementalUpdate_vvc.sql 2) change to the workbook SplashMothership_vvc.xlsb
lzim commented 3 years ago

Discussed at hq_workgroup wk1 feb_epic