Select to view content in your preferred language

Use Electric 3.4 with ArcGIS Enterprise 10.8.1 / Pro 2.6?

1331
7
08-11-2020 03:59 AM
SergueiSokolov
Regular Contributor

While we are waiting for the release of a new version the Electric Utility Network Foundation template for Enterprise 10.8.1....Is there an upgrade process for the template v3.4 to work with ArcGIS 10.8.1/Pro2.6?

We have configured ArcGIS Enterprise 10.8.1 (ArcGIS Pro 2.6, Oracle 19c) and tried to load the Electric v3.4 template. Apply package with loading the sample data failed with the following message:
Current state: Run apply Package - failed\
The asset package has the following warnings (1/1):
Item Field OID Value Message Level
0 ElectricCrossarm length None None This field name is reserved on some database platforms warning
C:\Users\XXXX\AppData\Local\ESRI\conda\envs\arcgispro-py3-clone\Lib\site-packages\pandas\core\frame.py:1487: FutureWarning: Using short name for 'orient' is deprecated. Only the options: ('dict', list, 'series', 'split', 'records', 'index') will be used in a future version. Use one of the above to silence this warning.
FutureWarning,
Failed to execute (AssetPackageToUtilityNetwork).

Should we use ArcGIS Pro 2.5 (untools2.5?) with ArcGIS Enterprise 10.8.1 to load & configure the Electric template 3.4? Again, we understand that the best approach is to wait for the next release of the Electric template, ... just want to explore options to move forward now with the sample data set and configuration from the Electric template v3.4.

0 Kudos
7 Replies
MikeMillerGIS
Esri Frequent Contributor

There is no need to wait for the new template.  We are not planning on releasing it till end of year.  You can use the existing solution to apply to a version 4 utility network in ArcGIS Pro 2.6. You will have to adjust the maps slightly, but the asset package will apply with no issue.

Can you verify the version of untools?  If you installed 2.6 on top of 2.5, I would suggest deleting your cloned python environment and install untools into that new clone to ensure you have the 2.6 version of the untools installed.

0 Kudos
SergueiSokolov
Regular Contributor

Thank you for a prompt reply. This was from the scratch installation of ArcGIS Pro 2.6. I have cloned python environment (no errors) and installed the version of 2.6 of the untools.

0 Kudos
MikeMillerGIS
Esri Frequent Contributor

could you try this?

Open Pro

Open the python window

type the following:

import pandas
pandas.__version__

Post what version of pandas is installed?  

0 Kudos
SergueiSokolov
Regular Contributor

import pandas
pandas.__version__
'1.1.0'

0 Kudos
MikeMillerGIS
Esri Frequent Contributor

It appears pandas was updated to 1.1.0 2 days ago.

https://anaconda.org/anaconda/pandas/files 

Here is the version histroy:

Release notes — pandas 1.1.0 documentation 

Looks like we are running into:

DataFrame.to_dict() has deprecated accepting short names for orient and will raise in a future version (GH32515) - What’s new in 1.1.0 (July 28, 2020) — pandas 1.1.0 documentation 

 We will work on a fix on our end, but in the mean time, can you try to downgrade pandas?

In Windows, Click Open - Python Command Prompt

This will open a command window pointing to your cloned python environment

Enter the following command:

conda install pandas=1.0.5

Click Yes to any prompts

0 Kudos
PaulLeBlanc1
Esri Regular Contributor

untools 2.6.0 build 1 is now available.You can uninstall and reinstall untools to pick up this change.

0 Kudos
SergueiSokolov
Regular Contributor

Thanks for sharing the fix.... It solved the issue.

Please note, as suggested, I have uninstalled untools2.6 in ArcGIS Pro. Validated that untools package was uninstalled. Refreshed packages and tried to install untools 2.6.0 again but it was still showing the previous version "pandas>=1.0 <2.0", so I've completed the following manual steps :

downloaded the suggested package untools-2.6.0-py36_1.tar.bz2

conda install "c:\temp\untools-2.6.0-py36_1.tar.bz2"

checked: pandas.__version__
'1.1.0'

so, I had to install 1.0.5 (as suggested)

conda install pandas=1.0.5

0 Kudos