Failed To Apply Patch For Package

0 views

Skip to first unread message

Ellyn Krucke

unread,

Aug 3, 2024, 2:33:28 PM8/3/24

to lyagranamlo

Hi @PaulLeBlanc1 . I am having the same issue. Below is the error message. I'm trying to apply the asset package to my test enterprise DB at version 5 of the UN. I see my asset package was 2.7 should I download the newest one for Electric and do you think that will solve my issue? I just upgraded to Pro to 2.9.1. I have run this asset package on a fgdb and it worked. Thanks for any help.

Hi @PaulLeBlanc1 , I tried running it with UNtools 2.9.1 but still get the same error (see below). I did not have the transmission tier checked or add data. Also, I am running the 2.7 asset package schema should that matter? I was trying not to have to go through downloading the newest one and configuring (e. g. reproject) the asset package. Any further thoughts? Thanks, Wendy

I am using ArcGIS Pro 2.7.3 to try and apply an asset package to a newly created utility network that I staged in a newly created enterprise geodatabase. I currently was trying to load schema only to test as the dataset is pretty significant. I am receiving the following warnings and errors:

Without seeing the data it's super hard to provide advice. But I would start with what fields am I trying to write TO and where is that data coming FROM. What type of field is it? Are there special characters in the physical information in some of those fields...the"/" indicates that there might be. This is just the surface of the problem. I am not sure if this is a given ESRI asset package that you loaded your data into and how did you load the data in as well. Look here: -loading/ To get started with data loading. FYI this process took us 6 months to work through to get our data loaded in successfully.

Thank you very much for your time and advice! This asset package was supplied to me from a client so I can then publish it to my Enterprise. I'm not sure what steps they took. I guess I will need to work with them to clean up the data before attempting to export an asset package that I can use.

From what it looks like, there is something going on where something is expecting to see some Editor Tracking fields and it cannot find them in the Asset Package. Is Editor Tracking perhaps enabled on the Enterprise Geodatabase you have your empty staged UN? If so, maybe try disabling it and then applying asset package?

Process 10/12 of the "Apply Asset Package" geoprocessing tool enables editor tracking. I did try to disable editor tracking on the dataset. When I did the dataset expanded, within the catalog pane, and it appeared as though the schema fully loaded in successfully even though the process said it failed to complete. I'm going to see if I can achieve similar results but with attempting to load in the data this time instead of schema only.

As you suspected, the process did finish correctly. What failed is an altered field to adjust the alias of the editor tracking fields. I would have to look up when the change occurred, thought it was 2.8, but when creating an UN, it no longer adds editor tracking fields. The Apply Asset Process was assuming these fields were added by the UN tools and making an Alter field call to adjust their alias to what was in the Asset Package. The newer version of the tools handle this change, but I do not think that was ported to the 2.7 version of the asset package tools.

I am not sure what is going on here. Something is not being applied correctly. Did you start with a new postgres gdb or was this an existing GDB? I see you have maps open, we have seen where these create locks on the Utility Network and cause corruption. Can you close all the maps, restart pro and apply the asset package to a new Postgres database?

Last thing I just tried is opening a blank Pro document, add just in my connections in and ran apply asset package. So far, always the same error. It gets to the UpdateSchema tool and gives errors of "ERROR 001958: No valid asset group was found. and also WARNING 000952: Failed to execute (AssignDomainToField).

It works going to an empty GDB but fails on the PostgreSQL db.

Yep. Brand New PostGres GDB. I'll try the copy/paste method and see if i run in to any issues.

I have these Utility Networks up and running on my DEV stack. Just trying to migrate them to Production by >>> Staging a Utility Network >>> Apply Asset Package generated from the Dev versions. I'm open to all options.

If you do the same process to a Mobile or File GDB, does it succeed? In the asset package folder, there is a ap_workspace folder, can you send me the gp log file? Could you send a schema only asset packge or UN that I can try to repo the issue? mmi...@esri.com

I don't understand why it's giving an error - isn't the point of the WMI filtering for it to go "ok, I got a 0 result on that query, onto the next step"? There is the option for "Continue on error" but no guides or information on this I have found describe having to enable this option.

Another question I have is - I'm applying the driver packages provided by HP - will this install only the required drivers from the package required by the PC or will it apply possibly wrong or unrequired drivers. The package contains drivers for say optional video cards that our models don't have - is the task going to install those drivers and say the nVidia control panel and such? Do I have to manually pick out the drivers that are unrequired prior to distributing the package to our DPs?

It still shouldn't error on the query itself should it? If the model I've given is wrong shouldn't it just fail on each query and continue? But in my case the Task Sequence fails and reports an error on the first of two queries.

I did originally test by using what a "WMIC computersystem get model" reported back within "%%", ie "%HP Compaq Elite 8300 SFF%" but received the same errors, only fell back to what I have above to test incase it was what it was querying.

Just to conform ran a test run, all appears successful now. only required drivers installed, no ATI or nVidia stuff installed as I feared either. I'm going to leave the WMI model query as is - just the number ie "%8300%" - I've found from getting the model on different HPs the order of words changes, ie:

It appears would be easier if it works (and it appears to without issue) to just stick with the model number itself, not the full model name to avoid issues in future when the next model changes the name order yet again. Sound like a plan or you think it is best to stick with the full name? We will only have 6 different driver packages in place for our environment.

If there are none, or none look related, then it's probably something else. Check carefully the output of the command you were trying when you got the error message, as there may be other clues in the full output from that command, aside from the error message.

Occasionally aptitude will be too eager to remove or downgrade large numbers of packages to satisfy your request, in which case retrying with -f changes its priorities and helps it come up with solutions that involve removing/downgrading fewer packages even if it means not all changes you requested can go ahead:

Edit: this is an old answer, and since it was written a newer APT front end, simply titled apt, has become the preferred command line APT interface for end-users. It is still not the Swiss army knife that aptitude is, but it's a bit more human-oriented than apt-get. I haven't taken the time to investigate how it would help in the above scenario, but it's worth using.

A Unable to correct problems, you have held broken packages error message often occurs because you are trying to install or have installed a package that you're not supposed to install. These are some fast and easy ways to fix the you have held broken packages error.

Open your sources.list file in /etc/apt/sources.list and check that there aren't any software sources for a different Ubuntu release than the Ubuntu release that you are currently using. If you find any incorrect release lines in sources.list, open the sources.list file with sudoedit /etc/apt/sources.list, comment out the incorrect lines in sources.list by preceding them with a # character, save the sources.list file, and run sudo apt update to update the list of available software packages.

If you have manually downloaded a .deb file simulate installing it by running apt install --simulate ./downloaded-deb-file.deb. If downloaded-deb-file.deb cannot be installed because of unmet dependencies the results of apt install --simulate ./downloaded-deb-file.deb will list these unmet dependencies. Since the command is a simulation you don't need and shouldn't use sudo or run it as root. Note the version requirements of each unmet dependency and search the web for a .deb file(s) that meets these version requirements. If you have found a .deb file that meets its version requirements, download it and run apt install --simulate ./new-downloaded-deb-file.deb to check if the newly downloaded .deb file has any unmet dependencies of its own.

In Synaptic in the left pane click the Custom Filters button which is marked by the mouse cursor in the below screenshot. From the list in the top left corner select Broken. In the center pane will be listed any broken packages that still need to be repaired.

Select the broken packages one at a time. Select a broken package, and then open the terminal and run apt policy . The results of this command will tell you if that broken package was installed from the default Ubuntu repositories or from some other source. If the broken package was installed from some other source, maybe that package can be removed along with its software source and replaced by a different version of the same package from the default Ubuntu repositories. Usually this means fixing a broken package by downgrading that package to an older version.