I'm trying to set up a NuGet server using ProGet and am hitting a brick wall when publishing a package and it doesn't appear in the feed. The package is written to disk and works in other NuGet feeds. Other packages also don't appear in the ProGet feed so I'm pretty certain that the package is fine and that the problem lies with ProGet.
I'm using the community version of ProGet, but I don't see why that would affect anything.
Any ideas are most welcome!
The ProGet service is responsible for indexing packages, so if it's not running, packages could be uploaded but not displayed in any feeds. Here are the common troubleshooting steps for this scenario:
Verify that the ProGet Windows service (INEDOPROGETSVC) is running.
Ensure that the user account hosting the ProGet service has access to the feed storage path. Since it is NETWORK SERVICE by default, it would not see your mapped drives, and may not have access to the UNC path where the packages are stored.
Try running the ProGet service interactively, i.e. stop the ProGet Windows service and run ProGet.Service.exe manually as a console application to see any live output. Remember to restart the ProGet service when you close the console application.
Check for feed indexing errors to see if there was a problem indexing a particular packages. I know in much older versions that single "poisoned" packages (bad .nuspec file, invalid directory structure, etc.) could halt the indexing altogether.
Thanks John Rasch, found that your first point gave me what i needed to look in the right direction.
I could not find the (INEDOPROGETSVC) service but I did find (ProGet Servie) - I restarted this and then refreshed the my feed and all the missing packages showed up.
My version of ProGet is v3.3.12
Thanks John.
Related
Recently I restarted my AWS instance and got a new IP address but after I restarted both Jenkins and AWS, the information about my previous jobs was no longer shown in Jenkins.
I checked the path and it still exists in the instance but it is not shown on the web. I tried to create another project and it still created in the same path just that only the newly created project is in. Any suggestions on how to recover my missing projects??
FYI
I have lots of old plugins that mentions "xxx failed to load" so I do not know if that is causing it.
one of my plugins does not match and all those that depends on it will fail to show on the installed section of the plugin. Thus I remove all the plugins by deleting it directly from the plugin folder and check for the working copy that was on my previous version and download the same version of plugins. After which, all the jobs come back on screen
We are migrating some code from CF 10 to CF 2016 virtual machines that needs to connect to a couple Access databases. I installed the necessary drivers and setup odbc datasources in Windows but have found that our CF 2016 VM was setup without the ODBC service and I have not been able to find a clear way to add it.
When I go to add/remove programs the only option is to uninstall CF and when I run a CF 2016 installer it will not let me go through the "Server configuration" process because an existing install already exists. Is there a standard approach for adding sub-components to a CF server that were not chosen on first install?
Looks like I found my answer after digging around for a while. The migration wizard needs to be run again to add the ODBC service but there could be various complications with this depending on permissions and other factors. I was not able to get this to work after a brief period of time so I am just going to wrap up my transition to using python for our limited Access needs but I do believe this to be the answer.
https://community.adobe.com/t5/ColdFusion/ColdFusion-11-ODBC-service/td-p/6207226
Here are the basic steps. See link above for various troubleshooting info from Charlie Arehart and others.
Navigate to adminconfig.xml at C:\ColdFusion11\cfusion\lib\ and open it with text editor (say notepad).
Change the value from “false” to “true” in runmigrationwizard
Change the value from “false” to “true” in odbc
Save the file and restart ColdFusion Service.
After restarting the service, access the cf admin url and you will get the migration wizard. Follow the onscreen wizard to continue.
I came upon the answer by way of a thread about getting Access data sources setup in CF11 and CF2016 which Charlie Arehart contributed to and linked over to info about the ODBC service setup.
I was able to create an offline installer with a command line like:
vs_enterprise__873301792.1489161815.exe --layout %CD%\vs2017offline --lang en-US
I got about 20 GB out of this (1901 files).
but, as it is already known, the Visual Studio installer needs an internet connection at least to initialize, even with such an offline layout.
(But it doesn't need to download the packages already existing in the layout/image, so it saves really download volume.)
In the MS web documentation for VS more parameters are mentioned:
https://learn.microsoft.com/en-us/visualstudio/install/use-command-line-parameters-to-install-visual-studio
AFAIK, nobody figured out how to use them to guarantee really an offline install without internet connection (it seems quite likely to me that channel and/or catalog have to be redirected for hindering VS installer to go online).
Has anyone?
Here are the parameters, which I mean:
--channelId Optional: The ID of the channel for the instance that will be installed. This is required for the install command, ignored for other commands if --installPath is specified.
--channelUri Optional: The URI of the channel manifest. This can be used for the install command; it is ignored for other commands.
--installChannelUri Optional: The URI of the channel manifest to use for the installation. The URI specified by --channelUri (which must be specified when --installChannelUri is specified) will be used to detect updates. If updates are not desired, --channelUri must be specified without an argument. This can be used for the install command; it is ignored for other commands.
--installCatalogUri Optional: The URI of the catalog manifest to use for the installation. If specified, the channel manager will attempt to download the catalog manifest from this URI before using the URI in the install channel manifest. This parameter is used to support offline install, where the layout cache will be created with the product catalog already downloaded. This can be used for the install command; it is ignored for other commands.
In the layout directory, there exist files like catalog.json, channelmanifest.json, response.json, vs_installer_version.json among others. Any idea?
You don't need to do anything special for the offline installer to work without any Internet connection. Even though channelURI in <offline folder>\Response.json points online, it will not try to update the component list or something each time, you need to explicitly request an update for that.
The only thing you need to make sure is to select a language that you have downloaded files for on the "Language packs" tab when selecting components -- otherwise, it will try to download the missing bits.
The installer will still try to access the Net
to post telemetry to https://dc.services.visualstudio.com/v2/track and https://vortex.data.microsoft.com/collect/v1 from Microsoft.VisualStudio.Telemetry.Net35.dll that's a part of the installer
to retrieve CRLs for relevant signing certificates from https://crl.microsoft.com
but both are optional, the installer won't fail it it can't do either -- I've just successfully installed stuff with a disabled network connection.
For an offline installer to work you need to 'fake' out the installer to make it think there is an Internet connection, but with bad connectivity.
To do this you need to add various addresses to your local 'hosts' file.
See Visual Studio 2017 Enterprise Offline Installation Fails for a list of names to add to your hosts file.
I tried installing certificates, different layout parameters for cache folder but none of them worked. I solved the issue by running vs_setup.exe in the cache folder, instead of running vs_community.exe with parameters.
We have a customised vCloud environment. We are trying to download the vAPP image as ovf file for migrating it to some other environment. I am following this procedure
Stop the VM.
Click on download button on setting
It asks for download location and type of image (ova/ovf).
It initiates the download.
Now my problem lies on 4th step. When I click download it initiates download and I could see "enabling download" when it happened. After some unknown time(can't predict the time may be 2hr, 3hr 4hr, 1hr) the process gets failed. I have to repeat the process multiple times(at least 3 to 5 times) to start the actual download process where it actually copies the VM image on disk.
I am not able to predict the actual time of VM download and why the process get failed many time before it start the actual export process.
Can someone tell me answers of below mentioned questions
Does vCloud enable download functionality before it allows us to download the VM? If it does how much time it takes for this functionality to enable.
Can we enable this functionality beforehand so that vCloud should just start the VM download process instantly once I shutdown the machine and start the VM export process?
Do you think using CLI tool like ovftool will make the process faster and prevent it from failing so that I will get to know the actual VM download time and we can prepare a plan for migration?
From my limited understanding of working with the API and SDK, I do not think 1 is possible... if it is... it's not straightforward.. at least to me
as for #3 if you are not using the CLI for scripting and automating purposes, yes it would definitely help
The Sitecore documentation provides some pretty clear instructions on how to configure a Sitecore instance as a processing server:
https://doc.sitecore.net/sitecore_experience_platform/xdb_configuration/configure_a_processing_server
However, many of those steps require enabling/disabling of files manually on the installed server. Has anybody seen or built a patch file (similar to SwitchMasterToWeb) that can disable/enable the appropriate functionality as a patch? I would rather not touch the default Sitecore install and instead rely on automated deployment of configuration patches.
I haven't seen this as a patch and not sure if its possible to do this with just one patch (would love to be proved wrong), but for something like this I've used a Powershell script.
I set up Octopus Deploy to run a Powershell script step after deploy to disable files and change settings if patch files can't do the job.
I can highly recommend the Powercore tools for this kind of thing.
https://github.com/adoprog/Sitecore-PowerCore/tree/master/Framework/ConfigUtils
If anybody else winds up looking for this, I've posted some work up on GitHub for patch files for a variety of versions for 8.0:
https://github.com/jst-cyr/Sitecore-Role-Configs
The patches there will do the 'disable/enable/change' for authoring, delivery, or processing. I don't have one for the reporting server.
Sitecore has evaluated POC for same. At this point of time applciable for Sitecore CMS 8.1 rev. 160302 (Update-2). See here-
https://github.com/Sitecore/Sitecore-Configuration-Roles