Posts made by rhessinger

Hi @22marat22_9029,

So turns out that we had a compatible library from our Conda implementation that I was able to use for zstd compression with Debian. I have created a ticket, PG-2242, to track the fix and it will be released in the next release of ProGet, 2022.14.

Thanks,
Rich

Hi @22marat22_9029,

Currently, we do not support the compression format of zstd (tar.zst) for the control or the data file. We only support .tar, .tar.gz, and .tar.xz fir the data and control files with the addition tar.bz2 and tar.lzma for the data file . This is currently a limitation of the third-party package we use to read tar files. Hopefully, this is something we can add support for in the future.

Thanks,
Rich

Hi @krampusdecorous_7076,

We have created a ticket, PG-2233, to fix this issue in ProGet. We are not exactly sure what is causing the issue as of yet, but we are able to recreate the issue and working on a fix. We are currently targeting Nov 18th for a release date on the fix.

Thanks,
Rich

Hi @jameswiseman76_4902,

I think you are looking for the npm search API. You would make a call to your npm feed using http://{proget-server}/npm/{feedName}/-/v1/search?text={Package}. It also looks like http://{proget-server}/npm/{feedName}/-/all will show you all local packages (it does not include remote packages from connectors) stored in ProGet. Please also note that the search API returns results in a paged fashion, you will need to use the from query string parameter to offset the results to get the following pages.

I hope this helps! Please let us know if you have any other questions.

Thanks,
Rich

Hi @andrewwickham_1641,

Thanks for giving us an update and letting us know that fixed your issue.

Thanks,
Rich

Hi @torgabor_4445 and @sdohle_3924,

I dug deeper into GitLab connectors and it looks like there are two ProGet issues here and a configuration issue.

1. PG-2210: GitLab does not return the "time" property when it returns package metadata. This causes ProGet to fail to return any packages.
2. PG-2211: There is a display issue when entering only a password with bearer authentication. The UI will show anonymous, even though it will use the token.
3. Configuration Issue: here is what you will need to configure a GitLab connector.
   • Connector URL: https://gitlab.com/api/v4/projects/12345/packages/npm/
   • User Name: leave blank
   • Password: Generate a personal access token with API access and set that value there.
   • Authentication: Bearer
   • Advanced Options: Check exact package match only. GitLab does not support partial searching and/or listing

I have added fixes for PG-2210 and PG-2211 and they will release tomorrow in ProGet 2022.9. Once you upgrade to 2022.9 and make the configuration updates, you should be able to pull GitLab packages via a connector.

Thanks,
Rich

Hi @torgabor_4445,

Can you please try something for me?

• Leave the username blank
• Put your token in for the password
• Use Bearer authentication
• Change your URL to https://gitlab.com/api/v4/projects/12345/packages/npm/:_authToken

If that does not fix it, I have another idea, but I will need to check a few more things offline.

Thanks,
Rich

Hi @marc-ledent_9164,

What version of BuildMaster do you have installed? Also what extensions are you trying to update that are currently giving you that error?

Thanks,
Rich

Hi @tkolb_7784,

If you are hosting on a windows machine, the easiest solution right now is to migrate your server to use IIS and then add an SSL binding to your site. If you do not want to purchase a new certificate and the self-signed certificate is too much work, you can use Let's Encrypt and configure it via winacme.

If you do not want to use IIS, then you will need to use a reverse proxy to handle SSL connections. Any reverse proxy can be used and a pretty simple one to configure is stunnel. Most reverse proxies can also be used with Let's Encrypt.

If you are hosting via Linux (Docker), then you will need to use a reverse proxy to handle SSL connections. We have a documentation page for different Linux-based reverse proxies including an example for setting up NGINX. These reverse proxies also support Let's Encrypt also.

Please let me know if you have any questions.

Thanks,
Rich

Hello,

This is related to a known issue that's been addressed in ProGet 6.0.19 and ProGet 2022.5. So, your best bet is to upgrade and the issue will become resolved :)

This is related to a few packages that have exceeded 2.2 billion downloads:

• Microsoft.NETCore.Platforms
• Microsoft.Extensions.Primitives
• Microsoft.Extensions.DependencyInjection.Abstractions

If the upgrade is impossible/difficult immediately, you can disable the connector as a workaround. Alternatively, you could block those packages with a connector filter and then upload them to your feed so that the counts won't come through the connector.

Thanks,
Rich

Hi @mike_2282,

Thank you for the feedback. I have made some changes to the documentation to address most of the issues you have brought up. Thanks again for the feedback!

Thanks,
Rich

Hi @mike_2282,

Thanks for alerting us to this. I'll get that added to our documentation. Please let us know the other issues you find and hopefully we can get those addressed in our documentation also.

Thanks,
Rich

Hi @v-makkenze_6348,

Can you please try to install the .NET 6.0 Web Hosting Bundle and see if that resolves your issue? Inedo Hub typically handles installing this, but I'm guessing that it did not detect your site in IIS.

Thanks,
Rich

Hi @pariv_0352,

I was able to fix the issue, PG-2160, and it will be released next week on ProGet 2022.2. If you would like to use the fix earlier, you can install a pre-release version of PrGet by installing ProGet 2022.0.2-CI.7 or higher. We have a walk-through on how to install prerelease products in our documentation.

Please let me know if you have any questions.

Thanks,
Rich

Hi @pariv_0352,

Thanks for all the information. I have been able to recreate the issue and I'm currently looking into it. I'll have an update for you soon.

Thanks,
Rich

Hi @gurdip-sira_1271,

I did some more research on this and I think I have been able to reproduce this issue. I have created a ticket, OT-477, to track the fix for this issue. I expect to have this fixed within the next couple of versions of Otter (2022.5 or 2022.6). It is possible that this is an issue with the git extension directly, but I will update you when I have a release date for the fix.

Thanks,
Rich

Hi @gurdip-sira_1271,

Would you be able to share a screen shot of your Git repository so we can see how the files and folder structure? You can send them to support@inedo.com if you do not feel comfortable sharing on here. If you do email it to us, please include [QA-882] in the subject.

Thanks,
Rich

Hi @gurdip-sira_1271,

There are a couple of things that can cause these to go out of sync. If you navigate to Administration -> Raft Repositorites and then click browse to the right of your Git raft. Do you see the scripts showing up in there? Also, which Git provider (GitHub, GitLab, etc...) are you using?

Thanks,
Rich

Hi @aries66_2180,

I just set up a clean ProGet 6.0.16 docker instance and tested this and it seems to be working for me. Can you send me an example image you are trying to push that fails? Also, would you be able to use a tool like WireShark or Fiddler to record your requests when trying to push an image and email it over to us at support@inedo.com with a subject of [QA-878] Docker push error?

Thanks,
Rich

Hi @aries66_2180,

If this happens on a clean Docker feed that is not set up the use common blob storage, then this is likely an issue with the image itself. Do all images have this issue or just that specific image? Are you using the Docker CLI to create this image or a different third-party tool?

Thanks,
Rich

Hi @aries66_2180,

We typically have seen this when there is a corrupted or orphaned blob in the database. Could you create a new temporary Docker feed, disable common blob storage, and attempt to push the image to that feed?

Thanks,
Rich

Hi @aries66_2180,

Can you please try running the "DockerGarbageCollection" scheduled task. This will do the same thing as the feed cleanup task, but for the common blob storage blobs and then attempt the push again?

Thanks,
Rich

Hi @aries66_2180,

Can you please check if you have common blob storage enabled for your Docker feed? You can do this by navigating to your Docker Registry and click the "Manage Feed" button in the upper right and then select the "Storage & Retention" tab. On there, you should see a block indicating if it is enabled or not.

Thanks,
Rich

Hi @chris-f_7319,

I'm sorry about that, I didn't realize the version of ProGet you are running was based on mono (we switched to .net 5 later in the lifecycle of ProGet 5.3). The command you will want to use is:

 sudo docker exec -it proget mono /usr/local/proget/service/ProGet.Service.exe resetadminpassword

If mono doesn't work, try using /usr/bin/mono instead of mono.

Thanks,
Rich

Hi @phillip-t_2200,

Azure SQL Managed Instances should work fine. We don't have any direct test cases for this, but based on our SQL implementation, that should work just fine.

Thanks,
Rich

Hi @chris-f_7319,

The command that you show is different than the command I sent. Can you confirm you tried:

docker exec -it proget exec /usr/local/proget/service/ProGet.Service resetadminpassword

It has a slightly different syntax than you provided. Also, what Docker host are you running?

Thanks,
Rich

Hi @chris-f_7319,

I think Dan was looking at the upcoming ProGet v2022 file structure. Can you please try:

docker exec -it proget exec /usr/local/proget/service/ProGet.Service resetadminpassword

Thanks,
Rich

Hi @gurdip-sira_1271,

Yes, this will be released with Otter later today.

Thanks,
Rich

Hi @marc-ledent_9164,

You will want to specify the "Domain to Search" as gcloud.dom,LDAPuser. For the secure credential, you will want to use just a username and password, unless the user logs in with a different suffix other than @gcloud.dom.

I think the issue is with the binddn. BuildMaster will connect to LADP/AD using the root OU. If you require a CN and OU to be specified, that will not work out of the box. Are those needed to connect to your domain controller?

Thanks,
Rich

Hi @marc-ledent_9164,

To specify a username/password to use to communicate with you domain, need to:

1. Add a Username/Password secure credential in the Administration -> Secure Credentials page and make note of the name you used (ex: ADDomainCreds)
2. Navigate to Administration -> Change User Directory (LDAP)
   1. Click "Advanced"
3. Find your Active Directory provider and click the name (most commonly named Active Directory (LDAP))
4. Change the "Search Mode" to Specific List
5. In the "Domains to search" enter your domain name and secure credentials name in the format of <DOMAIN_SUFFIX>,<CREDENTIAL_NAME> (ex: kramerica.local,ADDomainCreds)
6. In the "Domain Controller Host", enter the host for your domain controller. Normally the suffix will work (ex: kramerica.local), but if not, enter the IP address of your domain controller.
7. Click "Save"
8. Restart your container(s)

Please let me know if that works for you.

Thanks,
Rich

Hi @gurdip-sira_1271,

Thanks for sending over the information. I have identified the issue, OT-472, and we plan to release a fix in Otter 2022.3 that is due out next Friday. I'll let you know if anything changes.

Thanks,
Rich

Hi @gurdip-sira_1271,

Can you share your OtterScript for that operation? Or does this happen when you try to add that OtterScript script as an operation?

Thanks,
Rich

Hi @jramon-ortega_2700,

We released a new version of our scripting extension, Scripting 2.0.1. You should be able to update the extension or upgrade Otter to 2022.02 and this should fix the issue for you.

Thanks,
Rich

Hi @gurdip-sira_1271,

If you navigate to Administration -> Raft Repositories, then click "browse" to the right of your Git repository, does your raft load or do you get an error?

Also a couple of notes:

• You should not enter a user name and password if you are using a Secure Credential. These may be conflicting.
• Git rafts use the generic git extension, so you will need to use a Username & Password credential instead of a GitHub Credential
  • Just enter your PAT as the password
• You may need to recreate your raft repository because clearing the password field normally just marks it as unchanged

Please give these a try (including browsing in your git raft) and see if that works for you.

Thanks,
Rich

Hi @kichikawa_2913,

Can you show me what your extensions page looks like?

Also, when you start your container, if you watch the output, do you see any errors when loading the extensions?

Thanks,
Rich

Hi @fabian_7019,

Thanks for bringing this to our attention. I was able to recreate the error and we will get this fixed in the next version of Otter, 3.0.25, which is due out next Friday. This looks to only be an issue when there is only one raft repository setup in Otter. As a workaround, if you create a second raft repository under Administration -> Raft Repositories, you will then be able to create new folders in the default raft.

The fix is being tracked in ticket OT-461.

Thanks,
Rich

@gurdip-sira_1271 said in Help with Git raft in Otter:

Could I not just do a git commit to the repo and then use the scripts in Otter?

Hi @gurdip-sira_1271,

Yes, you can commit your scripts directly to Git by adding them to the Scripts folder. You can also modify your scripts directly in your Git repository as well or you can use the editor directly in Otter. We also just added a new text editor based on Monaco (the same editor as VS Code) and a new visual editor for OtterScript available as a preview feature in the latest version of Otter (3.0.24).

We don't have direct end-to-end steps on adding scripts directly in Git because that has not been the most typical way users have used this feature. Typically, a user adds the script via Otter and then will edit them in git.

The other way git rafts are used is with git branches. A raft is created for each git branch where the editing and testing of scripts are done in one branch and the production scripts are stored in another branch. This can become tricky though when calling scripts for different rafts in OtterScripts.

Hope this helps!

Thanks,
Rich

Hi @gurdip-sira_1271,

I think the safer option would be to upload your scripts on the "Scripts" page in Otter. If you navigate to the "Scripts" page, click "Add Script", and then select "Upload Scripts & Assets". You can then select your Git raft and then bulk upload all your script files. This way, they all get put in the correct folder automatically.

Thanks,
Rich

Just as a follow-up on the solution. This error came from a Git Repository Monitor. The Docker Image for Build Master does not include git installed out of the box. This will require you to either install git on the running container or use a BuildMaster Agent/SSH server to be added and to run this on there. The other issue was that the repository monitor was using a secure resource from a specific application, so the repository monitor needs to specify the application it uses as well.

Hi @luke_1024,

It looks like the error happens when trying to decompress your control file in the deb package. Something that cargo-deb is doing compresses and attaches that differently than dpkg-deb does. I took a quick look through cargo-deb's docs, but couldn't find anything to specify the compression. Something to try would be using the --fast flag when running cargo deb .

Please let me know if --fast fixes it. If not, I will set aside some time next week to debug through these packages more.

Thanks,
Rich

Hi @marc-ledent_9164,

Thanks for confirming that for me. How often do you see this issue? Is it always for the same application or does it happen on any application?

Thanks,
Rich

Hi @marc-ledent_9164,

Do you have anything that could be blocking connections to my.inedo.com? I'm curious if there is something preventing the connection that would cause that to hang. Also, if you open up your browser's dev tools, do you see any errors show in the console?

Thanks,
Rich

Hi @marc-ledent_9164,

Can you please try to restart your container and then attempt to activate again? Sometimes I have seen that the key is cached and it does not clear correctly, so a simple restart fixes it.

Thanks,
Rich

Hi @marc-ledent_9164,

Can you please tell me what version of BuildMaster you are running?

Thanks,
Rich

Hi @luke_1024,

Could you also try using gz compression instead of xz compression on your control and data archive? I don't know cargo-deb, but what I see in their docs, it looks like you need to do this:

deb_contents.push(compress::gz(&control_archive, &control_base_path)?);
deb_contents.push(compress::gz(&data_archive, &data_base_path)?);

Thanks,
Rich

Hi @luke_1024,

Would you be able to share the .deb file created by cargo-deb and the one created by dpkg-deb -b? I think I see where the issue is, but that would allow me to confirm the issue. You can share these via email by sending it to support@inedo.com with a subject of [QA-816] Package File Examples.

Thanks,
Rich

Hi @luke_1024,

I just wanted to let you know that we are currently researching your issue and we should have more information for you tomorrow.

Thanks,
Rich

Hello @nuno-ildefonso_8876,

Can you please tell me what version of BuildMaster you are using and what version of the GitLab extension you have installed?

Thanks,
Rich

Hi @pariv_0352,

The "FeedCleanup" job will do two main things; clearing the Multipart Upload Temp files and running retention rules. For Docker feeds, it will also run the deletion of blobs, but only for Feed specific blobs. If you are using common blob storage (which is enabled by default in feeds created in ProGet 5.3+), then that is where the "DockerGarbageCollection" job comes into play. That one handles cleaning up/deleting the blobs that are shared across multiple feeds.

Hope this helps!

Thanks,
Rich

Hi @testing-avram1_6229,

Can you please tell me what version of BuildMaster you are using and what version of the BuildMaster agent you have installed? Also, if you change the BuildMaster agent to a Delayed Start instead of an Automatic start on your remote server, do you see the same issue?

Thanks,
Rich