maxwai

Members
  • Posts

    6
  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

maxwai's Achievements

Noob

Noob (1/14)

1

Reputation

  1. Hi, Today I added a Docker Container that is on a custom Container Registry and not on DockerHub. The Registry is a Gitlab Registry hosted by my University. I've added the login credentials via the CLI and it worked without problems. I could add the Docker Container in the GUI and Unraid pulled and started it normaly. I immediately noticed that on the Docker Page that for the container it showed "not available" for the version. So I made an update to the docker image (I'm pushing the updates so I now that there is one available), and tried to force the update via the advance view. But there it just showed my "TOTAL DATA PULLED: 0 B" and restarted the container. Of course the container was not updated. I then tried doing a docker pull in the CLI with the same URL as shown in the GUI and there it worked without a problem. As soon as I did the pull, I went back and retried the forced update, there of course, as excepted, it also pulled 0B and restarted the container with the new image since it now was present locally. So I think the Problem may be that the URL has a port after the domain and so the Unraid updater thinks the port and beyond is already the tag and not part of the URL anymore. If that is really the problem it should be easily fixed by looking for ":" on the far right instead of the first. The Diagnostic file was generated right after trying to update the container in the GUI so maybe there is something there that can help. storinator-diagnostics-20230716-2139.zip
  2. Found that a pre-compiled package is available here: https://slackonly.com/pub/packages/15.0-x86_64/development/git-lfs/ I have tested that the package works.
  3. I also opened up a issue on the GitHub Repo but I think it may be favorable to document how a package can be added to the plugin. With this, the community could do the heavy lifting of finding and maybe even compiling the packages and adding packages would go faster. Of course this introduces the problem of verifying that a package wasn't modified from the source. For that I propose that for packages that were already found compiled, the link to the checksum could be given (so not the checksum directly but the link of where to find the checksum in combination with the package itself) to verify the package. For packages that need to be compiled manually because they are not available in the public package lists, I propose, instead of adding the compiled package directly, which can't be easily verified, to open a Issue with the necessary commands to compile it. So that would be the commands to download the source, extract it, build and compile the package, and package it again for the plugin. Link to the issue
  4. Closed since already reported here:
  5. So following up with this Report that is falsely marked as solved: clearing cookies does help but only temporarily. I did not yet find out what exactly is the condition but think I did find out which cookies are the reason for it (so deleting only these cookies already helps for a moment): rxd-init and txd-init Here is the error message when vnc doesn't work: May 21 20:15:15 Storinator nginx: 2022/05/21 20:15:15 [error] 29980#29980: *351315 recv() failed (104: Connection reset by peer) while reading upstream, client: 192.168.44.43, server: , request: "GET //wsproxy/5700/ HTTP/1.1", upstream: "http://127.0.0.1:5700/", host: "192.168.44.12"