Can not fetch update image on deployent


I recently install mender 3.4 with Docker compose.
The server was working, I have already launched deployments on devices.
We received a new device from a supplier, I configured the Mender client and tried to update it.
The deployment failed with the following logs:

2023-11-27 09:30:08 +0000 UTC info: Running Mender client version: 2.6.0-dirty
2023-11-27 09:30:08 +0000 UTC error: Can not fetch update image: Get "https:///mender-artifact-storage/62ff65b5-6cb8-49b4-9fee-c46ec57cc4dc?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=mender-deployments%2F20231127%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20231127T093008Z&X-Amz-Expires=86400&X-Amz-SignedHeaders=host&response-content-type=application%2Fvnd.mender-artifact&X-Amz-Signature=d5f7eeed59f27dfc06052a697afde00e0d59e2d0487f94e6e579489242c4cdb7": http: no Host in request URL
2023-11-27 09:30:08 +0000 UTC error: Update fetch failed: update fetch request failed: Get "https:///mender-artifact-storage/62ff65b5-6cb8-49b4-9fee-c46ec57cc4dc?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=mender-deployments%2F20231127%2Fus-east-1%2Fs3%2Faws4_request&X-Amz-Date=20231127T093008Z&X-Amz-Expires=86400&X-Amz-SignedHeaders=host&response-content-type=application%2Fvnd.mender-artifact&X-Amz-Signature=d5f7eeed59f27dfc06052a697afde00e0d59e2d0487f94e6e579489242c4cdb7": http: no Host in request URL
2023-11-27 09:30:08 +0000 UTC info: State transition: update-fetch [Download_Enter] -> fetch-install-retry-wait [Download_Enter]
2023-11-27 09:31:08 +0000 UTC info: State transition: fetch-install-retry-wait [Download_Enter] -> update-fetch [Download_Enter]

The supplier manages to update remotely without encountering this problem.

I don’t understand where this problem comes from.
Did anyone have an idea?

Thank you in advance for your answers

@Trankil nothing that immediately strikes me. Client version 2.6…0-dirty sounds a bit suspicious, it indicates an old and possibly patched up thing. Does this also happen on the current maintained version?

Plus, does manually pulling the artifact work?

@robgio maybe can share some thoughts too.


@Trankil could it be a network problem in the new device? http: no Host in request URL

Can you try to run some troubleshooting command from the device? E.g.:

dig mender-artifact-storage
ping mender-artifact-storage
curl https:///mender-artifact-storage/62ff65b5-6cb8-49b4-9fee-c46ec57cc4dc?X-Amz-Algorithm=.<redacted>

The supplier provided us with the equipment with this version, we are not supposed to modify it…

What do you mean by manually pulling the artifact ?

dig: command not found
curl: (6) Could not resolve host: mender-artifact-storage
ping: mender-artifact-storage: Name or service not known

mender-artifact-storage is not a local address for docker compose?

@Trankil see the post by @robgio, the curl invocation does manually download (slang: “pull”) the artifact from the server.

Please understand that especially concerning outdated packages and self-hosted instances of the open source backend our support options are very limited.

Concerning your answer to @robgio: if you cannot ping a name, then name resolution does not work. Hence, no tool will be able to download anything from it. So something with the network configuration is wrong in this case.


I don’t understand what the domain name “mender-artifact-storage” means because it is not that of my server.
What does this domain name correspond to?
Where can I modify it?

Look at /etc/mender.conf, as documented here. This is probably set by your supplier. I strongly suggest you contact them to get it sorted out.