A recent build has failed with the following error.
[interface] Step 1/24 : FROM balenalib/raspberrypi3-node:12-stretch-build as build
[interface] Cannot overwrite digest sha256:39e1529973265976a80a22cc17650daefc2bd0a24cb12fd2a370fc17747190a1
Between this build and the previous we had not made any updates to the interface service.
I am attempting to build the same commit again at the moment, it has started the build from the beginning, therefore, I assume that this issue will be resolved.
I am not looking for help with the above but thought it would be useful to share.
The following build failed with a “no such image” error (although the error message is slightly more verbose than when I have seen this error before).
Here is a section from the build output, the controller service has not been changed in this push so it should be using a cache for all layers of the dockerfile.
[controller] Step 18/22 : WORKDIR /app
[controller] Using cache
[controller] ---> 26dd7c91f442
[controller] Step 19/22 : COPY --from=build /app .
[Info] Still Working...
[controller] failed to get destination image "sha256:26dd7c91f4427270d3aae1c9cf428836297156758e5d5218032bb9f3ff670de3": No such image: sha256:26dd7c91f4427270d3aae1c9cf428836297156758e5d5218032bb9f3ff670de3
I am running another build and I will share the results from my next build when it finishes.
There are 5 services which create images between 0.2 and 1.2 GB. I have not cancelled the build and will try starting it again. Mostly they take 10 - 20 minutes to build but we have not been able to get any of our code onto devices today.
We have spent some time today moving our CI to do local building, however facing some issues with emulated builds where the context and dockerfile for a service are in different folders.
Hi. We’re looking into the builder errors. Are the logs you posted everything or is there something after the Cannot overwrite digest.... and failed to get destination image....?
Fairly certain we’ve had it in the past, but not to a noticeable degree. Seems to be having a field day today, though, which is rather appreciably slowing my progress.
Thanks, sorry for the delay, we are still looking into this, there have been some issues with the builders lately and we are trying to understand if this error you reported is related to those. Will update you as soon as we have a definite answer on this.