Bug 1625457
| Summary: | Pull of image during docker build failed with "Cannot overwrite digest" for the same digest | |||
|---|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Clayton Coleman <ccoleman> | |
| Component: | Containers | Assignee: | Miloslav Trmač <mitr> | |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | weiwei jiang <wjiang> | |
| Severity: | medium | Docs Contact: | ||
| Priority: | unspecified | |||
| Version: | 3.11.0 | CC: | amurdaca, aos-bugs, ccoleman, jokerman, mitr, mmccomas, mpatel, wking, xtian | |
| Target Milestone: | --- | |||
| Target Release: | 3.11.0 | |||
| Hardware: | Unspecified | |||
| OS: | Unspecified | |||
| Whiteboard: | ||||
| Fixed In Version: | Doc Type: | Bug Fix | ||
| Doc Text: |
Cause: The image pull implementation released a lock between checking whether an image with a specific digest has already been pulled and actually pulling it and recording the results.
Consequence: Concurrent requests to pull the same image could result, depending on the specific timing, in some of them failing with an “error pulling image …: Cannot overwrite digest …” error message.
Fix: Repeated attempts to record the same remote digest for the same locally pulled image are no longer treated as errors.
Result: Concurrent requests to pull the same image no longer fail with the ”Cannot overwrite digest” error.
|
Story Points: | --- | |
| Clone Of: | ||||
| : | 1628262 (view as bug list) | Environment: | ||
| Last Closed: | 2018-12-21 15:23:35 UTC | Type: | Bug | |
| Regression: | --- | Mount Type: | --- | |
| Documentation: | --- | CRM: | ||
| Verified Versions: | Category: | --- | ||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | ||
| Cloudforms Team: | --- | Target Upstream Version: | ||
| Embargoed: | ||||
|
Description
Clayton Coleman
2018-09-05 02:32:52 UTC
That error would have to be coming from the docker daemon (we're just pulling the image via the node's docker daemon), so starting w/ the containers team to tell us how this can happen. It is either a docker bug, or it's an issue w/ the content the internal registry is serving. Also what is the build that's happening here? Is this a docker strategy build where the FROM is docker-registry.default.svc:5000/ci-op-rn4204r8/pipeline@sha256:7ecadbdff313447670cf4b2ff21e827cf87d0f2df0d76f8273756f20dc9764a6 ? if not, what is the flow that is leading to the attempt to pull that image? (how is the image reference by the buildconfig?) https://bugzilla.redhat.com/show_bug.cgi?id=1628262 has this (or at least something that looks very similar) as fixed in docker-1.13.1-75.git8633870.el7_5. Maybe we need a version bump? This seems to be the upstream patch: https://github.com/moby/moby/pull/37781 It is merged here - https://github.com/projectatomic/docker/pull/321 and should be in docker -75 build. Closing bugs that were verified and targeted for GA but for some reason were not picked up by errata. This bug fix should be present in current 3.11 release content. The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days |