Bug 1451769
Summary: | Image and Container GC failing at set thresholds | ||
---|---|---|---|
Product: | OpenShift Container Platform | Reporter: | Matthew Robson <mrobson> |
Component: | Containers | Assignee: | Jhon Honce <jhonce> |
Status: | CLOSED ERRATA | QA Contact: | DeShuai Ma <dma> |
Severity: | low | Docs Contact: | |
Priority: | high | ||
Version: | 3.3.1 | CC: | aos-bugs, decarr, dwalsh, erjones, imcleod, jhonce, jokerman, mmccomas, mrobson, sjenning, vgoyal |
Target Milestone: | --- | Flags: | jhonce:
needinfo-
|
Target Release: | 3.7.0 | ||
Hardware: | All | ||
OS: | All | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Enhancement | |
Doc Text: |
Feature: POOL_META_SIZE configuration item
Reason: thin pool metadata size was set to .1% of free space of volume group.
Result: POOL_META_SIZE allows the operator to customize the size of thin pool metadata volume size to meet their workload.
|
Story Points: | --- |
Clone Of: | Environment: | ||
Last Closed: | 2017-11-28 21:55:46 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
Matthew Robson
2017-05-17 13:28:11 UTC
Don't know what does "context deadline exceeded" mean? Is there some timeout in openshift/kube. If yes, how do we know time is sufficient to finish operation. Can you run "docker info" on the node. How do we know docker is not responding? I think openshift/kube needs to fill in more details here (especially with docker logs) showing that docker is dead, before it comes to platform team. May be docker is not running at all at the node. I mean without having more details about state of node, don't assume this is a docker issue. If the customer has a node in this condition do this: $ curl -H "Authorization: Bearer $(oc whoami -t)" --insecure https://localhost:10250/metrics | grep -e docker Should give us some indication as to docker's responsiveness as seen from the openshift node process. Going by the docker logs, docker is refusing creation of new containers/images because it thinks metadata is almost full. So you need to either grow metadata lv or remove some images/containers and that should reduce usage of metadata volume. time="2017-08-14T17:28:55.161557210-07:00" level=error msg="Handler for POST /containers/create returned error: devmapper: Thin Pool has 7863 free metadata blocks which is less than minimum required 7884 free metadata blocks. Create more free metadata space in thin pool or use dm.min_free_space option to change behavior" Verify on ocp3.7, when imagegc trigger we can seed th imageGCManager trying to cleanup the unused image. # openshift version openshift v3.7.0-0.146.0 kubernetes v1.7.6+a08f5eeb62 etcd 3.2.1 //logs like: Oct 10 04:36:30 ip-172-18-4-102 atomic-openshift-node: I1010 04:36:30.520176 32692 image_gc_manager.go:270] [imageGCManager]: Disk usage on "rhel-docker--pool" () is at 28% which is over the high threshold (10%). Trying to free 2904974950 bytes Oct 10 04:41:30 ip-172-18-4-102 atomic-openshift-node: I1010 04:41:30.710926 32692 image_gc_manager.go:335] [imageGCManager]: Removing image "sha256:220072a1dafa8eb7f5739606393285301888277b21fe65f14614ad86377e6332" to free 208663627 bytes Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2017:3188 |