Bug 1943258
| Summary: | [Assisted-4.7][Staging][Advanced Networking] Cluster install fails while waiting for control plane | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Trey West <trwest> |
| Component: | assisted-installer | Assignee: | vemporop |
| assisted-installer sub component: | assisted-service | QA Contact: | Yuri Obshansky <yobshans> |
| Status: | CLOSED ERRATA | Docs Contact: | |
| Severity: | unspecified | ||
| Priority: | high | CC: | alazar, aos-bugs, itsoiref, yobshans |
| Version: | 4.7 | ||
| Target Milestone: | --- | ||
| Target Release: | 4.8.0 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | AI-Team-Core | ||
| Fixed In Version: | OCP-Metal-v1.0.20.1 | Doc Type: | Bug Fix |
| Doc Text: |
Cause:
In rare cases an arbitrary service may take the .10 IP address in service network, preventing DNS service from starting up properly because it requires this address.
Consequence:
Many subsystems malfunction because DNS service fails to bind to the .10 address in service network. Error message "failed to create service for dns default: failed to create dns service: Service \"dns-default\" is invalid: spec.clusterIPs: Invalid value: []string{\"172.30.0.10\"}: failed to allocated ip:172.30.0.10 with error:provided IP is already allocated" can be seen in OpenShift logs.
Fix:
This issue should be fixed at OpenShift level, but as a workaround Assisted Installer will currently delete and recreate any service that takes the .10 address, in order to free the address up for DNS service.
Result:
The cluster starts up properly, DNS service can take the required IP address even if it was previously taken by another service. Such service will be recreated and will receive a free IP address in service network.
|
Story Points: | --- |
| Clone Of: | Environment: | ||
| Last Closed: | 2021-07-27 22:55:47 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
Trey West
2021-03-25 16:52:17 UTC
@itsoiref Is this similar to some of the the other bugs you handled? this is the issue with dns pod not starting due to already taking ip
from dns operator :
2021-03-25T05:54:54.973779004Z time="2021-03-25T05:54:54Z" level=error msg="failed to reconcile request /default: failed to ensure dns default: failed to create service for dns default: failed to create dns service: Service \"dns-default\" is invalid: spec.clusterIPs: Invalid value: []string{\"172.30.0.10\"}: failed to allocated ip:172.30.0.10 with error:provided IP is already allocated"
*** Bug 1918531 has been marked as a duplicate of this bug. *** @vemporop is there a fixed in version for this? The fix is in the controller, the image quay.io/ocpmetal/assisted-installer-controller:latest already includes it, if you can test with a custom controller image. No release yet. Verified on v1.0.21.1 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 (Moderate: OpenShift Container Platform 4.8.2 bug fix and security update), 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-2021:2438 |