Bug 2483872

Summary: cri-o failed to start after upgrade to fedora coreos 44.20260510.3.1 (stable) due to selinux issue
Product: [Fedora] Fedora Reporter: Junjie Yuan <yuan>
Component: cri-o1.35Assignee: Brad Smith <bradley.g.smith>
Status: CLOSED DUPLICATE QA Contact:
Severity: high Docs Contact:
Priority: unspecified    
Version: 44CC: bradley.g.smith, go-sig
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: ---
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2026-06-02 14:06:32 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 Junjie Yuan 2026-06-02 03:01:50 UTC
Description of problem:
cri-o failed to start due to selinux issue, upgrade coreos to newer version (testing 44.20260523.2.1) can't resolve it.

Version-Release number of selected component (if applicable):
fedora coreos 44.20260510.3.1
cri-o1.35-1.35.3-1.fc44.x86_64
container-selinux-2.247.0-1.fc44.noarch

How reproducible:

Steps to Reproduce:
1. systemctl enable --now crio
2. upgrade fedora core from previous version
3. crio failed to start

Actual results:
crio failed to start due to selinux issue

Expected results:
crio running


Additional info:

selinux:

time->Tue Jun  2 02:58:34 2026
type=AVC msg=audit(1780369114.463:4115): avc:  denied  { execmem } for  pid=33265 comm="crio" scontext=system_u:system_r:kernel_t:s0 tcontext=system_u:system_r:kernel_t:s0 tclass=process permissive=0
----
time->Tue Jun  2 02:58:44 2026
type=AVC msg=audit(1780369124.712:4121): avc:  denied  { execmem } for  pid=33305 comm="crio" scontext=system_u:system_r:kernel_t:s0 tcontext=system_u:system_r:kernel_t:s0 tclass=process permissive=0
----
time->Tue Jun  2 02:58:54 2026
type=AVC msg=audit(1780369134.962:4127): avc:  denied  { execmem } for  pid=33354 comm="crio" scontext=system_u:system_r:kernel_t:s0 tcontext=system_u:system_r:kernel_t:s0 tclass=process permissive=0
----
time->Tue Jun  2 02:59:05 2026
type=AVC msg=audit(1780369145.212:4131): avc:  denied  { execmem } for  pid=33392 comm="crio" scontext=system_u:system_r:kernel_t:s0 tcontext=system_u:system_r:kernel_t:s0 tclass=process permissive=0

crio log:

Jun 02 02:57:02 k8s-gpu-worker-001 systemd[1]: crio.service: Scheduled restart job, restart counter is at 695.
Jun 02 02:57:02 k8s-gpu-worker-001 systemd[1]: Starting crio.service - Container Runtime Interface for OCI (CRI-O)...
Jun 02 02:57:02 k8s-gpu-worker-001 (crio)[32879]: crio.service: Referenced but unset environment variable evaluates to an empty string: CRIO_CONFIG_OPTIONS, CRIO_METRICS_OPTIONS, CRIO_NETWORK_OPTIONS, CRIO_RUNTIME_OPTIONS, CRIO_STORAGE_OPTIONS
Jun 02 02:57:02 k8s-gpu-worker-001 crio[32879]: /usr/bin/crio: error while loading shared libraries: cannot make segment writable for relocation: Permission denied
Jun 02 02:57:02 k8s-gpu-worker-001 systemd[1]: crio.service: Main process exited, code=exited, status=127/n/a
Jun 02 02:57:02 k8s-gpu-worker-001 systemd[1]: crio.service: Failed with result 'exit-code'.
Jun 02 02:57:02 k8s-gpu-worker-001 systemd[1]: Failed to start crio.service - Container Runtime Interface for OCI (CRI-O).

Comment 1 Junjie Yuan 2026-06-02 03:05:42 UTC
Even after restoring the SELinux context of the binary, it still fails to start.

Running `setenforce 0` can temporarily work around the issue.

crio selinux (default, no changed):

core@k8s-gpu-worker-001:~$ ls -lZ /usr/bin/crio
-rwxr-xr-x. 1 root root system_u:object_r:container_runtime_exec_t:s0 75423640 Jan  1  1970 /usr/bin/crio

Comment 2 Brad Smith 2026-06-02 14:06:32 UTC
Thank you for the report. Please see https://bugzilla.redhat.com/show_bug.cgi?id=2477939 for more information. There is also information at https://github.com/coreos/fedora-coreos-tracker/issues/2148.

*** This bug has been marked as a duplicate of bug 2477939 ***