Troubleshooting: Volume attachment fails due to SELinux denials in Fedora downstream distributions

| June 7, 2023

Applicable versions

All Longhorn versions.

Likely environment

Commonly seen in Fedora downstream distributions (e.g. Fedora, RHEL, Rocky, CentOS, etc.) when container-selinux is updated beyond version 2.189.0. This can happen unexpectedly and can catch administrators off guard.

Symptoms

Symptoms are the same as those discussed in a previous KB article that focused on OKD.

All volumes are stuck in an attach/detach loop. dmesg and ausearch on storage nodes reveal SELinux issues:

-> dmesg
...
[Sat Dec 10 18:52:01 2022] audit: type=1400 audit(1670698321.515:7214): avc:  denied  { dac_override } for  pid=231579 comm="iscsiadm" capability=1  scontext=system_u:system_r:iscsid_t:s0 tcontext=system_u:system_r:iscsid_t:s0 tclass=capability permissive=0
[Sat Dec 10 18:52:01 2022] audit: type=1300 audit(1670698321.515:7214): arch=c000003e syscall=83 success=no exit=-13 a0=55b9035185c0 a1=1f8 a2=ffffffffffffff00 a3=0 items=0 ppid=231163 pid=231579 auid=4294967295 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=4294967295 comm="iscsiadm" exe="/usr/sbin/iscsiadm" subj=system_u:system_r:iscsid_t:s0 key=(null)
[Sat Dec 10 18:52:01 2022] audit: type=1327 audit(1670698321.515:7214): proctitle=697363736961646D002D6D00646973636F76657279002D740073656E6474617267657473002D700031302E3133312E312E31363

-> ausearch -m AVC -ts recent
...
----
time->Wed Jun  7 19:33:17 2023
type=PROCTITLE msg=audit(1686166397.967:2849): proctitle=697363736961646D002D6D00646973636F76657279002D740073656E6474617267657473002D700031302E38382E302E39
type=PATH msg=audit(1686166397.967:2849): item=1 name="/var/lib/iscsi/nodes/iqn.2019-10.io.longhorn:vol-name/10.88.0.9,3260,1" nametype=CREATE cap_fp=0 cap_fi=0 cap_fe=0 cap_fver=0 cap_frootid=0
type=PATH msg=audit(1686166397.967:2849): item=0 name="/var/lib/iscsi/nodes/iqn.2019-10.io.longhorn:vol-name/" inode=54532024 dev=fc:01 mode=040600 ouid=0 ogid=0 rdev=00:00 obj=unconfined_u:object_r:iscsi_var_lib_t:s0 nametype=PARENT cap_fp=0 cap_fi=0 cap_fe=0 cap_fver=0 cap_frootid=0
type=CWD msg=audit(1686166397.967:2849): cwd="/"
type=SYSCALL msg=audit(1686166397.967:2849): arch=c000003e syscall=83 success=no exit=-13 a0=562f10bf5540 a1=1f8 a2=fffffffffffffef0 a3=0 items=2 ppid=53345 pid=53577 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 egid=0 sgid=0 fsgid=0 tty=(none) ses=1 comm="iscsiadm" exe="/usr/sbin/iscsiadm" subj=unconfined_u:system_r:iscsid_t:s0 key=(null)
type=AVC msg=audit(1686166397.967:2849): avc:  denied  { dac_override } for  pid=53577 comm="iscsiadm" capability=1  scontext=unconfined_u:system_r:iscsid_t:s0 tcontext=unconfined_u:system_r:iscsid_t:s0 tclass=capability permissive=0

Reason

The root cause is the same as was discussed in a previous KB article that focused on OKD. A permissions issue in open-iscsi causes iscsiadm to create directories under /var/lib/iscsi without the execute bit. When it later needs to access these directories, it cannot do so without the dac_override capability. Updated versions of container-selinux do not grant iscsiadm this capability when run from a container.

This issue was fixed in open-iscsi v2.1.4, but a patch used to build the iscsi-initiator-utils RPM overrode the change in a couple of lines, leaving Fedora downstream distributions affected. Now that many of these distributions are running with an updated container-selinux package, a workaround is required.

Solution

Short term

It is generally not recommended (and difficult) to install an out-of-band version of open-iscsi not provided by your distribution maintainers. Instead, the dac_override permission must be provided to iscsid_t. This can be done by running the following command (or similar) on all nodes before using Longhorn.

echo '(allow iscsid_t self (capability (dac_override)))' > local_longhorn.cil && semodule -vi local_longhorn.cil

While it is not ideal to have to grant the dac_override capability, it is important to recognize that affected versions of iscsi-initiator-utils cannot run without it, even outside of Longhorn. Usually, iscsiadm is executed by an unconfined root user, and uses this capability implicitly.

Multiple Kubernetes distributions already apply this fix by default:

If you are running an affected OS with an up-to-date container-selinux and your Kubernetes distribution doesn’t already apply the fix for you, we provide a DaemonSet that can make the necessary change on all nodes.

git clone https://github.com/longhorn/longhorn.git
cd longhorn
kubectl apply -f deploy/prerequisite/longhorn-iscsi-selinux-workaround.yaml

Long term

A PR was submitted to the Fedora project. If it is accepted, additional capabilities will no longer be required as downstream distributions adopt the changes.

Back to Knowledge Base

Recent articles

Troubleshooting: NoExecute taint prevents workloads from terminating
Troubleshooting: Orphan ISCSI Session Error
Failure to Attach Volumes After Upgrade to Longhorn v1.5.x
Kubernetes resource revision frequency expectations
SELinux and Longhorn
Troubleshooting: RWX Volume Fails to Be Attached Caused by `Protocol not supported`
Troubleshooting: fstrim doesn't work on old kernel
Troubleshooting: Failed RWX mount due to connection timeout
Space consumption guideline
Troubleshooting: Unexpected expansion leads to degradation or attach failure
Troubleshooting: Failure to delete orphaned Pod volume directory
Troubleshooting: Volume attachment fails due to SELinux denials in Fedora downstream distributions
Troubleshooting: Volumes Stuck in Attach/Detach Loop When Using Longhorn on OKD
Troubleshooting: Velero restores Longhorn PersistentVolumeClaim stuck in the Pending state when using the Velero CSI Plugin version before v0.4.0
Analysis: Potential Data/Filesystem Corruption
Instruction: How To Migrate Longhorn Chart Installed In Old Rancher UI To The Chart In New Rancher UI
Troubleshooting: Unable to access an NFS backup target
Troubleshooting: Pod with `volumeMode: Block` is stuck in terminating
Troubleshooting: Instance manager pods are restarted every hour
Troubleshooting: Open-iSCSI on RHEL based systems
Troubleshooting: Upgrading volume engine is stuck in deadlock
Tip: Set Longhorn To Only Use Storage On A Specific Set Of Nodes
Troubleshooting: Some old instance manager pods are still running after upgrade
Troubleshooting: Volume cannot be cleaned up after the node of the workload pod is down and recovered
Troubleshooting: DNS Resolution Failed
Troubleshooting: Generate pprof runtime profiling data
Troubleshooting: Pod stuck in creating state when Longhorn volumes filesystem is corrupted
Troubleshooting: None-standard Kubelet directory
Troubleshooting: Longhorn default settings do not persist
Troubleshooting: Recurring job does not create new jobs after detaching and attaching volume
Troubleshooting: Use Traefik 2.x as ingress controller
Troubleshooting: Create Support Bundle with cURL
Troubleshooting: Longhorn RWX shared mount ownership is shown as nobody in consumer Pod
Troubleshooting: `MountVolume.SetUp failed for volume` due to multipathd on the node
Troubleshooting: Longhorn-UI: Error during WebSocket handshake: Unexpected response code: 200 #2265
Troubleshooting: Longhorn volumes take a long time to finish mounting
Troubleshooting: `volume readonly or I/O error`
Troubleshooting: `volume pvc-xxx not scheduled`

© 2019-2024 Longhorn Authors | Documentation Distributed under CC-BY-4.0


© 2024 The Linux Foundation. All rights reserved. The Linux Foundation has registered trademarks and uses trademarks. For a list of trademarks of The Linux Foundation, please see our Trademark Usage page.