Bug 1085081
| Summary: | Docker sshd closing connection RHEL 6.5 | ||
|---|---|---|---|
| Product: | [Fedora] Fedora EPEL | Reporter: | aschulz <arnoschulz> |
| Component: | docker-io | Assignee: | Lokesh Mandvekar <lsm5> |
| Status: | CLOSED NEXTRELEASE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | el6 | CC: | acathrow, admiller, arnoschulz, dwalsh, golang-updates, jkeck, lsm5, mattdm, mgoldman, skottler, vbatts |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2014-06-03 20:31:34 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
aschulz
2014-04-07 18:42:59 UTC
Running sshd -dddd I've got debug1: SELinux support enabled debug3: ssh_selinux_setup_exec_context: setting execution context ssh_selinux_getctxbyname: Failed to get default SELinux security context for root ssh_selinux_setup_exec_context: security_getenforce() failed debug1: do_cleanup which pointed me to https://groups.google.com/forum/#!msg/docker-user/7EyZthXHcww/B3YAV0XsxNAJ Disabling SELinux worked, setting it to permissive didn't. There's a pull requested that pull request regarding SELinux support in docker that was merged 12 days ago, perhaps will be included in the next rpm release https://github.com/dotcloud/docker/pull/4211. I've checked and SELinux pull was included in Docker 0.10.0 and I got the RPM (ftp://195.220.108.108/linux/epel/testing/6/x86_64/docker-io-0.10.0-2.el6.x86_64.rpm) for RHEL installed but I still get the same error as Guy debug1: SELinux support enabled debug3: ssh_selinux_setup_pty: setting TTY context on /dev/pts/0 ssh_selinux_getctxbyname: ssh_selinux_getctxbyname: security_getenforce() failed debug1: do_cleanup debug1: PAM: cleanup debug1: PAM: closing session Similar by disabling SELinux everything works like a charm, but setting it to either Enforcing or Permissive and it fails. No logs on either the host or container under /var/log/messages for SELinux errors (In reply to aschulz from comment #3) > Similar by disabling SELinux everything works like a charm, but setting it > to either Enforcing or Permissive and it fails. Copying dwalsh. We've updated the images (yum update) and now SSH works with SELinux enabled (tested with both Ubuntu and Fedora containers) tested on RHEL 6.5 host. No idea as to what caused the initial issue. One final note, The issue seems to be still present for Ubuntu 12.04 containers but 12.10, 13.04, 13.10 and 14.04 containers work without a hitch. http://people.redhat.com/dwalsh/SELinux/RHEL6/ contains an update libselinux which should be added to rhel6.5 image, which should allow stuff like useradd/groupadd to work. When we ship an image it will have the fixed libselinux. Until then use the library I provided. |