Bug 201671
| Summary: | rpc.mountd won't start when nss_ldap configured with TLS | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | darkness <darkness-keyword-redhat.2bd9c5> |
| Component: | selinux-policy-targeted | Assignee: | Daniel Walsh <dwalsh> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 5 | ||
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | i386 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Current | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2007-03-28 20:05:08 UTC | Type: | --- |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Fixed in selinux-policy-2.3.7-2.fc5 Change to modified Closing bugs |
Description of problem: When nss_ldap is configured for the services database, and nss_ldap is instructed to use SSL/TLS, rpc.mountd fails to start because it cannot access /dev/random and/or /dev/urandom. Version-Release number of selected component (if applicable): selinux-policy-targeted-2.3.3-8.fc5 nfs-utils-1.0.8-2.fc5 How reproducible: Consistently Steps to Reproduce: 1. Configure nss_ldap to use SSL/TLS (I'm using "ssl start_tls" in /etc/ldap.conf). 2. Configure the services database to use LDAP: "services: files ldap" in /etc/nsswitch.conf. 3. Run rpc.mountd. Actual results: rpc.mountd hangs, messages in /var/log/messages: Aug 8 02:30:47 foo mountd[11031]: nss_ldap: reconnecting to LDAP server (sleeping 4 seconds)... SELinux audit messages: audit(1155018584.459:6336): avc: denied { read } for pid=11002 comm="rpc.mountd" name="urandom" dev=tmpfs ino=2730 scontext=user_u:system_r:nfsd_t:s0 tcontext=system_u:object_r:urandom_device_t:s0 tclass=chr_file audit(1155018584.459:6337): avc: denied { read } for pid=11002 comm="rpc.mountd" name="random" dev=tmpfs ino=2723 scontext=user_u:system_r:nfsd_t:s0 tcontext=system_u:object_r:random_device_t:s0 tclass=chr_file audit(1155018647.383:6340): avc: denied { getattr } for pid=11031 comm="rpc.mountd" name="urandom" dev=tmpfs ino=2730 scontext=user_u:system_r:nfsd_t:s0 tcontext=system_u:object_r:urandom_device_t:s0 tclass=chr_file audit(1155018647.383:6341): avc: denied { getattr } for pid=11031 comm="rpc.mountd" name="random" dev=tmpfs ino=2723 scontext=user_u:system_r:nfsd_t:s0 tcontext=system_u:object_r:random_device_t:s0 tclass=chr_file Expected results: rpc.mountd starts immediately. Additional info: The "denied { getattr }" messages only appear after I've made an SELinux module to allow read on random/urandom. Using the following policy module allowed rpc.mountd to start here: module mountd_fix 1.1; require { class chr_file { getattr read }; type nfsd_t; type random_device_t; type urandom_device_t; role system_r; }; allow nfsd_t random_device_t:chr_file { getattr read }; allow nfsd_t urandom_device_t:chr_file { getattr read };