Bug 232652
| Summary: | runcon shouldn't reorder parameters | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Ulrich Drepper <drepper> |
| Component: | coreutils | Assignee: | Ondrej Vasik <ovasik> |
| Status: | CLOSED RAWHIDE | QA Contact: | |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | rawhide | CC: | meyering |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2007-10-29 11:54:20 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: | |||
Hi Uli, Thanks for the report. I've fixed it on the selinux branch, upstream: http://git.sv.gnu.org/gitweb/?p=coreutils.git;a=commitdiff;h=9aee8e78ea Built as coreutils-6.9-9.fc9 , closing RAWHIDE |
Description of problem: The runcon program reorders parameters since it uses the default getopt_long setup. But runcon differs from su here since su does not take a command and its parameters on the command line. Trying to run runcon -t unconfined_t /usr/sbin/prelink --no-update-cache some/binary fails since --no-update-cache is no option for runcon. But runcon shouldn't use it in the first place. Of course one can add -- as a parameter before the program name but this is not how other programs with similar interface work, they don't require this kind of knowledge from the user (strace, sh, ...). Version-Release number of selected component (if applicable): coreutils-6.7-9.fc7 How reproducible: always Steps to Reproduce: 1.echo 'int main(){return 0;}' > u.c 2.gcc -o u u.c 3.runcon -t unconfined_t /usr/sbin/prelink -N ./u Actual results: runcon: invalid option -- N [...] Expected results: Success Additional info: The solution is easy: add + as the first character in the option string of the getopt_long call. I very much doubt any script today depends on this level of option reordering so it should not introduce any regression.