Lukas Vrabec wrote:
On 05/16/2016 03:39 PM, m.roth@5-cent.us wrote:
We're working on several new CentOS 7 systems, moving users from
CentOS 6. Now, the users have had some *sigh* custom stuff, like their own version of Perl (please do *not* ask, and I would *love* to get them off it, but....)
Anyway, in the directory it's in, I did a semanage fcontext -e /usr/bin, and now I'm seeing errors in the log of selinux complaining it can't find the rpm (because there's not one for this).
What's the correct way to deal with this - different labelling, a local policy, or ?
*Speaking* of could not find rpm... we're required to use, for some websites, CA's SiteMinder. Now, I could go off on a rant about how CA, that's the megacompany, Computer Associates) are utterly clueless and incompetent... no, I *will* have a short rant: the damn stuff is a tarball, *not* a package, *and* it knows nothing whatever about selinux, so forget having a CA-provided SiteMinder selinux policy... anyway, that stuff is installed in /opt, and with no policy... do I have to do what I was doing before, with special local policies on the machines it runs on, to make setroubleshoot stop complaining it can't find the .rpm, or...?
mark