Description of problem: I accidentally typed the same option twice and sealert refused to work. The message is kind of funny.: "cannot run as browser, already selected browser". Version-Release number of selected component (if applicable): setroubleshoot-server-2.2.94-1.el6.i686 setroubleshoot-plugins-2.1.60-1.el6.noarch setroubleshoot-2.2.94-1.el6.i686 setroubleshoot-doc-2.2.94-1.el6.i686 How reproducible: always Steps to Reproduce: # sealert -a -a cannot run as analyze, already selected analyze -b --browser Launch the browser -h --help Show this message -s --service Start sealert as a dbus service -S --noservice Start sealert without dbus service as stand alone app -l --lookupid id Lookup alert by id, id may be wildcard * to lookup all alerts -a --analyze file Scan a log file, analyze it's AVC's -H --html_output Ouput in html -v --verbose log INFO level and higher messages to console -V --debug log DEBUG level and higher messages to console -u --user logon user name -p --password logon password -f --fix fix avc with the given uuid # sealert -s -s cannot run as service, already selected service -b --browser Launch the browser -h --help Show this message -s --service Start sealert as a dbus service -S --noservice Start sealert without dbus service as stand alone app -l --lookupid id Lookup alert by id, id may be wildcard * to lookup all alerts -a --analyze file Scan a log file, analyze it's AVC's -H --html_output Ouput in html -v --verbose log INFO level and higher messages to console -V --debug log DEBUG level and higher messages to console -u --user logon user name -p --password logon password -f --fix fix avc with the given uuid # sealert -S -S cannot run as noservice, already selected noservice -b --browser Launch the browser -h --help Show this message -s --service Start sealert as a dbus service -S --noservice Start sealert without dbus service as stand alone app -l --lookupid id Lookup alert by id, id may be wildcard * to lookup all alerts -a --analyze file Scan a log file, analyze it's AVC's -H --html_output Ouput in html -v --verbose log INFO level and higher messages to console -V --debug log DEBUG level and higher messages to console -u --user logon user name -p --password logon password -f --fix fix avc with the given uuid # sealert -b -b cannot run as browser, already selected browser -b --browser Launch the browser -h --help Show this message -s --service Start sealert as a dbus service -S --noservice Start sealert without dbus service as stand alone app -l --lookupid id Lookup alert by id, id may be wildcard * to lookup all alerts -a --analyze file Scan a log file, analyze it's AVC's -H --html_output Ouput in html -v --verbose log INFO level and higher messages to console -V --debug log DEBUG level and higher messages to console -u --user logon user name -p --password logon password -f --fix fix avc with the given uuid Actual results: sealert terminates with an error message Expected results: sealert works as expected, duplicated option is ignored
sealert -a -a cannot run as analyze, already selected analyze -b --browser Launch the browser -h --help Show this message -s --service Start sealert as a dbus service -S --noservice Start sealert without dbus service as stand alone app -l --lookupid id Lookup alert by id, id may be wildcard * to lookup all alerts -a --analyze file Scan a log file, analyze it's AVC's -H --html_output Ouput in html -v --verbose log INFO level and higher messages to console -V --debug log DEBUG level and higher messages to console -u --user logon user name -p --password logon password -f --fix fix avc with the given uuid This is what I am seeing in F14. I am not sure what you would want the error message to say.
I would like to see sealert doing the analysis instead of printing an error message. I didn't choose two antagonistic functions of sealert. sealert should quietly ignore other occurrences of '-a' on the command line, because they do not conflict with each other.
-a expects a file to be handed to it.
OK, I'm taking back my note about '-a', it works fine when a file is attached to it. However, sealert still pretends that duplicated '-b' or '-s' or '-S' are in conflict.
I would prefer to push off setroubleshoot to see how well the redesign in Fedora 15 goes, and then update to that package in 6.2.
Fixed in setroubleshoot-3.0.31-1.el6
Clearing qa_ack, setroubleshoot has been removed from 6.2 approved list.
Well, the question is whether this is a bug. If yes, I don't think this is so important and we can move it to RHEL6.3.
I agree.
We need to backport a patch from Fedora.
Fixed in setroubleshoot-3.0.45-1.el6
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2012-0781.html