Bug 480668
Summary: | start cups on demand using systemd | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Bill Nottingham <notting> | ||||||||
Component: | cups | Assignee: | Tim Waugh <twaugh> | ||||||||
Status: | CLOSED DUPLICATE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||||
Severity: | medium | Docs Contact: | |||||||||
Priority: | low | ||||||||||
Version: | rawhide | CC: | adel.gadllah, harald, j.kopkova, jpazdziora, k.georgiou, michael.monreal, pachoramos1, rdieter, rvokal, tuju, twaugh, ykopkova | ||||||||
Target Milestone: | --- | Keywords: | FutureFeature | ||||||||
Target Release: | --- | ||||||||||
Hardware: | All | ||||||||||
OS: | Linux | ||||||||||
Whiteboard: | |||||||||||
Fixed In Version: | Doc Type: | Enhancement | |||||||||
Doc Text: | Story Points: | --- | |||||||||
Clone Of: | Environment: | ||||||||||
Last Closed: | 2011-06-07 08:29:24 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: | |||||||||||
Attachments: |
|
Created attachment 329384 [details]
xinetd service file
Created attachment 329385 [details]
SELinux policy module
SELinux policy requires changes for cups to work when it's launched from xinetd so that it can properly handle the socket it's passed.
ping? Busy with more pressing stuff at the moment... Too risky a change until 1.4 has settled down a lot more. This bug appears to have been reported against 'rawhide' during the Fedora 11 development cycle. Changing version to '11'. More information and reason for this action is here: http://fedoraproject.org/wiki/BugZappers/HouseKeeping (In reply to comment #5) > Too risky a change until 1.4 has settled down a lot more. ping? This will be a *lot* easier to do once mDNS is the default discovery and browse mechanism in CUPS. That won't be until CUPS 1.5 though. The basic problem is that if cupsd is not around to hear CUPS Browse packets it might be a long time (minutes) after cupsd is started before they are re-sent. So the user might be sat in front of a print dialog for several minutes before they can get on with printing. With mDNS this would not be a problem as avahi would be already be doing the listening. Tim - given systemd, should this bug be closed in favor of a systemd implementation, or at least retitled/retargeted? Yes, retitling. We are fairly close to full Bonjour support now, and Lennart has told me he has patches for systemd integration in cupsd. possible duplicate bug 690766 *** This bug has been marked as a duplicate of bug 690766 *** |
Created attachment 329383 [details] cups changes required Description of problem: (Filing for completeness sake - have already talked a bit with Tim about this over e-mail) Looking at cups, it has code to start on demand via launchd when a connection attempt is made to the cups socket address. It's possible to use a similar mechanism to start cups on demand via xinetd, so it can be started on demand instead of always running. Open issues: - you trade starting cups always for starting xinetd always - you need to do 'something' for when you share printers - the xinetd file requires lockstep agreement with any 'Listen' statements in cupsd.conf Version-Release number of selected component (if applicable): cups-1.4-0.b2.2