Description of problem: 6_s2 The following are the dangling links: /usr/sbin/ip6tables-restore-translate -> xtables-compat-multi /usr/sbin/ip6tables-translate -> xtables-compat-multi /usr/sbin/iptables-restore-translate -> xtables-compat-multi /usr/sbin/iptables-translate -> xtables-compat-multi Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Clean install of Mageia-6-sta2-x86_64-DVD.iso with enabled non-free and tainted media. 2. Pick Custom package install 3. in Package Group Selection screen set all package groups selected except Other Graphical Desktops. In that one select Gnome, Plasma and Xfce. Complete install, reboot 4. click up a terminal 5. su - root 6. urpmi symlinks 7. apply all updates, then run 8. symlinks -r / | grep dangling
Status comment: (none) => 6_s2
$ urpmq -g iptables System/Kernel and hardware/iptables So assigning to the Kernel maintainers
CC: (none) => marja11Assignee: bugsquad => kernel
Please don't put 6_s2 in the status comment field, it's not a status comment. A status comment is supposed to be a short summary of the current state of the bug. As for the bug, those four links showed up as new files in the iptables 1.6.1 release. I have no idea what the xtables-compat-multi that they're pointing at is supposed to be.
Status comment: 6_s2 => (none)
(In reply to David Walser from comment #2) > Please don't put 6_s2 in the status comment field, it's not a status > comment. A status comment is supposed to be a short summary of the current > state of the bug. > Well, he was told to neither add that to summary of the bug report, nor to use the "6sta2" keyword (unless it's a iso bug). However, I think using the keyword is OK, because iso bugs will have the release component or the installer component and/or be filed against stage2 or draklive-install. For me it is easier to see when a cauldron bug was last confirmed to be valid if that keyword is set, than when it isn't and I need to wade through the comments. Let's see whether anyone complains if I set the 6sta2 keyword.
Keywords: (none) => 6sta2
OK, from looking at iptables/Makefile.am, I see that xtables-compat-multi is supposed to be installed if ENABLE_NFTABLES is enabled, and it is. Also, from the build log, it loos like xtables-compat-multi *is* installed in /sbin: http://pkgsubmit.mageia.org/uploads/done/cauldron/core/release/20170128111738.akien.duvel.5222/iptables-1.6.1-1.mga6/build.0.20170128111802.log So I don't know why RPM didn't complain about it not being in the files list.
(In reply to David Walser from comment #2) > Please don't put 6_s2 in the status comment field, it's not a status > comment. Seems like it is to me. I did post on dev/QA discussion list and no one really seemed to have a problem with me using it. > A status comment is supposed to be a short summary of the current > state of the bug. Well, I thought that was a pretty "short summary" of the state of the bug. I am trying very hard to help reduce the work load on anyone wanting to know if the bug is still valid. But, hey, I have no problem modifying my script to use "still fails on 6_s2".
Installing "iptables-nft" fixes this dangling links issue: $ urpmf /xtables-compat-multi iptables-nft:/sbin/xtables-compat-multi $ So I think that main iptables package should require iptables-nft, WDYT?
CC: (none) => geiger.david68210
Or moved the four following files to the subpackage "iptables-nft": /usr/sbin/ip6tables-restore-translate /usr/sbin/ip6tables-translate /usr/sbin/iptables-restore-translate /usr/sbin/iptables-translate
Fixed in next iptables update! - iptables-1.6.1-2.mga6
Status: NEW => RESOLVEDResolution: (none) => FIXED