| Summary: | 3_a1: user 'halt': program '/sbin/halt' does not exist | ||
|---|---|---|---|
| Product: | Mageia | Reporter: | Bit Twister <bittwister2> |
| Component: | RPM Packages | Assignee: | Mageia Bug Squad <bugsquad> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | Normal | CC: | mageia, timpul |
| Version: | Cauldron | ||
| Target Milestone: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | 3alpha2 | ||
| Source RPM: | systemd-189-4.mga3.src.rpm | CVE: | |
| Status comment: | |||
|
Description
Bit Twister
2012-09-08 02:13:42 CEST
Opps, work around solution should be ln -s /bin/systemctl halt there was a thread on the mageia-dev ml about that iirc, colin something for you ? Whiteboard:
(none) =>
3alpha1 Hmm, this should have been after mga3a1, as this was only pushed in yesterday (after release). Anyway, the binaries were move to /usr/bin as there was no need for the consolehelper wrappers that were in there. I guess somethings are still calling halt with an absolute path. Can you tell me what actually triggered the error. Obviously calling halt from the command line should work fine. You should see the error if you run steps 2 through 4 of test procedure. sorry, that should be steps 2 through 4 of Steps to Reproduce. Gah, I'm blind. Ahh right, so it's because the "halt" user has /sbin/halt set as it's shell. Is the halt user even used these days? I think it was typically used such that you could login as the halt user and have the machine power down.. these days normal users can be granted permissions to halt the machine in different ways so I think I'd propose just dropping the user. What do you reckon? (In reply to comment #6) > Is the halt user even used these days? You'll need to ask sys admins. > I think it was typically used such that you could login as the halt user and > have the machine power down.. these days normal users can be granted > permissions to halt the machine in different ways so I think I'd propose just > dropping the user. What do you reckon? I am just a dumb user, whatever you/upstream decide is fine with me. cd /sbin ln -s /bin/systemctl halt is an easy workaround to suppress the pwck -r complaint. (In reply to comment #8) > bug still valid ? looks like it to me # pwck -r user 'adm': directory '/var/adm' does not exist user 'halt': program '/sbin/halt' does not exist user 'news': directory '/var/spool/news' does not exist user 'uucp': directory '/var/spool/uucp' does not exist user 'avahi': directory '/var/avahi' does not exist user 'avahi-autoipd': directory '/var/avahi' does not exist user 'vboxadd': directory '/var/run/vboxadd' does not exist
Bit Twister
2012-10-20 22:01:19 CEST
Keywords:
NEEDINFO =>
(none)
Manuel Hiebel
2012-10-20 22:15:40 CEST
Whiteboard:
3alpha1 =>
3alpha2 I think the general consensus was to remove the halt user, tho' I don't think I actually actioned that yet. When this is done, pwck will still produce the same results for upgrades, but new installs should be fine. 3_alpha3 workaround:
cd /sbin
ln -s /usr/bin/systemctl halt
I'm afraid 'halt' is no longer used. Use 'poweroff' instead. CC:
(none) =>
timpul (In reply to Robert Ottlovich from comment #12) > I'm afraid 'halt' is no longer used. Use 'poweroff' instead. I can agree with you as far as the command line is concerned, run the command grep halt /etc/passwd Gah, I clearly forgot to do this as per comment 10. *facepalm* OK, finally committed to setup pkg in cauldron svn. Sorry for forgetting :( In the same vein of tidying up pointless users however, do you think we should also nuke the "sync" user? (In reply to Colin Guthrie from comment #15) > OK, finally committed to setup pkg in cauldron svn. Sorry for forgetting :( Don't worry about it. I have a workaround/patch script for those kinds of bugs. Of course changing the Assigned To: bug report field will keep you from forgetting. :-D > In the same vein of tidying up pointless users however, do you think we > should also nuke the "sync" user? I am just a lowly user and do not know the ramifications of such a change. problem solved in mga 4 alpha1 Status:
NEW =>
RESOLVED commit c65674bb929b03c3521be7c878077fd8e823dea9
Author: Colin Guthrie <colin@...>
Date: Wed Aug 31 11:40:28 2016 +0200
Nuke the 'shutdown' and 'halt' users (mga#7389)
They serve no purpose these days
---
Commit Link:
http://gitweb.mageia.org/software/setup/commit/?id=c65674bb929b03c3521be7c878077fd8e823dea9
|