FS#25819 - [initscripts] add shutdown_preumount hook

Attached to Project: Arch Linux
Opened by Devin J. Pohly (djpohly) - Monday, 29 August 2011, 22:17 GMT
Last edited by Tom Gundersen (tomegun) - Saturday, 26 November 2011, 12:00 GMT
Task Type Feature Request
Category Arch Projects
Status Closed
Assigned To Tom Gundersen (tomegun)
Architecture All
Severity Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description:
rc.shutdown has a postumount hook but not a preumount hook. In contrast, rc.sysinit has both a premount and postmount hook.

preumount would be useful for filesystem monitoring software which needs to capture all changes to the filesystem. Using the next earliest hook, it would miss writes to the random seed and wtmp files.


Additional info:
* initscripts-git
This task depends upon

Closed by  Tom Gundersen (tomegun)
Saturday, 26 November 2011, 12:00 GMT
Reason for closing:  Implemented
Comment by Tom Gundersen (tomegun) - Tuesday, 30 August 2011, 15:51 GMT
In principle another hook is ok. Please post a git formatted patch to arch-projects@archlinux.org (you have to sign up), so people can comment.
Comment by Devin J. Pohly (djpohly) - Wednesday, 09 November 2011, 01:06 GMT
  • Field changed: Percent Complete (100% → 0%)
Patch was posted [http://mailman.archlinux.org/pipermail/arch-projects/2011-August/001780.html]; did I need to say something here too?
Comment by Tom Gundersen (tomegun) - Wednesday, 09 November 2011, 01:07 GMT
Sorry about this, I must have lost track of that patch. Thanks for the link, I'll follow up on the mailignlist before the December release.
Comment by Devin J. Pohly (djpohly) - Wednesday, 09 November 2011, 01:22 GMT
No problem, was wondering what happened to it. :)
Comment by Tom Gundersen (tomegun) - Saturday, 26 November 2011, 11:58 GMT
Merged! Sorry for taking so long for such an obvious patch ;-)

I was hesitating a bit as now (due to some recent shuffling) the hook will coincide with the postkillall hook. However, I decided it would be best if hook authors don't have to worry about changes to initscripts and just rely on the semantics of the hooks as described in /etc/rc.d/functions.

Loading...