Newbie, found bug and possible fix in SW 14 /etc/rc.d/rc.6 (LVM)
 

Hello,

I am learning linux, since I'm unemployed and have little better to do. I chose Slackware 14 after a few distros, and I like it - it's a bastard just like me!

I used the standard menu installation, all packages, and I had to modify the lilo.conf and fstab, since I'm mixing my PATA and SATA drives out of sheer stubbornness. I'm ctrl-alt-deleting from the main console (not in X), and the reboot locks up after stdout message " 0 logical volume(s) in volume group "vol_grp1" now active".

I added "set -x" to the /etc/rc.d/rc.6 script so I could get some more detail, and I see it's locking up on line 247:

I looked at the options for vgchange, and it included --noudevsync, which made me wonder if udevd had been shut down already. At line 171 it was:

So I added "--noudevsync" to the vgchange arguments, since udevd was off anyway, and it fixed the reboot problem - no more hang.

Being a newb, (first post), I didn't want to log this anywhere else.

FYI, for any fixers.

Slackware 14 64, kernel 3.2.29 with SMP on AMD.
rc.6 version: 2.47 Sat Jan 13 13:37:26 PST 2001
author: Miquel van Smoorenburg, Patrick J. Volkerding.
multiuser mode.

May be the fact that I haven't mounted any of the LVMs yet, but I was trying to just get a full boot/reboot cycle to work before I added anything beyond a boot drive.

Jeff

I am just a little unclear on what the problem is.....At one point in you post you said that you had fixed the problem. Please clarify!:D

Cheers,
Nbiser

Hi Nbiser,

The problem was that my Slackware 14 installation would not reboot when I pressed ctrl-alt-delete. It started the shutdown cycle and hung after the message " 0 logical volume(s) in volume group "vol_grp1" now active" displays on the screen. I was able to isolate what caused the hang. It was because the /etc/rc.d/rc.6 script first shuts down udevd, and then tries to shut down lvm. But the lvm daemon tries to sync with the udev daemon, so I added --noudevsync to the lvm shutdown command vgchange, and the reboot was fixed. My installation now reboots perfectly.

I was hoping someone who owned the rc.6 script could modify it more appropriately than I could. I may be the only one experiencing this bug. It may only have occurred because I had no active/mounted logical volumes when I tried to go to level 6.

I probably just made it more confusing. Sorry.

I'd post the rc.6 script, but I haven't got any communication up between my Windows computer and my linux box yet. I'm trying to isolate installation issues in Slackware. Maybe I can enhance the Slackware installer? It needs enhancement for us weirdos with crazy boxes.

I've never had this problem. However, it seems to me that you have fixed it. As long as the system continues to function properly you should be OK.