First Last Prev Next    No search results available      Search page      Enter new bug
Bug#: 87265
Alias:
Product:
Component:
Status: RESOLVED
Resolution: FIXED
Assigned To: Gentoo's Team for Core System packages <base-system@gentoo.org>
Hardware:
OS:
Version:
Priority:
Severity:
Reporter: Neil Darlow <neil@darlow.co.uk>
Add CC:
CC:
Remove selected CCs
URL:
Summary:
Status Whiteboard:
Keywords:

Filename Description Type Creator Created Size Actions
Create a New Attachment (proposed patch, testcase, etc.) View All

Bug 87265 depends on: Show dependency tree
Bug 87265 blocks:
Votes: 0    Show votes for this bug    Vote for this bug

Additional Comments: (this is where you put emerge --info)


Not eligible to see or edit group visibility for this bug.






View Bug Activity   |   Format For Printing   |   XML   |   Clone This Bug

Description:   Opened: 2005-03-30 06:02 0000 
The /etc/init.d/halt script installed by apcupsd performs a two minute wait and
exit 1 after issuing the killpower command to the UPS.

This essentially means the final halt command isn't executed under powerfail
conditions and software RAID superblocks won't be updated.

As a member of the apcupsd development team, I've had to report this bug on
every distribution I've used. The apcupsd project no longer advocates the wait
and exit 1 under powerfail conditions because of the problems it causes.

Reproducible: Always
Steps to Reproduce:
1.
2.
3.

------- Comment #1 From Frank 2005-04-23 06:45:53 0000 ------- 
So what are you recommending /etc/init.d/halt does instead ?

------- Comment #2 From Neil Darlow 2005-04-23 07:06:21 0000 ------- 
/etc/init.d/halt should do the following:

1) Issue apcupsd --killpower
2) Terminate as it normally would

The RAID superblocks will be updated and the system will do whatever it should for a halt condition i.e. display "Power down" or power itself off.

The UPS will power itself off after its configured KILLDELAY setting expires.

The apcupsd documentation (which has been recently revised) discusses powerdown situations and configuration for powerdown in detail.

Regards,
Neil Darlow

------- Comment #3 From Neil Darlow 2005-04-29 12:39:13 0000 ------- 
During the conversion of my system to UTF-8, I upgraded baselayout from 1.9.4
to 1.11.11 and noticed that /etc/init.d/halt.sh in the newer version does
exactly what I described for apcupsd during halt.

In fact, I can't see that /etc/init.d/halt is actually executed as part of the
halt process (correct me if I'm wrong).

Regards,
Neil Darlow

------- Comment #4 From SpanKY 2005-04-29 14:10:43 0000 ------- 
i dont have an /etc/init.d/halt on my system ... or any of my systems ...

they're all /etc/init.d/halt.sh

------- Comment #5 From SpanKY 2005-04-29 14:12:10 0000 ------- 
actually i lied, i managed to find 1 system which had it ... it isnt part of
baselayout and is never executed by baselayout

so if halt.sh (which is part of baselayout) does the right thing, can we punt
this bug ?

------- Comment #6 From Neil Darlow 2005-04-29 14:40:29 0000 ------- 
Looks like /etc/init.d/halt is installed by apcupsd (and should be removed).

Somewhere after baselayout-1.9.4, support for apcupsd was added to /etc/init.d/halt.sh so systems running that version won't turn the UPS off.

Either some backporting, or some effort made to get the appropriate newer version of baselayout marked stable, is required.

Regards,
Neil Darlow

------- Comment #7 From SpanKY 2005-04-29 16:19:39 0000 ------- 
does apcupsd still install /etc/init.d/halt ?  if so, we should fix the package

------- Comment #8 From SpanKY 2005-05-04 19:31:28 0000 ------- 
apcupsd no longer installs /etc/init.d/halt
First Last Prev Next    No search results available      Search page      Enter new bug