Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 27836 - apcupsd non-NIS net slave abends immediately with no error or message generated.
Summary: apcupsd non-NIS net slave abends immediately with no error or message generated.
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] Core system (show other bugs)
Hardware: x86 Linux
: High major (vote)
Assignee: Martin Holzer (RETIRED)
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2003-09-02 21:13 UTC by Steve Kutnar
Modified: 2003-12-04 14:21 UTC (History)
3 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Steve Kutnar 2003-09-02 21:13:54 UTC
I have two Gentoo Linux 1.4 systems sharing an APC SmartUPS.

One machine is directly connected to the UPS via a serial connection - the master.  Apcupsd works fine on this machine when configured with a UPSCLASS of standalone or netmaster and a UPSMODE of disable or net.

The second machine derives power from the UPS, but has no direct connection.  So, this machine should be a net slave.  I'm not using the NIS master/slave setup.

Here are the config parameters (/etc/apcupsd/apcupsd.conf) for the master:
UPSCABLE 940-0024C
UPSTYPE smartups
DEVICE /dev/ttyS0
LOCKFILE /var/lock
BATTERYLEVEL 5
MINUTES 3
TIMEOUT 0
ANNOY 300
ANNOYDELAY 60
NOLOGON disable
KILLDELAY 0
NETSERVER on
NISIP 0.0.0.0
NISPORT 3351
EVENTSFILE /var/log/apcupsd.events
EVENTSFILEMAX 10
UPSCLASS netserver
UPSMODE net
NETTIME 60
NETPORT 6666
SLAVE <slave-ip>
STATTIME 0
STATFILE /var/log/apcupsd.status
LOGSTATS off
DATATIME 0

...and for the slave:
UPSCABLE ether
UPSTYPE smartups
DEVICE /dev/ttyS0
LOCKFILE /var/lock
BATTERYLEVEL 0
MINUTES 0
TIMEOUT 0
ANNOY 300
ANNOYDELAY 60
NOLOGON disable
KILLDELAY 0
NETSERVER on
NISIP 0.0.0.0
NISPORT 3351
EVENTSFILE /var/log/apcupsd.events
EVENTSFILEMAX 10
UPSCLASS netslave
UPSMODE net
NETTIME 60
NETPORT 6666
MASTER <master-ip>
STATTIME 0
STATFILE /var/log/apcupsd.status
LOGSTATS off
DATATIME 0

Note that I tried adjusting UPSCABLE, DEVICE, BATTERYLEVEL, MINUTES, and NETSERVER to see if any of those had an effect on the situation, but they did not.

I didn't make this a critical bug because there is the possibility of using NIS instead of this simpler master/slave configuration.  However, someone expecting or relying on this particular method are out of luck for now.

Reproducible: Always
Steps to Reproduce:
1. Configure master and slave with the respective apcupsd.conf files listed.
2. Do normal rc-update procedure for apcupsd on master and run via "/etc/init.d/apcupsd start".
3. Just try launching apcupsd directly from the CLI on the slave. --OR-- also handle like master.

Actual Results:  
Slave apcupsd returns immediately to the command-line without error.
Doing a "ps aux" shows that apcupsd is not running.
/var/run/apcupsd.pid contains a PID that doesn't exist.
/var/log/apcupsd.events is created, but empty.
No information logged to /var/log/messages, syslog, or daemon.log

If run via /etc/init.d, apcupsd on the slave also abends, but leaves behind the entry in /var/lib/init.d/started.

Expected Results:  
If run via the CLI, apcupsd should not return to the command-line under normal circumstances.

If run via /etc/init.d, it should behave as most daemons.
Comment 1 Steve Kutnar 2003-11-15 16:42:01 UTC
As of today (11/15/2003) I no longer experience this problem.

I did a system update and apcupsd is now at version 3.10.6.  The prior version was 3.10.5-r3.
Comment 2 Heinrich Wendel (RETIRED) gentoo-dev 2003-12-04 14:21:28 UTC
fine, so this is fixed now :)