Apcupsd UPS control software

Thanks Adam,
Your intructions were great!


On Sat, Apr 19, 2008 at 1:49 PM, Adam Kropelin <adk0212@...> wrote:

> Eloy Fonseca wrote:
> > Hi everyone,
> >
> > I have installed the apcupsd in my iMac and now i want to remove it
> > because OSX can no longer see my APC back up. How is the best way to
> > remove the application and restore OSX nobreak features?
>
> It is sad that Apple's package installer does not support the concept of
> uninstall! I have been meaning to write a simple uninstall script to work
> around that problem, but for now you can do this (as root) at a terminal:
>
>    rm -rf /Library/StartupItems/apcupsd
>    rm -rf /System/Library/Extensions/ApcupsdDummy.kext
>    rm -rf /System/Library/Extensions.mkext
>    rm -rf /System/Library/Extensions.kextcache
>
> Then reboot...
>
> --Adam
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
> Don't miss this year's exciting event. There's still time to save $100.
> Use priority code J8TL2D2.
>
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Apcupsd-users mailing list
> Apcupsd-users@...
> https://lists.sourceforge.net/lists/listinfo/apcupsd-users
>

Artem Volk wrote:
> Gentlemen,
> 
> Thank you for a quick replies for my request and sorry for delayed
> response--we have Easter here in Ukraine.
> 
> It seems that reduced battery lifetime was caused by low line voltage.
> My UPS has the following settings (note: 220V is the normal voltage):
> 
> LOTRANS : 195.0 Volts
> HITRANS : 255.0 Volts
> 
> But I noticed ONLINE BOOST status when the LINEV was 210 Volts.

This might not be unusual. This means if the line voltage goes below 195
volts, it turns the boost on. There is hysteresis in there to keep it from
clicking back and forth. If the boost is 15% (I think it is), then the line
voltage can go up about 15% before the boost is turned off. Similarly at the
high end. So it will go up to 212.75 volts before the boost is turned off.
N.B.: the percentage of boost and hysteresis are just my recollection of the
numbers, and are perhaps not that exact anyway. But if my recollections are
correct, what you see would be normal.

> The
> UPS clicks (like when transferring to batteries) in this moments, but
> the LED remains green. No beeps are generated too. Is this a normal
> behaviour? 

I do not remember if it beeps or not when starting smart boost. If it does,
it is probably a single beep. I know it does not beep all the time when it
does a smart boost.

I am not sure if there is a difference between what you mean by ONLINE BOOST
and what APC calls Smart Boost. This may be a difference between what APC
calls it and what apcupsd calls the same thing, or maybe you have a
different feature altogether.

This may matter. I know what APC calls Smart Boost _does not_ use the
battery. If what you call ONLINE BOOST does use the battery, it is something
that I do not find in the APC data I have. Anyhow, APC have this to say
about this:

Transfer points
The Smart-UPS may be configured to transfer to on-battery operation over a
wide range of utility voltages. Loads designed to operate
over a wide input voltage range can be protected without the annoyance of
frequent transfers.
Understanding Smart-UPS transfer points is not as easy as it seems though
not really difficult, either.
Let's take SUA1000XL as an example. With its nominal voltage specified at
120V, the available upper transfer point options are 127,
130, 133 and 136V. The available lower transfer points are 106, 103, 100 and
97V.
At the same time the specification for SUA1000XL (which at the moment this
application note was being prepared could be found at
http://www.apc.com/resource/include/techspec_index.cfm?base_sku=SUA1000)
states the input voltage range as:
Table 1 ­ Input voltage ranges for SUA1000XL
Input voltage range for mains operations                                82-144 V
Input voltage adjustable range for mains operation
75-154 V
What is the relationship between these numbers?
The key fact here is that the upper and lower transfer points specify the
maximum and minimum output voltages, respectively. E.g.,
setting the upper transfer point to 127V and the lower transfer point to
106V would guarantee that the load will never see the voltage
beyond the range of 106 to 127V.
Let's consider a scenario. Suppose the upper transfer point of a SUA1000XL
is set to 127V and the lower transfer point is set to 106V.
At the beginning let input voltage be 120V, hence the output voltage of
SUA1000XL being 120V as well. Then the input voltage begins
to decrease. The output voltage decreases along with the input voltage until
it hits the lower threshold of 106VAC. At this point the
Automatic Voltage Regulator (AVR) kicks in and boosts the output voltage by
a certain extent, which is 15% in the case of SUA1000XL.
The output voltage rises up to 106 x 1.15 = 121.9V. If the input voltage
continues to decrease, the output voltage follows it, staying 15%
higher than the input voltage. But once the output voltage hits the lower
threshold of 106V again, the input voltage at this point being
92.2V, the AVR doubles the boost (in models where equipped, and enabled by
operating conditions) by yet another 15% (in case of
SUA1000XL). The output voltage rises to 92.2 x 1.30 = 119.8VAC again. Should
the input voltage keep on decreasing, the output
voltage would also decrease proportionately until it hits the lower
threshold of 106VAC yet one more time. At this point the input voltage
is 82VAC, and the AVR has no more taps to switch to. Only now the UPS goes
to battery and the output voltage rises to 120V.
On the opposite side, when the voltage increases above nominal, Smart-UPS
works in a similar way. The AVR in this case makes only
one attempt to trim the input voltage (by 13% in case of SUA1000XL) in order
to keep the output voltage below the threshold of
127VAC. If the input voltage increases further the UPS goes to battery,
generating output voltage of 120V.

> 
> If the ONLINE BOOST is a abnormal status may be this status change
> should be logged?

IIRC, Smart Boost is a normal status event that is logged.

> I hope this will help many people to save their
> batteries.
> 
> P.S. I'd like to thanks Adam for a well-written, good documented and
> really supported software. Hope, my apcupsd integration module for
> phpSysInfo can be counted as a small gratitude.
> 



-- 
  .~.  Jean-David Beyer          Registered Linux User 85642.
  /V\  PGP-Key: 9A2FC99A         Registered Machine   241939.
 /( )\ Shrewsbury, New Jersey    http://counter.li.org
 ^^-^^ 14:35:01 up 40 days, 19:37, 2 users, load average: 4.77, 4.20, 4.07

Well, I guess confessional debugging even works on a mailing list, I solved 
the problem shortly after posting:


I'll post my solution for anyone who runs into this type of error.

What I was missing was the kernel driver "Device Drivers->HID 
Devices->/dev/hiddev raw HID device support" aka CONFIG_USB_HIDDEV=y in 
your .config.

If you read the help on this it basically says its a driver for USB devices 
which are technically not user interfaces such as monitor controls and UPSs.

Hope this helps someone someday,

Dave


On April 24, 2008, David Flogeras wrote:
> Hello,
>
> I have an APC BX900R UPS, which uses a USB connection to talk to the host.
>
> I had previously been using the UPS with an embedded Network Storage unit,
> which I know runs apcupsd 3.10.15.  It had no trouble detecting the UPS and
> communicating with it via the USB cable.
>
> Now I am using the UPS on a Gentoo Linux box with apcupsd-3.12.4, but it
> keeps failing with:
>
> Thu Apr 24 23:13:45 ADT 2008  apcupsd FATAL ERROR in linux-usb.c at line
> 649 Cannot find UPS device --
> For a link to detailed USB trouble shooting information,
> please see <http://www.apcupsd.com/support.html>.
>
>
>
> The config file is very simple, and was taken from a different gentoo box
> running the same version of apcupsd, but with a different model UPS
> (BackUPS650). Basically (the important stuff ) is simply:
>
> UPSCABLE usb
> UPSTYPE usb
> DEVICE
>
>
>
> I tried upgrading to apcupsd-3.14.2, but to no avail. The same thing
> happens, except the "FATAL ERROR in linux-usb.c" moves to line 600 (I
> assume due to code changes in the driver).
>
>
> Can anyone shed some light on what might be going wrong? My system is far
> from exotic, its an older (2004ish) well supported Asus motherboard, and I
> regularly use USB for thumbdrives, and my mp3 player.
>
> I am running 2.6.24 kernel.
>
>
> Thanks for any help,
>
> Dave
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
> Don't miss this year's exciting event. There's still time to save $100.
> Use priority code J8TL2D2.
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/java
>one _______________________________________________
> Apcupsd-users mailing list
> Apcupsd-users@...
> https://lists.sourceforge.net/lists/listinfo/apcupsd-users

Artem Volk wrote:
> Gentlemen,
>
> I've Back-UPS ES 525 bought three monthes ago. I use it to protect
> FreeBSD box with apcupsd installed. All data I mention below is
> gathered using apcaccess 3.14.2.
>
> I monitor the UPS and noticed that time remaining on battery degrades
> from 42 minutes to 20 minutes two weeks ago, then to 20 minutes, and
> now for approximately 3 minutes. The load is the same --
> approximately 17%,
> the battery charge level is 100%.
>
> During its life the UPS has three big power outages (discharge to
> 10%) and it seems that it have "ON-LINE BOOST" state very often.
> acptest do not have calibrate option for my UPS.
>
> Why the battery became weak so fast?

The BOOST status indicates that the UPS believes the line voltage is below 
spec and is using battery power to boost it. This may explain the reduced 
battery runtime figure.

Try using apctest to reduce the UPS low voltage threshold (you may need 
3.14.3 to see this option).

--Adam

----- Original Message ----- 
From: "Adam Kropelin" <akropel1@...>
To: "Apcupsd Discussion List" <apcupsd-users@...>
Sent: Tuesday, April 22, 2008 14:35
Subject: Re: [Apcupsd-users] Time on battery degrades very fast


> Artem Volk wrote:
>> Gentlemen,
>>
>> I've Back-UPS ES 525 bought three monthes ago. I use it to protect
>> FreeBSD box with apcupsd installed. All data I mention below is
>> gathered using apcaccess 3.14.2.
>>
>> I monitor the UPS and noticed that time remaining on battery degrades
>> from 42 minutes to 20 minutes two weeks ago, then to 20 minutes, and
>> now for approximately 3 minutes. The load is the same --
>> approximately 17%,
>> the battery charge level is 100%.
>>
>> During its life the UPS has three big power outages (discharge to
>> 10%) and it seems that it have "ON-LINE BOOST" state very often.
>> acptest do not have calibrate option for my UPS.
>>
>> Why the battery became weak so fast?
>
> The BOOST status indicates that the UPS believes the line voltage is below
> spec and is using battery power to boost it. This may explain the reduced
> battery runtime figure.
>
> Try using apctest to reduce the UPS low voltage threshold (you may need
> 3.14.3 to see this option).
>
> --Adam
>
Weird. I would think that the BOOST feature stays away from using the 
battery,
or at least tries to. I wonder if the UPS is switching to battery more often 
than
APCUPSD is being informed of? My BR-1500 has switched to battery power
probably over 200 times in the last few month, but only maybe 60 or 70 of
those were actually caught by the computer, simply because the transfers 
were too
quick.

You may also have gotten a bum battery in your unit. That has been known to
happen on occasion.

Cheers!

Mike Shepherd Administrator: http://www.ve7prt.bc.ca CB: The Sheepster (base 
or mobile) Ham: VE7PRT (base/mobile/portable)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Adam Kropelin wrote:
> Artem Volk wrote:
>> Gentlemen,
>>
>> I've Back-UPS ES 525 bought three monthes ago. I use it to protect
>> FreeBSD box with apcupsd installed. All data I mention below is
>> gathered using apcaccess 3.14.2.
>>
>> I monitor the UPS and noticed that time remaining on battery degrades
>> from 42 minutes to 20 minutes two weeks ago, then to 20 minutes, and
>> now for approximately 3 minutes. The load is the same --
>> approximately 17%,
>> the battery charge level is 100%.
>>
>> During its life the UPS has three big power outages (discharge to
>> 10%) and it seems that it have "ON-LINE BOOST" state very often.
>> acptest do not have calibrate option for my UPS.
>>
>> Why the battery became weak so fast?
> 
> The BOOST status indicates that the UPS believes the line voltage is below 
> spec and is using battery power to boost it.

That does not explain it because a Smart-UPS diddles the tap on a
transformer if the line voltage is too low just to avoid using the battery.
APC call that SMART BOOST. It would use the battery only if the line voltage
were so low that changing the transformer tap was not enough. (Similarly if
the line voltage were too high and it used SMART TRIM).

What is your actual line voltage? Mine says this at the moment:

LINEV    : 123.5 Volts

You can look at your system log file (/var/log/messages) for apcupsd events
to see what your line voltage is doing. Mine says stuff like this (in part)

Apr 22 21:16:06 trillian apcupsd[3601]:
122.2,123.5,122.8,55.05,60.00,23.4,35.5,000.0,000.0,122.8,100.0,0
Apr 22 22:16:37 trillian apcupsd[3601]:
122.8,123.5,122.8,55.05,60.00,24.9,35.1,000.0,000.0,122.8,100.0,1
Apr 22 23:17:07 trillian apcupsd[3601]:
122.2,122.8,122.2,55.05,60.00,25.4,35.1,000.0,000.0,122.2,100.0,0

> This may explain the reduced 
> battery runtime figure.
> 
> Try using apctest to reduce the UPS low voltage threshold (you may need 
> 3.14.3 to see this option).
> 
Do not set this arbitrarily. I set mine so the voltage window where it lets
things alone is as wide as possible without going out of the range that the
power supply promises to regulate properly. So if my line voltage goes under
97 volts, SMART BOOST is started. If it goes over 129 volts, SMART TRIM goes
into effect. YMMV.

LOTRANS  : 097.0 Volts
HITRANS  : 129.0 Volts



- --
  .~.  Jean-David Beyer          Registered Linux User 85642.
  /V\  PGP-Key: 9A2FC99A         Registered Machine   241939.
 /( )\ Shrewsbury, New Jersey    http://counter.li.org
 ^^-^^ 23:35:01 up 36 days, 4:37, 2 users, load average: 4.38, 4.28, 4.17
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)
Comment: Using GnuPG with CentOS - http://enigmail.mozdev.org

iD8DBQFIDrDUPtu2XpovyZoRArMjAJ9rihJXMiLmntXFCW87l/+JyT0SNwCgxV7J
Q/1i+q19z3hj3Unb5hhyQYU=
=tq5j
-----END PGP SIGNATURE-----

On Tue, Apr 22, 2008 at 11:45 PM, Jean-David Beyer
<jeandavid8@...> wrote:
>  Adam Kropelin wrote:
>  > Artem Volk wrote:
>  >> Gentlemen,
>  >>
>  >> I've Back-UPS ES 525 bought three monthes ago. I use it to protect
>  >> FreeBSD box with apcupsd installed. All data I mention below is
>  >> gathered using apcaccess 3.14.2.
>  >>
>  >> I monitor the UPS and noticed that time remaining on battery degrades
>  >> from 42 minutes to 20 minutes two weeks ago, then to 20 minutes, and
>  >> now for approximately 3 minutes. The load is the same --
>  >> approximately 17%,
>  >> the battery charge level is 100%.
>  >>
>  >> During its life the UPS has three big power outages (discharge to
>  >> 10%) and it seems that it have "ON-LINE BOOST" state very often.
>  >> acptest do not have calibrate option for my UPS.
>  >>
>  >> Why the battery became weak so fast?
>  >
>  > The BOOST status indicates that the UPS believes the line voltage is below
>  > spec and is using battery power to boost it.
>
>  That does not explain it because a Smart-UPS diddles the tap on a
>  transformer if the line voltage is too low just to avoid using the battery.
>  APC call that SMART BOOST. It would use the battery only if the line voltage
>  were so low that changing the transformer tap was not enough. (Similarly if
>  the line voltage were too high and it used SMART TRIM).

As you say, SMART BOOST is a SmartUPS feature. Artem has a BackUPS. As
far as I know, there is no "boost without using the battery" on a
BackUPS.

--Adam

Adam Kropelin wrote:
> On Tue, Apr 22, 2008 at 11:45 PM, Jean-David Beyer
> <jeandavid8@...> wrote:
>>  Adam Kropelin wrote:
>>  > Artem Volk wrote:
>>  >> Gentlemen,
>>  >>
>>  >> I've Back-UPS ES 525 bought three monthes ago. I use it to protect
>>  >> FreeBSD box with apcupsd installed. All data I mention below is
>>  >> gathered using apcaccess 3.14.2.
>>  >>
>>  >> I monitor the UPS and noticed that time remaining on battery degrades
>>  >> from 42 minutes to 20 minutes two weeks ago, then to 20 minutes, and
>>  >> now for approximately 3 minutes. The load is the same --
>>  >> approximately 17%,
>>  >> the battery charge level is 100%.
>>  >>
>>  >> During its life the UPS has three big power outages (discharge to
>>  >> 10%) and it seems that it have "ON-LINE BOOST" state very often.
>>  >> acptest do not have calibrate option for my UPS.
>>  >>
>>  >> Why the battery became weak so fast?
>>  >
>>  > The BOOST status indicates that the UPS believes the line voltage is below
>>  > spec and is using battery power to boost it.
>>
>>  That does not explain it because a Smart-UPS diddles the tap on a
>>  transformer if the line voltage is too low just to avoid using the battery.
>>  APC call that SMART BOOST. It would use the battery only if the line voltage
>>  were so low that changing the transformer tap was not enough. (Similarly if
>>  the line voltage were too high and it used SMART TRIM).
> 
> As you say, SMART BOOST is a SmartUPS feature. Artem has a BackUPS. As
> far as I know, there is no "boost without using the battery" on a
> BackUPS.
> 
I cannot find a Back-UPS 525 ES on the APC site. But some models of the
Back-UPS seem to have Smart boost and trim. The Back-UPS ES 8 Outlet 650VA
does not have it. The BACK-UPS 1200VA 120V does. The Back-UPS LS and Pro
series seem to as well.

-- 
  .~.  Jean-David Beyer          Registered Linux User 85642.
  /V\  PGP-Key: 9A2FC99A         Registered Machine   241939.
 /( )\ Shrewsbury, New Jersey    http://counter.li.org
 ^^-^^ 10:00:02 up 36 days, 15:02, 2 users, load average: 4.34, 4.28, 4.21

Kirill S. Bychkov wrote:
> Hi everybody!
> I still getting some unpleasant errors from powerflute on OpenBSD.
> Apcupsd is
> 3.14.3.
> When I start it, I see the next things:
> Mutex lock failure. ERR=Resource deadlock avoided
> FATAL ERROR in newups.c at line 71
> and this longs a lot, than it coredumps.
>
> powerflute compiles without any errors and, as I see, with all needed
> flags: c++ -c -fno-exceptions -fno-rtti -O2 -pipe -Wall -pthread
> -I/usr/ports/sysutils/apcupsd/w-apcupsd-3.14.3-ncurses/apcupsd-3.14.3/include
> powerflute.c
> cc -g -O -pthread powerflute.o
> /usr/ports/sysutils/apcupsd/w-apcupsd-3.14.3-ncurses/apcupsd-3.14.3/src/lib/libapc.a
>  -lsupc++ -lncurses -lmenu -lform -lpanel -o powerflute
>
> Sometimes I can see borders of powerflute interface, but they are
> quickly changing to error.

I looked into this a week or two ago for James Dean. I responded privately 
and forgot to forward my message to the list...

Powerflute is unfixable as it stands today. It requires the recently 
removed "raw status" interface over NIS. The "raw status" interface was 
removed because it is broken by design (it exports the raw UPSINFO 
structure, and does so in host byte order). Thus is only works between 
idential apcupsd versions and does not work at all across differing 
architectures.

Even when it did work, Powerflute was not terribly useful. The same 
information is available using apcaccess, multimon, or apctray (win32).

Powerflute can be made to work again if someone modifies it to use the 
ascii status interface instead. Doing that is not high on my priority list. 
But I accept patches :)

--Adam

Hans Georg Giese wrote:
> Hi,
> thanks a lot for your software. It works!
>
> Some hints to improve it:
> - the current package from the Debian Server (fpt.de.debian.org)
> installs without problems and the daemon works. Apctest dies throwing
> some errors (unable to establish lock file, unable to connect to ups,
> syntax errors in device.c).

You must shut down apcupsd before running apctest. I believe the error
messages you get from apctest when apcupsd is still running do suggest this
course of action.

--Adam

On Friday 18 April 2008, Dameon Wagner wrote:

> I think the main issue is the bandwidth, and if I could get apctest to
> talk at 9600, I could probably do everything I want.
 
No. The protorol is changed too. I wrote about it 10/03/08.
(Subject: [Apcupsd-users] Problem with new version of Smart-UPS RT 5000)

I use SNMP Driver now.

-- 
Regards,
Sergey

On Fri, Apr 18, 2008 at 05:13:59PM +0500, Sergey scribbled
 in "Re: [Apcupsd-users] Newer APCs with Serial comms at 9600bps":
> On Friday 18 April 2008, Dameon Wagner wrote:
> 
> > I think the main issue is the bandwidth, and if I could get apctest to
> > talk at 9600, I could probably do everything I want.
>  
> No. The protorol is changed too. I wrote about it 10/03/08.
> (Subject: [Apcupsd-users] Problem with new version of Smart-UPS RT 5000)

Ah, that sucks...

> I use SNMP Driver now.

I use SNMP for a few of our boxes too, but I'd still like the
possibility for normal/native control/monitoring as I currently do
for all the other UPS' we have.  I'm guessing that it's being worked
on in CVS?  I'm in no rush, but thought I'd mention it anyway, just
incase no-one else had come across it.

Cheers.

Dameon.

Dameon Wagner wrote:
> On Fri, Apr 18, 2008 at 05:13:59PM +0500, Sergey scribbled
> in "Re: [Apcupsd-users] Newer APCs with Serial comms at 9600bps":
>> On Friday 18 April 2008, Dameon Wagner wrote:
>>
>>> I think the main issue is the bandwidth, and if I could get apctest
>>> to talk at 9600, I could probably do everything I want.
>>
>> No. The protorol is changed too. I wrote about it 10/03/08.
>> (Subject: [Apcupsd-users] Problem with new version of Smart-UPS RT
>> 5000)
>
> Ah, that sucks...
>
>> I use SNMP Driver now.
>
> I use SNMP for a few of our boxes too, but I'd still like the
> possibility for normal/native control/monitoring as I currently do
> for all the other UPS' we have.  I'm guessing that it's being worked
> on in CVS?  I'm in no rush, but thought I'd mention it anyway, just
> incase no-one else had come across it.

The new protocol is not actively being worked on for a simple reason: I
don't have any specs for it, nor do I have an UPS to test it on (or reverse
engineer it from). New protocols are easy to add to apcupsd, so adding
support should be straight forward once someone documents it. I will
investigate some more to see if anyone has the protocol spec yet.

--Adam

Hallo,

you must configure DEVICE 10.1.1.166:apc:admin user phrase
you must delete your passphrase and set the old one Administrator -->  
user -->apc
device --> admin user --> password: phrase

you can not change there everything


Zitat von Jeff Thurston <jthurston@...>:

> First of all, this software is great! I am sure this problem is due to user
> error and not apcupsd.
>
> So thank you in advance for everything.
>
>
>
> Platform: Debian Etch, AMD64 & Debian Etch x86 (same problem for both)
>
> UPS model: "Smart-UPS RT 6000 RM XL"
>
> UPS Network Management Card: "AP9619" WEB/SNMP/PowerChuteNET.
>
> apcupsd version: 3.14.3
>
>
>
> Configured with PCNET enabled, and everything works beautifully, until I try
> to use PCNET with auth.
>
> If I do not configure DEVICE ip:user:passphrase, it works, except for
> shutdown of course gets permission denied or somesuch. But all of the tests
> in the manual, aside from actual live shutdown will work.
>
>
>
> I have specified my own PowerChute passphrase as well as using the default.
>
> Also the APC manual/help documentation say the username is Administrator,
> tried that as well as admin.
>
>
>
> No matter what I do, I always get this error after about 30 seconds if I
> have the "DEVICE" line configured.
>
>
>
> apccontrol: Warning serial port communications with UPS lost.
>
> But I am not using serial, so what gives? Or is that simply a last resort.
>
>
>
> My Config:
>
> ## apcupsd.conf v1.1 ##
>
> UPSCABLE ether
>
> UPSTYPE pcnet
>
> LOCKFILE /var/lock
>
> DEVICE 10.1.1.166:Administrator:admin user phrase
>
> UPSCLASS standalone
>
> UPSMODE disable
>
> TIMEOUT 60
>
>
>
> My Config.out:
>
> Configuration on Thu Apr 17 13:55:19 PDT 2008:
>
>
>
>   Host:                       x86_64-unknown-linux-gnu -- debian 4.0
>
>   Apcupsd version:            3.14.3 (20 January 2008)
>
>   Source code location:       .
>
>   Install binaries:           /sbin
>
>   Install config files:       /etc/apcupsd
>
>   Install man files:          /usr/share/man
>
>   Nologin file in:            /etc
>
>   PID directory:              /var/run
>
>   LOG dir (events, status)    /var/log
>
>   LOCK dir (for serial port)  /var/lock
>
>   Power Fail dir              /etc/apcupsd
>
>   Compiler:                   g++ 4.1.2
>
>   Compiler flags:             -g -O2 -Wall
>
>   Linker:                     gcc
>
>   Linker flags:               -g -O
>
>   Host and version:           debian 4.0
>
>   Shutdown Program:           /sbin/shutdown
>
>   Port/Device:                /dev/ttyS0
>
>   Network Info Port (CGI):    6543
>
>   UPSTYPE                     apcsmart
>
>   UPSCABLE                    smart
>
>
>
>   drivers (no-* are disabled): apcsmart dumb net no-usb no-snmp pcnet
> no-test
>
>
>
>   enable-nis:                 yes
>
>   with-nisip:                 0.0.0.0
>
>   enable-cgi:                 no
>
>   with-cgi-bin:               /etc/apcupsd
>
>   with-libwrap:
>
>   enable-nls:                 no
>
>   enable-libintl:             no
>
>   enable-powerflute:          yes
>
>   enable-pthreads:            yes
>
>   enable-dist-install:        yes
>
>   enable-gapcmon:             no
>
>
>
>



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

Jeff Thurston wrote:
> Configured with PCNET enabled, and everything works beautifully,
> until I try to use PCNET with auth.
>
> If I do not configure DEVICE ip:user:passphrase, it works, except for
> shutdown of course gets permission denied or somesuch. But all of
> the tests in the manual, aside from actual live shutdown will work.
>
> I have specified my own PowerChute passphrase as well as using the
> default.
>
> Also the APC manual/help documentation say the username is
> Administrator, tried that as well as admin.

Be careful not to confuse the PCNET username/password with the console
username/password that you use to log in to the Web/SNMP card. They are
distinct. The PCNET password must be set from the PCNET configuration
section of the card. If you don't see a PCNET section in the Web GUI,
you're probably running old firmware. Try upgrading your Web/SNMP card
firmware.

> No matter what I do, I always get this error after about 30 seconds
> if I have the "DEVICE" line configured.
>
> apccontrol: Warning serial port communications with UPS lost.

That is a sign of the authentication failing.

> But I am not using serial, so what gives? Or is that simply a last
> resort.

That's just a poorly-worded message. It means Apcupsd cannot communicate
with your UPS, whatever the connect type is. Back in the day the only
connections we supported were over serial links, so the text hardcoded in
the apccontrol scripts always said "serial". I'm guessing you were running
the "safe.apccontrol" script? I fixed all the other scripts a year or more
ago (just removed the reference to serial), but a grep of the tree shows I
missed the safe.apccontrol script. I've now fixed that one, too.

--Adam