Re: [Nut-upsuser] driver for HP/Compaq T750 ?

2009-02-26 Thread Arnaud Quette
2009/2/26 Charles Lepple clep...@gmail.com

 On Wed, Feb 25, 2009 at 1:54 PM, KP Kirchdoerfer
 kap...@bering-uclibc.de wrote:
  Hi,
 
  I own a UPS HP T750.
 
  Is there a driver for this UPS (either serial or USB)?

 As you may have already seen, we don't have an entry for this UPS in
 data/drivers.list yet.

 However, HP often repackages other vendors' equipment. Does the
 documentation mention what software is used to monitor the UPS on
 Windows? Often this is a clue as to what is inside.


knowing the OS you're running would also help...
for example, on Linux for USB, you could run lsusb to help determine the
rebrand originator and so the potential driver needed. My guess is that it's
either Powerware (so bcmxcp_* or usbhid-ups) or Apc (so apcsmart or
usbhid-ups)

Arnaud
-- 
Linux / Unix Expert RD - Eaton - http://www.eaton.com/mgeops
Network UPS Tools (NUT) Project Leader - http://www.networkupstools.org/
Debian Developer - http://people.debian.org/~aquette/
Free Software Developer - http://arnaud.quette.free.fr/
___
Nut-upsuser mailing list
Nut-upsuser@lists.alioth.debian.org
http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser

Re: [Nut-upsuser] early shutdown of VMware VMs

2009-02-26 Thread Arnaud Quette
Hi David,

just a forward to the list since others might be interested in that info.

2009/2/25 David Newman dnew...@networktest.com

 On 2/17/09 5:26 AM, Arnaud Quette wrote:
 
  I've appended a script excerpt I've once made to address the guest
  shutdown from the host.
  it was for ESX, and at that time, I was really surprised to not see
  anything to address automagically an ordered shutdown (of the VMs)...
  hope this helps.

 Thanks for this -- it is helpful.

 For VMWare Server 2 (and probably future versions of ESX) there is a new
 VIX API that replaces vmware-cmd.

 VIX has a bunch of stuff I haven't tried yet, but I do know hosts start
 and stop using the vmrun command. The syntax for starting a guest host
 is like this:

 vmrun -T server -h https://myvmhost:8333/sdk -u root -p password start
 [standard] vmware/FreeBSD.vmx

 dn


 
  Arnaud
  --
  Linux / Unix Expert RD - Eaton - http://www.eaton.com/mgeops
  Network UPS Tools (NUT) Project Leader - http://www.networkupstools.org/
  Debian Developer - 
  http://people.debian.org/~aquette/http://people.debian.org/%7Eaquette/
  Free Software Developer - http://arnaud.quette.free.fr/
 
  2009/2/14 Doug Parsons d...@parsonsemail.com
  mailto:d...@parsonsemail.com
 
  Thanks, Doug. I'm newly on VMWare Server 2, which has different
 tools
  than the 1.x ones I'm familiar with. AFAIK the 1.x tools did not
 have
  any sort of UPS or shutdown awareness capability; I'll need to check
  whether that's true with 2.x.
  
  dn
 
 
  Unfortunately Server is the only version that I don't have running
  anymore.
  But as Arjen noted most versions can handle the guest shutdown
  natively. One
  catch is that this extends the shutdown and startup times quite a
  bit and so
  the battery level at which you start the shutdown would need to be
  higher.
  Another catch I believe is that the guest must be running the VMWare
  tools.
  If VMWare Server can't do it (check under the advanced tab) then you
 may
  want to consider VMWare 3i as it can and is also free. Hit the
  VMWare forums
  as they should be able to fill in all the details.
 
  Doug
 
 
 
  # Customisable Vmware ESX shutdown
  stopVMWare()
  {
# Test if we have a Vmware ESX v3 setup
if [ -x /usr/bin/vmware ]
then
  ESXV3=`/usr/bin/vmware -v | grep ESX Server 3`
  if [ -n $ESXV3 ]
  then
  # Get the VM list
VMLIST=`/usr/bin/vmware-cmd -l`
 
  for VM in $VMLIST
  do
# Get the VM state
VMSTATE=`/usr/bin/vmware-cmd $VM getstate -q`
 
  # Guest OS shutdown if VMSTATE is equal to on
  if [ $VMSTATE == on ]
  then
/usr/bin/vmware-cmd $VM stop trysoft
 
# delay a bit the next sequence
sleep 2
  fi
done
# Delay for 1 minute to give the VMs more time
# to cleanly shutdown
sleep 60
  fi
fi
  }


cheers,
Arnaud
-- 
Linux / Unix Expert RD - Eaton - http://www.eaton.com/mgeops
Network UPS Tools (NUT) Project Leader - http://www.networkupstools.org/
Debian Developer - http://people.debian.org/~aquette/
Free Software Developer - http://arnaud.quette.free.fr/
___
Nut-upsuser mailing list
Nut-upsuser@lists.alioth.debian.org
http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser

[Nut-upsuser] RFC: Use tcp-wrapper for all connections to upsd

2009-02-26 Thread Joerg Pulz
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1


Hi,

after some experimenting and digging through the code i found no solution 
how to completely disable access to upsd from specific hosts.
In previous versions (before r1233) it was possible to allow or deny 
access to upsd completely by using ACL, ACCEPT and REJECT entries in 
upsd.conf. As this functionality was removed and tcp-wrappers support was 
introduced i thought it would be possible to use some rules in hosts.allow 
to get the same functionality as before. Unfortunately, thats not the 
case.
Only authenticated commands like SET or INSTCMD are protected by 
tcp-wrappers, all other commands like GET or LIST can be used from 
everywhere by everyone which is IMO a regression.

For me, the right solution would be to protect all incoming connections by 
tcp-wrappers.

What do others think about this?

Kind regards
Joerg

- -- 
The beginning is the most important part of the work.
-Plato
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.9 (FreeBSD)

iD8DBQFJpnrkSPOsGF+KA+MRAl5qAJ4giiMOPNrSjAnI3p7Fa0NHSLCSbQCgjJBr
kTpLEuSEJJAKdLutFMZxmFE=
=DBDA
-END PGP SIGNATURE-

___
Nut-upsuser mailing list
Nut-upsuser@lists.alioth.debian.org
http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser


Re: [Nut-upsuser] driver for HP/Compaq T750 ?

2009-02-26 Thread KP Kirchdoerfer
Am Donnerstag, 26. Februar 2009 12:01:41 schrieb Arnaud Quette:
 2009/2/26 Charles Lepple clep...@gmail.com

  On Wed, Feb 25, 2009 at 1:54 PM, KP Kirchdoerfer
 
  kap...@bering-uclibc.de wrote:
   Hi,
  
   I own a UPS HP T750.
  
   Is there a driver for this UPS (either serial or USB)?
 
  As you may have already seen, we don't have an entry for this UPS in
  data/drivers.list yet.
 
  However, HP often repackages other vendors' equipment. Does the
  documentation mention what software is used to monitor the UPS on
  Windows? Often this is a clue as to what is inside.

AFAIK native HP software.

 knowing the OS you're running would also help...

yes, sorry; I'm running Debian Lenny.

 for example, on Linux for USB, you could run lsusb to help determine the

lsusb just shows Hewlett Packard...

Bus 001 Device 002: ID 03f0:1f06 Hewlett-Packard

 rebrand originator and so the potential driver needed. My guess is that
 it's either Powerware (so bcmxcp_* or usbhid-ups) or Apc (so apcsmart or
 usbhid-ups)

Unfortunately none of these worked; the errors are varying:

with apcsmart
Unable to detect an APC Smart protocol UPS on port /dev/ttyS1
Check the cabling, port name or model name and try again
similar message with the bcmxcp driver;

the upscode2 driver gives:
Missing UPCL after UPCL
No contact with UPS, delaying init.

And the upshid-ups driver ends with:
No matching HID UPS found
Driver failed to start (exit status=1)

regards
kp








___
Nut-upsuser mailing list
Nut-upsuser@lists.alioth.debian.org
http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser


Re: [Nut-upsuser] driver for HP/Compaq T750 ?

2009-02-26 Thread Arjen de Korte
Citeren KP Kirchdoerfer kap...@bering-uclibc.de:

[...]

 And the upshid-ups driver ends with:
 No matching HID UPS found
 Driver failed to start (exit status=1)

Try this driver again, but make sure to add

 productid = 1f06

to the ups.conf entry for this device. Chances are that it is a  
re-branded Tripplite unit (we've seen other HP units that are actually  
this brand).

Best regards, Arjen
-- 
Please keep list traffic on the list


___
Nut-upsuser mailing list
Nut-upsuser@lists.alioth.debian.org
http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser


Re: [Nut-upsuser] RFC: Use tcp-wrapper for all connections to upsd

2009-02-26 Thread Arjen de Korte
Citeren Joerg Pulz joerg.p...@frm2.tum.de:

 after some experimenting and digging through the code i found no solution
 how to completely disable access to upsd from specific hosts.

On multi-homed servers the LISTEN directive will deal with this, by  
only listening on interfaces from which clients are allowed to  
connect. If this isn't fine grained enough, your firewall will keep  
out unwanted connections much more efficiently than tcp-wrappers (or  
the now obsolete ACL mechanism) ever will.

 In previous versions (before r1233) it was possible to allow or deny
 access to upsd completely by using ACL, ACCEPT and REJECT entries in
 upsd.conf. As this functionality was removed and tcp-wrappers support was
 introduced i thought it would be possible to use some rules in hosts.allow
 to get the same functionality as before. Unfortunately, thats not the
 case.

This is by design.

 Only authenticated commands like SET or INSTCMD are protected by
 tcp-wrappers, all other commands like GET or LIST can be used from
 everywhere by everyone which is IMO a regression.

 For me, the right solution would be to protect all incoming connections by
 tcp-wrappers.

Using tcp-wrappers for source address access control alone is a *huge*  
waste of effort, therefor NUT no longer supports this.

 What do others think about this?

The tcp-wrappers support in NUT is only meant to deal with the case  
where you want to allow access for certain users from a specific set  
of machines (for instance, administrative access). This means we  
require the username and password, hence this only works for commands  
that require to be logged into the server.

The previous ACL mechanism was too inefficient (in terms of resources)  
to be really useful in countering attacks on the server. By the time  
the decision to allow or deny a client access was made, most of the  
effort that was needed to process the incoming connection would  
already have been spent, so there really wasn't that much to gain  
anymore (other than restrict clients to see what is going on on the  
server). This is the reason we dropped the ACL mechanism.

Best regards, Arjen
-- 
Please keep list traffic on the list


___
Nut-upsuser mailing list
Nut-upsuser@lists.alioth.debian.org
http://lists.alioth.debian.org/mailman/listinfo/nut-upsuser