Re: my isp has trouble with its own dhcp assignment

From: spellberg_robert <emailrob_at_risebroadband.net>
Date: Thu, 20 Oct 2022 22:17:37 UTC
On 10/18/22 12:47 AM, David Christensen wrote:
> On 10/17/22 12:41, spellberg_robert wrote:
>> 22_oct_17_mon 19.35.utc
>>
>>
>>
>> howdy , folks ; long time , no post .
>> why ?
>> because , when compared to 20 - 40 years ago , when i --loved-- it ,
>>    email has become the most_royal of pains to use .
>>
>> please forgive this brief digression ; it is relevant .
>>
>> i spent 6 hours , yesterday ,
>>    typing a detailed account of the symptoms and
>>    the eventual identification of that
>>    which i believe to be the actual problem
>>    with my isp's dhcp server .
>>
>> when sending ,
>>    i got the password wrong [ correct chars , wrong order ] .
>> believing that i was closing the dialog_box for the password ,
>>    i closed the composition window .
>> after wrestling w/ thunderbird for an hour w/o finding the message ,
>>    i resorted to the tried_and_true bsd approach ;
>>    with hd and vi , i found , finally , a copy in a file called "Sent" .
>> because the email was never sent ,
>>    this is the exact place in which to save it .
>> i believe that the jargon for this characteristic is "feature" .
>>
>> ok ; so , i did not lose the 6 hours of typing [ whew ! ] .
>>
>> i copied the relevant lines from "Sent" into its own file .
>> --but-- , for the life of me , i can not get thunderbird to
>>    bring this file into a new composition window ,
>>    except as an "attachment" [ which i can not see ] .
>>
>> iirc , email_"attachment"s are strongly discouraged on this list .
>> so , it is --not-- included w/ this post .
>> how_ever , i could be wrong .
>> if it --is-- permissible to do so ,
>>    then i will be happy to send it w/ a subsequent post .
>>
>>
>>
>> here is the abridged version of my problem .
>>
>> i have built backup_hardware for my gateway box .
>>
>> in the event of a hardware failure ,
>>    a]  power_down failed_box ,
>>    b]  detach cables ,
>>    c]  swap boxen ,
>>    d]  attach cables ,
>>    e]  power_up good_box w/ auto_boot w/ dhclient .
>>
>> during the testing of the new box ,
>>    i have found that the isp's dhcp server returns an address
>>    which works only partially or not_at_all
>>    [ it is a different address , always ; should it be ? ] .
>>
>> because the existing box has not failed ,
>>    ifconfig tells me its real identity .
>>
>> if i use , manually , the "add" form of the ifconfig command ,
>>    with this address , then ,
>>    suddenly , everything works perfectly .
>>
>> --but-- , i did this manually , not by using dhcp .
>> i must assume that i do not know the prior address .
>>
>> the handbook and various man_pages discuss , at length ,
>>    the case of a roving lap_top , w/ multiple isps .
>> in this case , i have multiple clients and a constant isp ;
>>    this is not discussed .
>>
>>
>>
>> is there a way for my client to persuade the isp's server to
>>    "bind" my client to an address which the isp will accept ?
>>
>> what is the etiquette regarding
>>    email_"attachment"s in postings to this list ?
>>
>> is there a way to get a text_file
>>    [ perhaps , created w/ vi ]
>>    into a thunderbird composition window ?
>>
>>
>>
>> tia
>>
>> rob
> 

22_oct_20_thu 22.10.utc



salut , david ---

a priori : thank you , thank you , thank you .

[ 20_thu 19.20.ish

   in proof_reading this , before sending it off ,
     i have found that the amount of information is extensive .
   there_fore , i revise the above to :

     a priori : merci infiniment .

   i appreciate , very_much , your willingness and your time .
]



another etiquette question :

   when i reply to the list , should i "cc" to the poster , also ?

erring on the side of caution , i have done so , this time ,
   as the deletion of an extra copy is easy to accomplish .



> 
> Regarding the IP address(es) assigned to your Internet service point by 
> your ISP, it depends upon your ISP, which Internet service plan you 
> have, and which hardware the ISP provided to you.
> 

this makes sense .

> 
> Regarding attachments to mailing list posts, my approach is as follows:
> 
> 1.  If the attachment is text and small, include the content in the post.
> 
> 2.  Otherwise, upload the content to a web server and include the URL in 
> the post.
> 

this is useful to know ; how_ever , ...

> 
> Regarding getting text file contents into an e-mail message using a 
> graphical workstation, one method is to open the file in Mousepad, 
> Notepad, etc., select the desired text, copy the selected text to the 
> clipboard, switch to the Thunderbird message composition window, and 
> paste the clipboard contents.
> 

   ... the above generated an idea .

i do not have the named "pad"s ;
   but , i tried the following and it worked .

open the text_file in a firefox "new_tab" .
select from the ul_corner to the lr_corner .
<ctrl_c>
position the cursor in the thunderbird composition_window .
<ctrl_v>

[ because this procedure worked ,
     i have rejected the "attachment" approach ;
     the original post has been replicated at the end of this post ,
     in its entirety , verbatim , including some headers ,
     between two horizontal rules of seven hyphens , each .
   much of the requested information is there , already .
]

this has the advantage of
   permitting an original creation , of a post , to be done in vi .
un_fortunately , this flavor of tb erases drop_down_lists immediately ,
   thus , i can not "save" the file any_where ; but , i can "send" it .
keying <ctrl_s> has the same behavior .

[ hmmm ... , per_haps , i can extract it from "mbox"
   [ i use , also , fetchmail ] .
]

> 
> If the above responses are insufficient, please answer the following 
> questions:
> 
> 1.  What is the location of your Internet service point?
> 

geographically , we are in rural mc_henry county , illinois , usa ,
   roughly mid_way between woodstock and harvard .
if you meant some_thing else , then please advise .

> 2.  Who is your ISP?  Please provide a URL.
> 

https://www.risebroadband.com

> 3.  What Internet service plan do you have?  Please provide a URL.
> 

https://www.risebroadband,com/residential

> 4.  What is the make and model of the access device provided by the ISP? 
>   Please provide a URL.
> 

if you mean the roof_top antenna , then i do not know .
how_ever , i can attempt to learn by making a phone_call .
it is my very serious hope that this is --not-- what you seek ;
   customer_service is --not-- this isp's strong_point
   [ --some-- clerks --are-- very_good ] .
possibly , this is because i have berkeley_unix boxen , only .

on the other hand ,
   if you mean the "thingy" , which the isp calls a "router" ,
   then we do not have one .
the gateway box is directly connected to the roof_top antenna ,
   with rj45 and poe .
the gateway does two things , packet_filtering and nat .
the isp never sees any internal traffic .

> 5.  You seem to be discussing several computers -- a gateway computer, a 
> backup gateway computer, and several clients (presumably, one is a 
> graphical workstation with Thunderbird).  Please clarify how many 
> computers you have and their roles.
> 

[ please refer to the post , below . ]

in general , there are many machines , on 192.168.100.0/24 ,
   but , for now , only three are involved with the out_side world .
these are :

   natasha	192.168.100.200		i386	 8.1
   catherine	192.168.100.143		amd64	11.4
   113		192.168.100.113		amd64	11.3

   natalie	192.168.100.201		amd64	11.3	as user
   natasha	192.168.100.200		amd64	11.3	as gateway

> 6.  Please pick one computer and let us work on that one.  When that is 
> fixed, pick another computer and start a new thread.
> 

agreed .

assuming that the 11.3_natasha config is successful , then_
   the plan is to replicate the fix onto 12.3_natasha ;
   this "should work" [ of course , i could be wrong ] .
with the existing 8.1_natasha , i will have two gateways .
this accomplishes partial redundancy .
now that the next 3 release schedules have been announced ,
   i will wait until 23_jan .
the remaining two i3_8100 boxen will be configured on 12.4 ;
   the 12.3 will be re_done on 12.4 and the 8.1 will be retired .

well , ... , it's a plan .

> 7.  What is the make and model of the computer?  Which CPU?  How much 
> memory?  What storage devices?  Please provide URL(s).
> 

i build my own mid_tower boxen [ see below ] .

the user boxen are i3/5/7 "skylake" , 64_GB ,
   1/2/4_TB seagate "constellation_es.3" .
next_year , some_time , "skylake" will be replaced with "rocket_lake"
   [ even though the "skylake" works just_fine ,
       i want to try_out the new "avx_512" instructions
   ] .

the new gateways are 16_GB , same seagate .

the 8.1 has , may_be , 256_MB and
   western_digital "cavier_blue" or "_black" .

> 8.  Which FreeBSD installer did you use?  What questions were you asked 
> and how did you answer them?
> 

the one which is included
   with the applicable amd64/dvd or i386/cd image_file .

uhhh , ... , all of them and honestly
   [ to the best of my under_standing ] .

broadly_speaking , i install every_thing , including the ports_tree .
hard_drives are extensively partitioned , to make full back_ups faster .

[ actually , 113 was not partitioned ,
     because of the confusingly worded question , in the installer .
   i wanted to use "fbsd only" --and--
     i wanted to "partition" [ --but-- , not for other "os"es ] .
   i guessed the wrong answer .
   because of this ,
     i chose to not finish the config any farther than necessary ;
     this includes not naming it "francoise" .
   it --has-- proved to be a useful second firefox platform ,
     because its ff_67 displays "pdf"_files ,
     while the ff_88 on 11.4_catherine , mostly , does_not do so .
]

i want to be helpful , here ;
   if the above is in_sufficient , then please advise .

> 9.  Please run the following commands and post your complete console 
> session -- prompt, commands entered, output printed:
> 
> # freebsd-version ; uname -a
> 

well ; you asked for it ; but tb forces wrap_around at 72 chars
   [ this --must-- be a "feature" ] ;
   i will indicate this situation w/ "[tb72]" .

note_that : all boxen have a non_privileged user in group "wheel" .

catherine
---------

--0--   pts/8   root @ catherine.ntrprnr.com.   2022_Oct_19_Wed[tb72]
    13:05:23.utc
/tmp

3888 # freebsd-version ; uname -a
11.4-RELEASE
FreeBSD catherine.ntrprnr.com 11.4-RELEASE FreeBSD 11.4-RELEASE #0[tb72]
  r362094: Fri Jun 12 18:27:15 UTC 2020    root@releng2.nyi.freebsd[tb72]
.org:/usr/obj/usr/src/sys/GENERIC  amd64

113
---

--0--   pts/2   root @ 113.ntrprnr.com.   2022_Oct_19_Wed[tb72]
    17:12:24.utc
/

310 # freebsd-version ; uname -a
11.3-RELEASE
FreeBSD 113.ntrprnr.com 11.3-RELEASE FreeBSD 11.3-RELEASE #0[tb72]
  r349754: Fri Jul  5 04:45:24 UTC 2019    root@releng2.nyi.freebsd[tb72]
.org:/usr/obj/usr/src/sys/GENERIC  amd64

natasha 8.1
-----------

--0--   pts/1   root@natasha.ntrprnr.com.   2022_Oct_19_Wed[tb72]
    17:32:30.gmt
/+shadow/etc/d.resolv_dot_conf

306 # freebsd-version ; uname -a
freebsd-version: Command not found.
FreeBSD natasha.ntrprnr.com 8.1-RELEASE FreeBSD 8.1-RELEASE #0[tb72]
: Mon Jul 19 02:55:53 UTC 2010    root@almeida.cse.buffalo.edu[tb72]
:/usr/obj/usr/src/sys/GENERIC  i386

natalie 11.3
------------

--0--   ttyv0   root @ natalie.ntrprnr.com.   2022_Oct_19_Wed[tb72]
    17:55:30.utc
/root

310 # freebsd-version ; uname -a
11.3-RELEASE
FreeBSD natalie.ntrprnr.com 11.3-RELEASE FreeBSD 11.3-RELEASE #0[tb72]
  r349754: Fri Jul  5 04:45:24 UTC 2019    root@releng2.nyi.freebsd[tb72]
.org:/usr/obj/usr/src/sys/GENERIC  amd64

> # cat /etc/rc.conf
> 
> 

in this case , i know what you meant ;
   but , i do not do this in the same manner as do many .

i start by copying /etc/defaults/rc.conf to a new file .
then , i add a comment marker to the beginning of every line .
last , i un_comment those lines which are desired to be operative .

additionally , i rationalize the formatting ,
   to eliminate the 80_column constraint [ i use 320 ] .
just like /etc/defaults/rc.conf , these files have hundreds of lines .

in this way , when editing ,
   i do not have to switch back_and_forth between two_or_more files .

/etc/rc.conf is , it_self , a sym_link .

thus , those which follow are the default over_rides , --only-- ;
   they appear in /etc/defaults/rc.conf order .

natalie 11.3
------------

note_that :
   the dummy_config of em0 eliminates some extra boot messages ;
   when booted as natalie , no cable is attached , here .

powerd_enable="YES"			# common
hostname="natalie.ntrprnr.com"
ifconfig_em0="inet 192.168.100.202/24"	# this is a dummy config .
ipv4_addrs_re0="192.168.100.201/24"
local_unbound_enable="YES"
sshd_enable="YES"			# common
ftpd_enable="YES"			# common
default_router="192.168.100.200"
keymap="us"				# common
keyrate="fast"				# common
cursor="blink"				# common
font8x16="iso-8x16.fnt"			# common
font8x14="iso-8x14.fnt"			# common
font8x8="iso-8x8.fnt"			# common
blanktime="NO"				# common
moused_enable="YES"			# common
dumpdev="AUTO"				# common

natasha 11.3
------------

powerd_enable="YES"			# common
hostname="natasha.ntrprnr.com"
ipfilter_enable="YES"
ipnat_enable="YES"
ipmon_enable="YES"
ifconfig_em0="DHCP"
ipv4_addrs_re0="192.168.100.200/24"
sshd_enable="YES"			# common
ftpd_enable="YES"			# common
gateway_enable="YES"
keymap="us"				# common
keyrate="fast"				# common
cursor="blink"				# common
font8x16="iso-8x16.fnt"			# common
font8x14="iso-8x14.fnt"			# common
font8x8="iso-8x8.fnt"			# common
blanktime="NO"				# common
moused_enable="YES"			# common
dumpdev="AUTO"				# common

natasha 8.1
-----------

hostname="natasha.ntrprnr.com."
ipfilter_enable="YES"
ipnat_enable="YES"
ipmon_enable="YES"
ifconfig_em0="DHCP"
ipv4_addrs_re0="192.168.100.200/24"
inetd_enable="YES"
named_enable="YES"
sshd_enable="YES"
ftpd_enable="YES"
gateway_enable="YES"
icmp_bmcastecho="YES"
keymap="us.iso"
keyrate="fast"
cursor="blink"
font8x16="iso-8x16"
font8x14="iso-8x14"
font8x8="iso-8x8"
blanktime="NO"
moused_enable="YES"
lpd_enable="YES"
linux_enable="YES"

catherine 11.4
--------------

powerd_enable="YES"
hostname="catherine.ntrprnr.com"
ifconfig_re0="inet 192.168.100.143 netmask 255.255.255.0"
local_unbound_enable="YES"
sshd_enable="YES"
ftpd_enable="YES"
default_router="192.168.100.200"
moused_enable="YES"
lpd_enable="YES"
dumpdev="AUTO"
linux_enable="YES"

113 11.3
--------

hostname="113.ntrprnr.com"
ifconfig_re0="inet 192.168.100.113 netmask 255.255.255.0"
inetd_enable="YES"
local_unbound_enable="YES"
sshd_enable="YES"
ftpd_enable="YES"
default_router="192.168.100.200"
icmp_bmcastecho="YES"
keymap="us"
keyrate="fast"
cursor="blink"
font8x16="iso-8x16.fnt"
font8x14="iso-8x14.fnt"
font8x8="iso-8x8.fnt"
blanktime="NO"
moused_enable="YES"
lpd_enable="YES"
dumpdev="AUTO"
linux_enable="YES"

> 10. If the computer is graphical, which window manager or desktop 
> environment are you using?
> 

user/client machines have x installed ; the wm is "twm" .

gateways do not have x
   [ well , strictly_speaking , one does ;
       but , it was never configured ;
       there_fore , it is never started
   ] .

> 11. Are you using ports or packages?  What ports and/or packages have 
> you installed?
> 
> 

packages .

11.4 catherine
--------------

xorg			7.7_3		used extensively
firefox			88.0.1_1,2	used extensively
thunderbird		78.10.2		used moderately		no_flush
fetchmail		6.4.18		used after thunderbird	w/ flush

nasm			2.15.05,1	dedicated_purpose

apache-openoffice	4.1.10		not_used
libreoffice6		6.4.7_9		not_used
xpdf4			4.03,1		not_used
xpdf			4.02,1		not_used
xpdf3			3.04_11		not_used

tcl86			8.6.11_1	great syntax , sloppy arithmetic
tcllib			1.20
tk86			8.6.11_2	not_used

dvd+rw-tools		7.1_3		used occasionally

11.3 113
--------

xorg			7.7_3		used extensively
firefox			67.0.2_2,1	used extensively
thunderbird		60.7.1		not_used

8.1 natasha
-----------

none

11.3 natasha/natalie
--------------------

xorg			7.7_3		not_used ; never_configured
firefox			67.0.2_2,1	not_used
thunderbird		60.7.1		not_used
m4			1.4.18_1,1	not_used

12.3 natasha/natalie
--------------------

bind			9.16.31		use tbd

> David
> 

-------
 From - Mon Oct 17 01:35:41 2022
X-Mozilla-Status: 0001
X-Mozilla-Status2: 00800000
X-Mozilla-Keys: 

BCC: emailrob@emailrob.com
Reply-To: emailrob@emailrob.com
To: freebsd-questions@freebsd.org
From: spellberg_robert <emailrob@risebroadband.net>
Subject: new gateway box runs slowly or partially - dhcp issue ?
Message-ID: <073d90ea-efe9-f571-f265-d41569f09f1d@risebroadband.net>
Date: Mon, 17 Oct 2022 01:33:43 +0000
User-Agent: Mozilla/5.0 (X11; FreeBSD amd64; rv:78.0) Gecko/20100101
  Thunderbird/78.10.2
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit

22_oct_16_sun 25.26.utc



howdy , folks ---

long time , no post .
here is hoping that all of you_and_yours are doing well ;
    we are doing ok .



my existing gateway has well_over a decade of 24/7 use ;
    fbsd i386 8.1 on micro_atx celeron_440 [ 65 nm , "cedar mill" ] .

my objective is to build two new boxen , so that ,
    when one fails , a known_good replacement can be swapped_in .
these would be rotated every 4 or 5 weeks .
i already have multiple "user" machines ,
    so this is [ nearly ] the last redundancy to be implemented
    [ i need a backup isp , also ] .
as we all learn , sooner_or_later , hardware fails .

i started this 2_1/2 years ago , so the parts for 4 boxen are
    based upon the i3_8100 , h310 chip_set .
at that time , the release was 11.3 ,
    so that is one of the configurations which i used .
later , i re_started the config w/ 12.3 ;
    how_ever , the recent success occurred w/ the 11.3 version ;
    this will be described below .
except for the differences between the 11.3 and 12.3 releases ,
    the two configs are identical
    [ of course , i could be wrong , but , i do not believe so ] .
thus , a "fix" for 11.3 "should work" on 12.3 .



the gory details :

     8.1  natasha  re0 192.168.100.200/24  <-->  em0 dhcp         gateway

    11.3  natasha  re0 192.168.100.200/24  <-->  em0 dhcp         gateway
    11.3  natalie  re0 192.168.100.201/24  <-->  em0 dummy_config user

all use ipfilter/ipnat/ipmon ;
    the ".rules" files are "ftp"_copies , from the 8.1 box .

natalie/natasha 11.3 has two rc.conf files :
    all development is done on natalie ; then ,
    change one sym_link , power_down , swap cables , boot as natasha ;
    same name , same 1918_address , make dhcp request .
when done testing ,
    change the sym_link , power_down , swap cables , boot as natalie .
the development process is fairly simple and
    this procedure mimics the expected future routine operation .

note_that : when the 11.3 box is natasha ,
    then natasha 8.1 remains booted ,
    but , its network cables [ .200 , dhcp ] are plugged into 11.3 ;
    also , the natalie cable [ .201 ] is detached and un_connected .

the current 8.1 dhcp assignment is 66.114.191.36/25
    [ iirc , i have seen , also , .39 ] ;
    when 11.3 is booted as natasha , the dhcp "bound to" address is ,
    recently , 172.19.65.233/27 [ also , .239 , .240 , .241 ] .
these come from [ DHCPOFFER , DHCPACK ] 66.114.187.249 [ always ] .



some observations :

when natalie and the other user_machines are going through 8.1 ,
    which is using the 66. address ,
    then , every_thing is "just_fine , thank_you" , as expected .
ping and traceroute , to 8.8.8.8 , establish a base_line of "normalcy"
    [ 7 - 15 ms , 13 hops [ 9 belong to my isp ] ] .
interestingly , on the 8.1_box , it_self ,
    ping and traceroute generate "Network is unreachable" messages .
also , on 12.3 [ but , not 11.3 ] ,
    natalie successfully retrieved the package for bind_9.16.31 ,
    from where_ever [ thus , resolution works ] .

--how_ever-- , when the user_machines are going through the 11.3 box ,
    which is using the 172. address ,
    then , for 8.8.8.8 , ping never returns and
    traceroute goes permanently "* * *" after , maybe , 6 hops
    [ iirc , i let it go for 30 - 40 ; much more than 13 ] .
ping and traceroute succeed w/ 66.114.187.249 , above ;
    this is hop 2 , after the gateway , 192.168.100.200 .
thus , they --do-- work ; they just do not go very far .
on the 11.3 box , it_self , ping and traceroute generate
    the same "Network is unreachable" messages .



the discovery :

although i had started this in 20_feb ,
    i have been rather obsessed since 22_apr .
spending at least part of almost every week_end on this ,
    because i was making no progress ,
    every_thing became suspect .
are my config_files correct ?
am i spelling commands correctly ?
did some_body change some code ,
    but not note the change in the man_page ?
i choose to believe that i can solve almost any fbsd problem
    [ at least , for the things that --i-- do ] ;
    but , after a_while , a person begins to doubt themselves .



22_sep_25_sun :

i had noticed that the two dhcp assignments were different .
i was reading the man_page for ifconfig .
i had seen the concept of "alias" and i had seen the option of "add" .
on the 8.1 box , netstat -inW reported the network as "66.114.191.0/"
    [ on 8.1 , the -W option has no effect ] .
so , on the previous sun [ sep_18 ] , i had tried

    ifconfig em0 inet 66.114.191.36/24 add

    but , it did not work .
how_ever , i noticed that the network specification was in_complete .
i had assumed that it was a /24 .
a check with ifconfig confirmed that my assumption was wrong
    [ that'll teach me ] .
try again :

    ifconfig em0 inet 66.114.191.36/25 add

MIRACLE_OF_MIRACLES !!!

it was like the parting of the red sea .

every_thing was wide_open .
ping and traceroute worked as well as they did on the 8.1 box .
firefox had no trouble w/ "radar.weather.gov" ,
    nor with "forecast.weather.gov" .

after watching this , with smug satisfaction , for several minutes ,
    i deleted the alias , because it had not been obtained by dhcp .
i have no way of knowing how long any address assignment will last .



finally , the question [ and the crowd goes wild ] :

un_fortunately , this discovery does not solve the problem ;
    it succeeds , only , in identifying it .

it seems to me that the isp should be assigning the existing address,
    rather than assign a new address and , then ,
    not_or_partially honor that new address .
after all , they already know the identity of the existing address .
why change it .

in reading various man_pages and the hand_book ,
    i find a bias in the examples which are given .
the common premise is that of a single [ constant ] lap_top ,
    attached to a wandering human ,
    which human appears at different locations ,
    for the purpose of connecting to different isps ,
    through the use of something , which is called "wi_fi" .
my situation is the inverse of this ;
    i have one [ constant ] isp and multiple clients .
this situation is not discussed .
if the client_machine is the constant ,
   then , probably , there exists a "data_base" file ,
   which file lists prior "leases" .
suppose that the removed_box is being replaced because
    its hd will_not "spin_up" .
there needs to be a way
    to get the isp's dhcp_server to assign an address ,
    which address will be accepted by the isp .

i suspect that i may have to talk to my isp about this ;
    but , then , maybe_not .
maybe , this is some_thing which can be solved at my end , completely
    [ that would be nice ] .
it may be that i need to spend money to get a "static" ip_assignment .
how_ever , if i am going to do that , then i may want to get a /29 ;
    i want my own mail_server and that means 2 authoritative name_servers
    [ an early version of bind_9 is on the 8.1 box ;
        i find bind to be straight_forward
    ] .
regard_less , before i inquire of the isp ,
    i wanted to learn whether_or_not any_one who reads this list has
    any experience with a problem of this type .



this is the premise :

    back_up__hardware is A Good Thing ;

this is the question :

    how do i get the back_up__hardware
    to protocol with the isp's dhcp_server
    to assign an address [ same or different , from recent use ]
    which will be accepted by the isp ?



tia

rob



ps ---

sorry about the length , but , on this list ,
    it is my opinion that
    it is better to err on the side of thorough_ness .
-------