Webramp vs. Sonicwall Firewall?
yu...@nospam.lansol.com
Sonicwall SOHO
Webramp 700S
Assuming the same number of user license.
** remove "NOSPAM" for e-mail replies.
Tolunay Orkun
I think Webramp was OEM of Sonicwall (different case and branding)
yu...@nospam.lansol.com
Jon Marshall
dealt with WebRamp's support in the past with their modem sharing devices,
and they're responsive and quick. With credit card, they'll overnight you a
replacement unit and then you return yours back to them. I have no dealings
with Sonicwall so I can't comment.
Also, you might want to look at the 3Com unit (3C16770) it also appears to
be a Sonicwall base clone and 3Com offers a lifetime warranty.
"Tolunay Orkun" <tolunay@-NOSPAM-usa.net> wrote in message
news:53bnaskcmv2eqld4e...@4ax.com...
> On Thu, 17 Feb 2000 06:36:52 GMT, yu...@NOSPAM.lansol.com wrote:
>
Jerry Asher
No hardware problems. Timely firmware updates with fixes and new features.
One concern I've always had with any oemmed products is how I get new
firmware and other drivers: do I get it from the original manufacturer and
if so, will I be treated equally to one of their customers? Do I have to
get it from the "branded" manufacturer (Webramp, 3COM?), and if so, does
that indicate there will be long delays in getting it to me, or worse,
incompatibilities? I have had different experiences/answers to this problem
with different items I have bought over the years. I am not saying this is
a reason to favor Sonicwall, I am only saying this has been a concern of
mine over the years.
Jerry
"Jon Marshall" <marshal...@augustmail.com> wrote in message
news:eQ9Td2Ve$GA.201@cpmsnbbsa03...
li...@noreply.com
to is who has the better support and better price?
James Schmidt
> So they are the same units with different names? What it comes down
> to is who has the better support and better price?
>
Sonicwall upgrades (supposidly) only work if your product is
registered. You submit your serial number and they issue a
registration number that you plug into the Sonicwall. Poof
it's registered.
Question arises if you can legally and ethically register
an OEM product with Sonicwall to use their firmware upgrades.
I would think you would register with your OEM and obtain
the upgrades from them. So now you are dependent on two
vendors (Sonicwall and the OEM) for support.
James
--
My return email address might be bogus.
It's really: schmidtj AT enteract DOT com
Mike Shepard
> firmware and other drivers: do I get it from the original manufacturer and
A big drawback of buying and registering Ramp Networks products is
you get spammed for life. Also, they charge you do download upgrades
to the jellyware in their NVM chips. And, they hide this - they
send you email that the upgrade is available, and download it now.
You download the huge file, then during install, it tells you need
to pay them before proceeding. This is based on my experience
with a WebRamp. I do use it, it works, but the company itself
is annoying. Sonicwall has better support and upgrade policies.
Will Kramer
"no don't send me email"? and the other gentleman never had this box
as an option or didn't check it off....
Will
John Thomas <jth...@quarnet.com> wrote:
>Well I think they have improved on both counts, but I don't under-
>stand why they are spamming you, but not me. They have my info for both
>upgrades, and I have received a grand total of 1 e-mail from them in the
>last 4 months- 1 copy of the Ramp news. Go figure...
>
>Mike Shepard wrote:
>>
>> > Is this for the 700s? I have recently done flash updates on 2 of them,
>> > and in neither case did I have to pay.
>>
>> Well to be fair, my complaint about them spamming me is very true -
>> but their policy of making you pay for every software update may
>> no longer be true. I bought a Webramp a couple of years ago, and
>> the first time they invited me to upgrade it's jellyware - is when
>> I got the "you must pay us" after I downloaded it and started the
>> update program. Maybe they fixed their upgrade policies - but
>> to this day, they still send me spam - even though I've asked
>> a dozen times for them to stop.
gz
news:Mjcr4.23106$0p1.5...@news4.giganews.com...
> The WebRamp also includes the Hub check their online docs out
> http://www.rampnet.com/support/700s/documentation/rel_not.html
> for what it's worth. It is the SAME unit, case, board, software, firmware,
> features, etc...
Beware that the hub has been added since it was first marketed. I purchased
my 700s almost a year ago and it does *not* have a hub - I didn't care
anyway as I already had an 8-port.
If you want one with the hub, check before ordering, especially over the web
>
> "bt" <@gingerroot.org> wrote in message
> news:JK6r4.604$PA2....@bgtnsc06-news.ops.worldnet.att.net...
> >
> > John Thomas <jth...@quarnet.com> wrote in message
> > news:38ACD24E...@quarnet.com...
> > <snip>
> > > The interface is identical.
> >
> > uhhh, has *ANYONE* mentioned the webramp comes with NO built-in hub
unlike
> > the sonicwall??
> >
> > here's a picture:
> > http://www.webramp.com/products/700s/
> > *notice the LAN port and the WAN port and NO OTHER ports?*
> >
> > here's the specs:
> > http://www.webramp.com/products/700s/tech.html
> >
> > lesson:
> > don't let anyone in a newsgroup do all the homework for you, check out
the
> > stuff yourself.
> >
> > i was impressed with the webramp wealth of support pages vs. sonicwall's
> > sparse marketing effort. i *really* liked the GUI tryout
> > (http://www.webramp.com/products/700s/700sDemo/index.html) but when it
> > dawned on me that i would have to connect a hub or small switch i
balked.
> i
> > like the all-in-one form factor of the sonicwall, to me it's worth the
> ~$50
> > price difference. funny that's just a little more than what a cheap
> 4-port
> > hub costs.
> >
> > bt
> >
> >
>
>
>
Pete Koolish
and not 10. I'm still using my Netgear hub downstream from the sonicwall
because of this.
Pete
Bob O`Brien
How about a swtich, instead of a mere hub?
<http://www.linksys.com/scripts/features.asp?part=befsr41>
Bob O`Bob
--
+ email replies without this line will be discarded cgpipe5p (expires 20mar00) +
No spammer could ever be too broke or too hungry.
John Navas
ob...@best.com (Bob O`Brien) wrote:
>In article <Nfwt4.13617$Ev4.3...@nnrp1-w.snfc21.pbi.net>,
>Pete Koolish <pkoo...@pdfgroup.com> wrote:
>>Speaking of hubs, I sure wish that the hub in the sonicwall was a 10/100 hub
>>and not 10. I'm still using my Netgear hub downstream from the sonicwall
>>because of this.
>
>How about a swtich, instead of a mere hub?
>
> <http://www.linksys.com/scripts/features.asp?part=befsr41>
Of little value with only 4 ports. ;-)
--
Best regards,
John Navas <http://navasgrp.home.att.net/>
CABLE/DSL TIPS: <http://navasgrp.home.att.net/tech/cable_dsl.htm>
Christopher Painter
However an inexpensive switch would get you 10/100.
The problem is you end up having 3 devices. One to convert ADSL/SDSL/CABLE
to Ethernet. One to act as firewall / NAT, and another to act as a switch /
hub. Starts to get expensive for the SOHO environment.
Chris
"Pete Koolish" <pkoo...@pdfgroup.com> wrote in message
news:Nfwt4.13617$Ev4.3...@nnrp1-w.snfc21.pbi.net...
> Speaking of hubs, I sure wish that the hub in the sonicwall was a 10/100
hub
> and not 10. I'm still using my Netgear hub downstream from the sonicwall
> because of this.
>
> Pete
>
>
Philip J. Koenig
(Christopher Painter) writes...
> The sonic wall has a wan and a lan port. They are both 10 and not 100.
> However an inexpensive switch would get you 10/100.
> The problem is you end up having 3 devices. One to convert ADSL/SDSL/CABLE
> to Ethernet. One to act as firewall / NAT, and another to act as a switch /
> hub. Starts to get expensive for the SOHO environment.
>
> Chris
Oh hardly. First of all, as John mentioned, a switch for a
4-port LAN is generally a waste of time. Secondly, the speeds
your internet connection are running at will never even tax a
10baseT port, so you certainly don't need it for "better internet
speed". Then you have the fact that the firewall vendor gets
to get intertwined in things like Nway negotiation compatibility
with whatever gippo NICS people decide to run in their "SOHO
LAN", so they may find themselves A) answering support questions
about things which have little to do with what their product is
about or B) customers getting disappointed in their product
simply because some aspect of the hub/switch part doesn't
work with their environment.
It's not surprising firewall vendors aren't so eager to throw
every little network piece in there. Single-speed hubs are
simple, generic and pretty foolproof. Switches are not.
> "Pete Koolish" <pkoo...@pdfgroup.com> wrote in message
> news:Nfwt4.13617$Ev4.3...@nnrp1-w.snfc21.pbi.net...
> > Speaking of hubs, I sure wish that the hub in the sonicwall was a 10/100
> hub
> > and not 10. I'm still using my Netgear hub downstream from the sonicwall
> > because of this.
> >
> > Pete
--
Philip J. Koenig The Electric Kahuna Organization [see below]
-----------------Computers & Communications for the New Millenium-------------
* This message is spam-resistant. To send email, use the following address *
* by removing numbers and spaces: pjkunet64 @ ekahuna27 . c o m *
* *
* Expose ORBS for what it is: abuse of the net. *
* Simple answers are for simple minds. Try a new way of looking at things. *
John Navas
For_...@ddress.see_be.low (Philip J. Koenig) wrote:
>In article <898pq6$74s$1...@bob.news.rcn.net>, cpai...@not-it.dalenet.com
>(Christopher Painter) writes...
>> The sonic wall has a wan and a lan port. They are both 10 and not 100.
>> However an inexpensive switch would get you 10/100.
>> The problem is you end up having 3 devices. One to convert ADSL/SDSL/CABLE
>> to Ethernet. One to act as firewall / NAT, and another to act as a switch /
>> hub. Starts to get expensive for the SOHO environment.
>Oh hardly. First of all, as John mentioned, a switch for a
>4-port LAN is generally a waste of time. Secondly, the speeds
>your internet connection are running at will never even tax a
>10baseT port, so you certainly don't need it for "better internet
>speed". Then you have the fact that the firewall vendor gets
>to get intertwined in things like Nway negotiation compatibility
>with whatever gippo NICS people decide to run in their "SOHO
>LAN", so they may find themselves A) answering support questions
>about things which have little to do with what their product is
>about or B) customers getting disappointed in their product
>simply because some aspect of the hub/switch part doesn't
>work with their environment.
>
>It's not surprising firewall vendors aren't so eager to throw
>every little network piece in there. Single-speed hubs are
>simple, generic and pretty foolproof. Switches are not.
Nevertheless, I do think that SonicWALL SOHO should include dual-speed
10/100 LAN ports, rather than just 10baseT LAN ports. For little more
than the cost of a 10baseT hub, a dual-speed hub provides most of the
benefits of a switch at the 4-port level, allowing LAN traffic at 100 Mbps
while the Internet connection runs at 10 Mbps.
Phil
<spamf...@navasgrp.dublin.ca.us> wrote:
>[POSTED TO ba.internet]
>For_...@ddress.see_be.low (Philip J. Koenig) wrote:
>
>>In article <898pq6$74s$1...@bob.news.rcn.net>, cpai...@not-it.dalenet.com
>>(Christopher Painter) writes...
>>> The sonic wall has a wan and a lan port. They are both 10 and not 100.
[[snip ]]
>Nevertheless, I do think that SonicWALL SOHO should include dual-speed
>10/100 LAN ports, rather than just 10baseT LAN ports. For little more
>than the cost of a 10baseT hub, a dual-speed hub provides most of the
>benefits of a switch at the 4-port level, allowing LAN traffic at 100 Mbps
>while the Internet connection runs at 10 Mbps.
I hafta think that Sonicsystem's roadmap includes a dual-speed hub or
switches. Only, it might happen after a hub/switch/router
manufacturer acquires them. How do you spell C-I-S-C-O?
If Sonic doesn't do it, one of their competitors will.
Phil
Phil Burton - Palo Alto,CA USA
Real address is: philb at alink dot net
Thanks for helping to reduce spam.
John Navas
ph...@alink.net (Phil) wrote:
>On Mon, 28 Feb 2000 07:07:10 -0800, John Navas
><spamf...@navasgrp.dublin.ca.us> wrote:
>>Nevertheless, I do think that SonicWALL SOHO should include dual-speed
>>10/100 LAN ports, rather than just 10baseT LAN ports. For little more
>>than the cost of a 10baseT hub, a dual-speed hub provides most of the
>>benefits of a switch at the 4-port level, allowing LAN traffic at 100 Mbps
>>while the Internet connection runs at 10 Mbps.
>
>I hafta think that Sonicsystem's roadmap includes a dual-speed hub or
>switches. Only, it might happen after a hub/switch/router
>manufacturer acquires them. How do you spell C-I-S-C-O?
3-C-o-m (as in
<http://www.sonicwall.com/Corporate_Info/press_release_archive/3Com_pr.html>)
>If Sonic doesn't do it, one of their competitors will.
Linksys EtherFast Cable/DSL Router
<http://www.linksys.com/scripts/features.asp?part=befsr41>
UMAX UGate-3000
<http://www.umax.com/networking/standard/ugate/product.cfm?model=UGate%2D3000>
Christopher Painter
I couldn't disagree with you more. The price difference between a hub
and a switch these days is very little. You have no idea the kind of
traffic that I can generate in my house between my various clients and
servers. Also its very nice to be able to plug into a port and autosense
100. Your making the assumption that the majority of the traffic on my lan
is destined for the internet, you couldn't be more wrong. I never claimed
the switch gives me better wan performance, it gives me better LAN
performance. But now that you mention it, it could give me better wan
performance. Lets say I'm using SAMBA to copy a 600MB ISO from my Linux
box to my Windows box where I have the burner. I'm chunking up 80-90%
utilization and got the collision light blinking like a banche. Now my wife
is on her IMac trying to surf the net. What do you think is going to do
better, a 10mb segment with 1 ethernet collision domain, or a 100mb switch
with 4 seperate ethernet collision domains? I also completly disagree with
you saying a single speed ethernet hub is simple to support and that a
switch is not. Have you ever used real networking hardware? I setup a
network for a small business once using a NetGear 16Port 10/100 switch.
Each port can autosense/autoswitch to 10/100 and full duplex/half duplex.
The switch is unmanaged and can hold over 8,000 mac addresses. What could
possible be hard to support about that? The god damn switch just sits
there, you don't even think about it. The whole thing is plugged into a
Netopia router running a T1. I've never heard one complaint about
performance or compatibility.
Chris
"Philip J. Koenig" <For_...@ddress.see_be.low> wrote in message
news:MPG.1323f7c4b...@news.brainstorm.net...
> In article <898pq6$74s$1...@bob.news.rcn.net>, cpai...@not-it.dalenet.com
> (Christopher Painter) writes...
> > The sonic wall has a wan and a lan port. They are both 10 and not 100.
> > However an inexpensive switch would get you 10/100.
> > The problem is you end up having 3 devices. One to convert
ADSL/SDSL/CABLE
> > to Ethernet. One to act as firewall / NAT, and another to act as a
switch /
> > hub. Starts to get expensive for the SOHO environment.
> >
> > Chris
>
>
> Oh hardly. First of all, as John mentioned, a switch for a
> 4-port LAN is generally a waste of time. Secondly, the speeds
> your internet connection are running at will never even tax a
> 10baseT port, so you certainly don't need it for "better internet
> speed". Then you have the fact that the firewall vendor gets
> to get intertwined in things like Nway negotiation compatibility
> with whatever gippo NICS people decide to run in their "SOHO
> LAN", so they may find themselves A) answering support questions
> about things which have little to do with what their product is
> about or B) customers getting disappointed in their product
> simply because some aspect of the hub/switch part doesn't
> work with their environment.
>
> It's not surprising firewall vendors aren't so eager to throw
> every little network piece in there. Single-speed hubs are
> simple, generic and pretty foolproof. Switches are not.
>
>
>
>
John Navas
PMJI
"Christopher Painter" <cpai...@not-it.dalenet.com> wrote:
>"Philip J. Koenig" <For_...@ddress.see_be.low> wrote in message
>news:MPG.1323f7c4b...@news.brainstorm.net...
>> ... as John mentioned, a switch for a
>> 4-port LAN is generally a waste of time. ...
> I couldn't disagree with you more. The price difference between a hub
>and a switch these days is very little.
Buy.com:
Netgear EN104TP 4-port 10baseT hub $23.52
Netgear DS104 4-port 10/100 hub $59.95
Netgear FS105 5-port 10/100 switch $81.95
(The FS104 4-port switch has apparently been discontinued.)
I'd call that a pretty big difference.
>You have no idea the kind of
>traffic that I can generate in my house between my various clients and
>servers.
With only 4 ports, you have at most:
DSL modem
Computer A
Computer B
Computer C
A can be talking to B at nearly full speed when C in talking to the
Internet, since C is limited to no more than 1.5 Mbps (usually less) out
of a total of 10 Mbps.
>Also its very nice to be able to plug into a port and autosense
>100.
True, but that doesn't require a switch, only a dual-speed hub.
Nevertheless, even a dual-speed hub is much more expensive than a 10baseT
hub.
>Your making the assumption that the majority of the traffic on my lan
>is destined for the internet, you couldn't be more wrong. I never claimed
>the switch gives me better wan performance, it gives me better LAN
>performance.
Not with only 4 ports. A dual-speed hub will give essentially the same
performance (since it's effectively 10 hub + 100 hub + smart bridge).
>But now that you mention it, it could give me better wan
>performance. Lets say I'm using SAMBA to copy a 600MB ISO from my Linux
>box to my Windows box where I have the burner. I'm chunking up 80-90%
>utilization and got the collision light blinking like a banche.
Not unless the NIC's are badly broken.
>Now my wife
>is on her IMac trying to surf the net. What do you think is going to do
>better, a 10mb segment with 1 ethernet collision domain, or a 100mb switch
>with 4 seperate ethernet collision domains?
Based on experience, I think the difference for your wife would be very
small.
>I also completly disagree with
>you saying a single speed ethernet hub is simple to support and that a
>switch is not. Have you ever used real networking hardware? I setup a
>network for a small business once using a NetGear 16Port 10/100 switch.
>Each port can autosense/autoswitch to 10/100 and full duplex/half duplex.
>The switch is unmanaged and can hold over 8,000 mac addresses. What could
>possible be hard to support about that? The god damn switch just sits
>there, you don't even think about it. The whole thing is plugged into a
>Netopia router running a T1. I've never heard one complaint about
>performance or compatibility.
There can be compatibility problems between certain switches and certain
NIC's (for that matter even dual-speed hubs and certain NIC's), an
important consideration for a vendor that needs to minimize support costs.
While I personally would prefer dual-speed 10/100 ports, I can understand
why SonicWALL went with just 10baseT (much less a switch).
John Navas
ph...@alink.net (Phil) wrote:
>Guess for now, Sonic/Ramp/3COM is the only game in town, and it's only
>a 10Mbit/sec game for now.
Which is of course more than enough for the Internet. Just add the
Netgear DS104 10/100 hub for $60 and you're in business.
Phil
<spamf...@navasgrp.dublin.ca.us> wrote:
When I saw this, I got all excited 'cuz the Linksys price is hundreds
below Sonic Systems. Then I read the user manaal PDF. It's not a
real firewall, although the router feartures are very attractive.
Guess for now, Sonic/Ramp/3COM is the only game in town, and it's only
a 10Mbit/sec game for now.
Phil
>[POSTED TO ba.internet]
>ph...@alink.net (Phil) wrote:
>
>>On Mon, 28 Feb 2000 07:07:10 -0800, John Navas
>><spamf...@navasgrp.dublin.ca.us> wrote:
>
>>>Nevertheless, I do think that SonicWALL SOHO should include dual-speed
>>>10/100 LAN ports, rather than just 10baseT LAN ports. For little more
>>>than the cost of a 10baseT hub, a dual-speed hub provides most of the
>>>benefits of a switch at the 4-port level, allowing LAN traffic at 100 Mbps
>>>while the Internet connection runs at 10 Mbps.
>>
>>I hafta think that Sonicsystem's roadmap includes a dual-speed hub or
>>switches. Only, it might happen after a hub/switch/router
>>manufacturer acquires them. How do you spell C-I-S-C-O?
>
>3-C-o-m (as in
><http://www.sonicwall.com/Corporate_Info/press_release_archive/3Com_pr.html>)
>
>>If Sonic doesn't do it, one of their competitors will.
>
>Linksys EtherFast Cable/DSL Router
><http://www.linksys.com/scripts/features.asp?part=befsr41>
>UMAX UGate-3000
><http://www.umax.com/networking/standard/ugate/product.cfm?model=UGate%2D3000>
Phil Burton - Palo Alto,CA USA
Christopher Painter
> Netgear EN104TP 4-port 10baseT hub $23.52
> Netgear DS104 4-port 10/100 hub $59.95
> Netgear FS105 5-port 10/100 switch $81.95
> (The FS104 4-port switch has apparently been discontinued.)
> I'd call that a pretty big difference.
I wouldn't call that a big difference. Consider what all of this stuff used
to cost, and how long you will use it. If you compare 23.50 to 60 thats 2.5x
the price for 10x the available bandwidth. By shopping for a good price on
some generic RealTek 10/100 network cards, one could afford the delta.
> With only 4 ports, you have at most:
> DSL modem
> Computer A
> Computer B
> Computer C
Sorry I miss spoke.... Its a 5 port FS105.... and one of the ports is
currently unused.
> A can be talking to B at nearly full speed when C in talking to the
> Internet, since C is limited to no more than 1.5 Mbps (usually less) out
> of a total of 10 Mbps.
My SDSL is 768Kbps. But I don't want A and B talking at 10 or even 20. The
I/O on linux server is an Abit BP6 ( UDMA66) running a stripeset of IBM
Deskstar 7200RPM 2MB Cache. I can *EASILY* exceed 10mbps. 200mbps is more
approriate. Even then 20MB/sec can be exceeded by my I/O. I guess I will
have to wait for gigabit ethernet over copper. I ran some premium cable so
hopefully I'll just have to replace my interconnects.
> >Also its very nice to be able to plug into a port and autosense
> >100.
>
> True, but that doesn't require a switch, only a dual-speed hub.
> Nevertheless, even a dual-speed hub is much more expensive than a 10baseT
> hub.
Personally I think dual-speed hubs are a bullshit marketing hack. Have you
seen how DLINK has been advertising theirs? They call it a "Dual Speed Hub
with Integrated Switch". I understand it, but I think someone going to
bestbuys looking for a Switch could get screwed. But as you say, why not
spend a little bit more.
> >But now that you mention it, it could give me better wan
> >performance. Lets say I'm using SAMBA to copy a 600MB ISO from my Linux
> >box to my Windows box where I have the burner. I'm chunking up 80-90%
> >utilization and got the collision light blinking like a banche.
>
> Not unless the NIC's are badly broken.
Must be the damn IMac 8) I can take the switch out, plug everything into
the router and show you......
> >I also completly disagree with
> >you saying a single speed ethernet hub is simple to support and that a
> >switch is not. Have you ever used real networking hardware? I setup a
> >network for a small business once using a NetGear 16Port 10/100 switch.
> >Each port can autosense/autoswitch to 10/100 and full duplex/half duplex.
> >The switch is unmanaged and can hold over 8,000 mac addresses. What
could
> >possible be hard to support about that? The god damn switch just sits
> >there, you don't even think about it. The whole thing is plugged into a
> >Netopia router running a T1. I've never heard one complaint about
> >performance or compatibility.
Didn't have much to complain here? I'll give you another advantage to a
switch in the above situation...
Security: If I used a hub someone could plug a laptop in the conference
room and start sniffing the segment.
If I use a switch they have to plug into the computer room where the servers
and routers are to try to get traffic destined to
that port ( this room is physically secured ) or they would have to put a
hub in place on the host they are trying to sniff
(which could not be done covertly)
> There can be compatibility problems between certain switches and certain
> NIC's (for that matter even dual-speed hubs and certain NIC's), an
> important consideration for a vendor that needs to minimize support costs.
>
> While I personally would prefer dual-speed 10/100 ports, I can understand
> why SonicWALL went with just 10baseT (much less a switch).
I just disagree. Bell Atlantic DSL claims that only 3Com network cars
3c509 for ISA and 3c900 for PCI and 3c589d for PCMCIA) work with their DSL
offering. This is just bullshit, and I bet they are making a good buck
selling 3Com cards. Now they are going integrated so they don't even have
ethernet. Ethernet is based on standards, and as long as I buy good stuff,
it should interoperate. Atleast Covad will support a lan.
Philip J. Koenig
> >For_...@ddress.see_be.low (Philip J. Koenig) wrote:
> >
> >>(Christopher Painter) writes...
> >>> The sonic wall has a wan and a lan port. They are both 10 and not 100.
>
> >Nevertheless, I do think that SonicWALL SOHO should include dual-speed
> >10/100 LAN ports, rather than just 10baseT LAN ports. For little more
> >than the cost of a 10baseT hub, a dual-speed hub provides most of the
> >benefits of a switch at the 4-port level, allowing LAN traffic at 100 Mbps
> >while the Internet connection runs at 10 Mbps.
>
> I hafta think that Sonicsystem's roadmap includes a dual-speed hub or
> switches. Only, it might happen after a hub/switch/router
> manufacturer acquires them. How do you spell C-I-S-C-O?
>
If anyone takes over Sonicwall, it's going to be 3com,
IMHO.. considering the fact that 3com already OEMs their
product and puts their name on it.
BTW, Sonicwall has no competition in the low-end space.
That won't last forever of course.
Philip J. Koenig
(Christopher Painter) writes...
> > There can be compatibility problems between certain switches and certain
> > NIC's (for that matter even dual-speed hubs and certain NIC's), an
> > important consideration for a vendor that needs to minimize support costs.
> >
> > While I personally would prefer dual-speed 10/100 ports, I can understand
> > why SonicWALL went with just 10baseT (much less a switch).
>
>
> I just disagree. Bell Atlantic DSL claims that only 3Com network cars
> 3c509 for ISA and 3c900 for PCI and 3c589d for PCMCIA) work with their DSL
> offering. This is just bullshit, and I bet they are making a good buck
> selling 3Com cards. Now they are going integrated so they don't even have
> ethernet. Ethernet is based on standards, and as long as I buy good stuff,
> it should interoperate. Atleast Covad will support a lan.
Read my lips: standards or not, it is a well-established fact
that "auto-negotiation" between various 10/100 switches or
multi-speed hubs DOESN'T WORK with certain combinations of
hardware.
Most low-end 10-100 parts don't have a way to manually
force a port into 10mbit, 100mbit, half or full-duplex.
Sooner or later, you will understand why this is important.
Bill Pitz
> If anyone takes over Sonicwall, it's going to be 3com,
> IMHO.. considering the fact that 3com already OEMs their
> product and puts their name on it.
> BTW, Sonicwall has no competition in the low-end space.
> That won't last forever of course.
I agree about 3Com. They're starting to edge into a lot of markets like
that.
There are some little companies starting up offering similar products
at prices that are roughly the same. Occasionally I see ads for them in
Boardwatch and the like.
-Bill
Bill Pitz
> Read my lips: standards or not, it is a well-established fact
> that "auto-negotiation" between various 10/100 switches or
> multi-speed hubs DOESN'T WORK with certain combinations of
> hardware.
> Most low-end 10-100 parts don't have a way to manually
> force a port into 10mbit, 100mbit, half or full-duplex.
> Sooner or later, you will understand why this is important.
The good news is that better 3Com switches will let you force the port on
the switch into specific modes. This is what I did to get some old SMC
10/100 boards to work at 100mbit fdx. (Which, btw are some of the best
cards I've used. Especially considering when they were made) They
scream now that I've got them up at 100mbit fdx.
-Bill
Christopher Painter
has dip switches on the back of it for duplex mode, and I picked it up for
under $400. Is that low end enough for you? Besides we were talking about
SOHO environment. Its not like I'm going to have a rats nest of equipment
tied to each other to troubleshoot.
Besides, where do you come off acting like you are the only person in the
world with any networking experience and that only your opinion matters?
Chris
"Philip J. Koenig" <For_...@ddress.see_be.low> wrote in message
news:MPG.13286a4a2...@news.brainstorm.net...
> In article <89gems$gs2$1...@bob.news.rcn.net>, cpai...@not-it.dalenet.com
> (Christopher Painter) writes...
>
> > > There can be compatibility problems between certain switches and
certain
> > > NIC's (for that matter even dual-speed hubs and certain NIC's), an
> > > important consideration for a vendor that needs to minimize support
costs.
> > >
> > > While I personally would prefer dual-speed 10/100 ports, I can
understand
> > > why SonicWALL went with just 10baseT (much less a switch).
> >
> >
> > I just disagree. Bell Atlantic DSL claims that only 3Com network cars
> > 3c509 for ISA and 3c900 for PCI and 3c589d for PCMCIA) work with their
DSL
> > offering. This is just bullshit, and I bet they are making a good buck
> > selling 3Com cards. Now they are going integrated so they don't even
have
> > ethernet. Ethernet is based on standards, and as long as I buy good
stuff,
> > it should interoperate. Atleast Covad will support a lan.
>
>
> Read my lips: standards or not, it is a well-established fact
> that "auto-negotiation" between various 10/100 switches or
> multi-speed hubs DOESN'T WORK with certain combinations of
> hardware.
>
> Most low-end 10-100 parts don't have a way to manually
> force a port into 10mbit, 100mbit, half or full-duplex.
> Sooner or later, you will understand why this is important.
>
>
>
Bill Pitz
> The NetGear FS-516 (rackmount 16port 10/100 autosensing ) that I use at work
> has dip switches on the back of it for duplex mode, and I picked it up for
> under $400. Is that low end enough for you? Besides we were talking about
> SOHO environment. Its not like I'm going to have a rats nest of equipment
> tied to each other to troubleshoot.
> Besides, where do you come off acting like you are the only person in the
> world with any networking experience and that only your opinion matters?
I've got a couple of those. As well as some 8-port ones, both rack and
non rack.
They're not really switches. They support full/half duplex, but when
running a sniffer, the individual nodes see *all* of the traffic from the
other nodes. This is not the behavior of a real switch (like a nice 3Com)
Those dip switches only select Full/Half duplex. You can't choose between
10mbit and 100mbit with those dipswitches, whereas the management software
in my 3Com switch lets me pick any of combination of these. And I can do
it remotely, without having to fidget with dip switches.
-Bill
John R Pierce
>They're not really switches. They support full/half duplex, but when
>running a sniffer, the individual nodes see *all* of the traffic from the
>other nodes. This is not the behavior of a real switch (like a nice 3Com)
the Netgear FS108 switch I have at work certainly doesn't send non-broadcast
traffic down the ports. Its a true store-and-forward packet switch. Frankly,
I'm not quite sure how you could do full duplex and NOT be a true switch?
-jrp
Christopher Painter
true. This behavior would be indicative of a hub not a switch, and I doubt
its that badly broken or designed.
I used this for a small business ( 90+ employees nationwide, around 12 hosts
here in a townhouse ) with a T1 on a Netopia router. I allocated every
active drop with its own drop straight on the switch, those who can ( most
hosts are machines with 3Com 3c905TX ) go 200MB. Now because its 1 host per
1 port on the switch its pretty easy to look at the activity lights on the
front of the switch and see the switching that is going on. A packet
sniffer will confirm or deny what you are reporting, but if what you said
was true,you could probably be looking at a class action lawsuit against
NetGear.
PS- Maybe we should move this thread to comp.dcom.lans.ethernet, we really
aren't talking about DSL anymore.
Chris
"Bill Pitz" <bi...@nospammm.svn.net> wrote in message
news:38bf...@news1.svn.net...
> They're not really switches. They support full/half duplex, but when
> running a sniffer, the individual nodes see *all* of the traffic from the
> other nodes. This is not the behavior of a real switch (like a nice 3Com)
>
> -Bill
Bill Pitz
> true. This behavior would be indicative of a hub not a switch, and I doubt
> its that badly broken or designed.
How could I make that up? In my office I have 6 Windows PCs with an uplink
to a 3Com Switch which provides connectivity to the LAN + the net.
I attached my laptop and ran NAI's "Sniffer Basic" and was able to see *all*
of the traffic from the Windows PCs. Try it. Maybe by some strange event I
ended up with two of them that behave the same way.
-Bill
Christopher Painter
Watch how you quote that Bill, I did not post the following below. I did put
a sniffer on a port of the NetGear FS516 today and I did not see any
crossover traffic.
Chris
"Bill Pitz" <bi...@nospammm.svn.net> wrote in message
news:38bf...@news1.svn.net...
Bill Pitz
> Bill,
> Watch how you quote that Bill, I did not post the following below. I did put
> a sniffer on a port of the NetGear FS516 today and I did not see any
> crossover traffic.
Perhaps it is "fixed" in the FS5xx series. I haven't tried it on those. I
did try it on an FS308 and an FS316 (same switch, just not rack mountable)
and I saw the traffic.
-Bill
Philip J. Koenig
(Christopher Painter) writes...
> The NetGear FS-516 (rackmount 16port 10/100 autosensing ) that I use at work
> has dip switches on the back of it for duplex mode, and I picked it up for
> under $400. Is that low end enough for you? Besides we were talking about
> SOHO environment. Its not like I'm going to have a rats nest of equipment
> tied to each other to troubleshoot.
The discussion revolved around the inclusion of a switch
(rather than hub) into a firewall product which sells for
under $500 all-inclusive. Hardly a candidate to include
$400 worth of switch hardware.
Low-end switches are less than $100 these days, and it
would seem fairly clear that these are the class of
devices (with no manual port-configuration facility) that
would likely be included in a firewall which has a retail
price of <$500, if at all.
Then of course you'll have all the people kvetching that
the firewall only has a 4-port hub or switch, but they
have 5 workstations.
I'd think people would rather have the flexibility to buy
whatever hub or switch they want, not have to pay for
something built into the firewall that may not meet their
needs, and then get whatever does suit their needs. (ie
a $99 5-port switch if they so desire)
If people think that two small boxes are "too complicated",
and those people are trying to put together a network,
their problems have only just begun.
> Besides, where do you come off acting like you are the only person in the
> world with any networking experience and that only your opinion matters?
>
> Chris
Only in the face of belligerent people who attempt to
authoritatively spew misinformation.
Philip J. Koenig
> > has dip switches on the back of it for duplex mode, and I picked it up for
> > under $400. Is that low end enough for you? Besides we were talking about
> > SOHO environment. Its not like I'm going to have a rats nest of equipment
> > tied to each other to troubleshoot.
>
> > world with any networking experience and that only your opinion matters?
>
> non rack.
>
> running a sniffer, the individual nodes see *all* of the traffic from the
> other nodes. This is not the behavior of a real switch (like a nice 3Com)
Oh my, that even surprises me. Sounds like something either
isn't working right, or they are engaging in deceptive advertising.
The only time you should see a switch leak traffic to other
nodes is A) broadcasts or B) if some switches are highly
congested (ie no more buffers), they will start bridging
traffic.
Philip J. Koenig
Most of what I have seen are vastly simpler than the SonicWall,
generally just dumb packet filters or simple proxies.
If you know of something that sells for less than $500, uses
stateful inspection and is dedicated (non-PC) hardware, I'd
be curious to get the company name and/or URL.
Clark Martin
See_email_@ddress_below.This_one_is.invalid (Philip J. Koenig) wrote:
>The only time you should see a switch leak traffic to other
>nodes is A) broadcasts or B) if some switches are highly
>congested (ie no more buffers), they will start bridging
>traffic.
C) when one machine first sends a packet to another machine (which is a
MAC broadcast even if it isn't a broadcast for the higher level
protocol).
D) When in promiscuous mode for debugging.
--
Clark Martin
Redwood City, CA, USA Macintosh / Internet Consulting
"I'm a designated driver on the Information Super Highway"
Philip J. Koenig
(Clark Martin) writes...
> In article <MPG.132b581b3...@news.brainstorm.net>,
> See_email_@ddress_below.This_one_is.invalid (Philip J. Koenig) wrote:
>
> >The only time you should see a switch leak traffic to other
> >nodes is A) broadcasts or B) if some switches are highly
> >congested (ie no more buffers), they will start bridging
> >traffic.
>
> C) when one machine first sends a packet to another machine (which is a
> MAC broadcast even if it isn't a broadcast for the higher level
> protocol).
>
> D) When in promiscuous mode for debugging.
True. Technically item C is called an "ARP Request". It's
not traffic per se, but it could be construed as a minor
security issue. (a sniffer would be able to ascertain what
MAC and/or IP addresses reside on the segment. Actually
dunno about MAC addresses, the responding host might unicast
it's reply back to an ARP request, have to look that up.)
John R Pierce
>In article <clarkm-0503...@wildcatl.martin.home>, cla...@pacbell.net
>(Clark Martin) writes...
>> In article <MPG.132b581b3...@news.brainstorm.net>,
>> See_email_@ddress_below.This_one_is.invalid (Philip J. Koenig) wrote:
>>
>> >The only time you should see a switch leak traffic to other
>> >nodes is A) broadcasts or B) if some switches are highly
>> >congested (ie no more buffers), they will start bridging
>> >traffic.
>>
>> C) when one machine first sends a packet to another machine (which is a
>> MAC broadcast even if it isn't a broadcast for the higher level
>> protocol).
>>
>> D) When in promiscuous mode for debugging.
>
>
>True. Technically item C is called an "ARP Request". It's
>not traffic per se, but it could be construed as a minor
>security issue. (a sniffer would be able to ascertain what
>MAC and/or IP addresses reside on the segment. Actually
>dunno about MAC addresses, the responding host might unicast
>it's reply back to an ARP request, have to look that up.)
it does.
only the *initial* ARP request is broadcast. Rest of the traffic is MAC to MAC,
including ARP renewals...
-jrp
Bill Pitz
> Oh my, that even surprises me. Sounds like something either
> isn't working right, or they are engaging in deceptive advertising.
> The only time you should see a switch leak traffic to other
> nodes is A) broadcasts or B) if some switches are highly
> congested (ie no more buffers), they will start bridging
> traffic.
It surprised me -- A LOT. However, the NAI sniffer software definitely
sees the traffic between the nodes on the switch. I was able to see
requests going through the switch for HTTP, SSH, etc. from other nodes.
The switch was far from overloaded. At the time there were only 2 users
using the network, with very light load.
-Bill
Philip J. Koenig
Next question: how MUCH of the other traffic did you see..
entire conversations? Just random packets? Packets from
both sender and receiver? Packet fragments?
I wonder if it was an overload problem. Were they 100mbit
hosts talking to 10mbit hosts? That might create a buffer
shortage.
Philip J. Koenig
> In ba.internet Philip J. Koenig <See_email_@ddress_below.this_one_is.invalid> wrote:
> > Oh my, that even surprises me. Sounds like something either
> > isn't working right, or they are engaging in deceptive advertising.
>
> > The only time you should see a switch leak traffic to other
> > nodes is A) broadcasts or B) if some switches are highly
> > congested (ie no more buffers), they will start bridging
> > traffic.
>
> It surprised me -- A LOT. However, the NAI sniffer software definitely
> sees the traffic between the nodes on the switch. I was able to see
> requests going through the switch for HTTP, SSH, etc. from other nodes.
>
> The switch was far from overloaded. At the time there were only 2 users
> using the network, with very light load.
>
> -Bill
Very weird. If I were you I'd call up their tech dept and
see what kind of response you get. (probably not much I
suppose.. Probably like those guys at Efficient Networks
that blew off John Pierce when he told them about the
problems he was having with his data-pattern-sensitive DSL
link)
That whole thing really opens up a whole train of thought
in regards to computer technology: that in many cases,
vendors can claim all sorts of things and *most* customers
would never be the wiser if their product didn't do exactly
what they claim.. yet they get the marketing capital by
making the claim anyway.
I read an article today (more like an advertorial if you
ask me) syndicated from an IDG publication (Java World?)
which was touting all the wonderful benefits of Java in
net-enabled phones and so forth. In the article the guy
tries to claim that Java would give the vendors access to
all those "plentiful java internet developers" which a
"C" based environment couldn't do.. ie trying to imply
that "C internet developers" were rare or something.
Haha! He may as well have tried to claim that the
Commodore 64 was the big upcoming thing in computing..
(OK, not really.. but it was pretty borderline)
John Navas
See_email_@ddress_below.This_one_is.invalid (Philip J. Koenig) wrote:
>That whole thing really opens up a whole train of thought
>in regards to computer technology: that in many cases,
>vendors can claim all sorts of things and *most* customers
>would never be the wiser if their product didn't do exactly
>what they claim.. yet they get the marketing capital by
>making the claim anyway.
Interesting coincidence -- I was making this same point to my wife
yesterday. I had asked her why she hadn't told me that her software
wouldn't do something that it was supposed to do. She said she had
thought it must have been something that she was doing wrong.
I think this is a major reason that we have to put up with such crappy
computer product quality -- now that the majority of the market is made up
of people who are in awe of and intimidated by computers, vendors know
that customers will tend to blame themselves rather than the vendor. It's
also a factor in the ISP business, where ISP's routinely blame "The
Internet" rather than admit responsibility for problems that are really
under their control. I think it also accounts for the popularity of AOL
-- AOL users I know have no idea how slow and limiting it is.
This all seems to have reached a new fever pitch, with computer product
marketing claims worthy of snake oil purveyors. What makes it work is
that much of the press has degenerated into little more than a recycling
mechanism for "infomercials." The sorry state of anti-virus protection is
but one of many examples -- for an interesting rant, see "CERT missed
Melissa's ultimate lesson" at
<http://kumite.com/myths/opinion/certmisd.htm>.
My fear is that firewall software will go the same way, given the way the
press has been dazzled with wildly exaggerated claims like these:
BlackICE Defender offers INSTANT INSTALLATION, no configuration
necessary, making it a product for the masses. BlackICE Defender
will detect intruder attempts, identify who they are and STOP them
dead in their tracks. Running on any Windows 95, 98, or NT based
system, BlackICE Defender consists of a sophisticated network
monitoring engine that can scan all inbound and outbound traffic on
your PC for suspicious activity. On finding an attempt to breach
your computer, BlackICE seamlessly and silently leaps into action
denying the hacker access to your computer while leaving your
legitimate traffic unaffected!
Gathers information about attackers using sophisticated
backtracing features.
Can automatically block all inbound traffic from an intruder
without any effect on legitimate traffic.
Silent, passive operation. Hackers cannot detect BlackICE
running and therefore cannot disable it.
Dynamic filters and advanced monitoring algorithms ensure
legitimate Internet or network traffic is unaffected.
Colorful graphs, simple user interface, and extensive
documentation make identifying network and Internet intrusions
easy for anyone.
Our Defender product contains personal firewall functionality.
It contains fewer "static" rules, but contains "dynamic"
automated rules. Existing personal firewalls do not detect
intrusions, they simply block undesirable traffic. This requires
the end-user to program exact what "undesirable" means. Defender
contains advanced intrusion countermeasures, among which is a
dynamic firewall programmed by the rest of the system. As a
result, no programming is required.
This is a competitive analysis of our product when stacked up
against AtGuard, Conseal, and the soon-to-be-announced personal
firewall products from Symantec/Norton, NAI, and others. Our
product is a "next-generation" technology, and is the only product
among the bunch that can detect hacker signatures regardless if the
traffic matches the firewall rules.
Our product does not ask questions. Because we can differentiate
between normal network activity and hacker activity, we
automatically block hacker activity while allowing normal activity
to flow. Our product is designed to protect the system with zero
configuration by the user.
p.s. To be clear, I'm not saying that BlackICE Defender is a bad product
-- I'm only picking on it because it's a particularly bad example of
marketing hype that chokes out reason.
Christopher Painter
that 1) A switch would be harder to support because of autoconfiguration
problems and that 2) a switch wouldn't really by any additional performance
with just a few hosts.
Since then you have come back to say
> If people think that two small boxes are "too complicated",
> and those people are trying to put together a network,
> their problems have only just begun.
Sounds like you have turned around on the first point.
> Only in the face of belligerent people who attempt to
> authoritatively spew misinformation.
I don't know who you think you are, but I have not "spew"ed one bit of
misinformation. Infact if you would like to challenge my credentials, I
would be happy to send you my resume. I don't know about where you work,
but in my group anyone who is as closed minded as you would be let go. Your
the one who seems to go around this newsgroup telling everyone that they are
wrong and that you have the right answer.
Chris
"Philip J. Koenig" <See_email_@ddress_below.This_one_is.invalid> wrote in
message news:MPG.132b571fa...@news.brainstorm.net...
> In article <89n6h8$gu2$1...@bob.news.rcn.net>, cpai...@not-it.dalenet.com
> (Christopher Painter) writes...
>
> > The NetGear FS-516 (rackmount 16port 10/100 autosensing ) that I use at
work
> > has dip switches on the back of it for duplex mode, and I picked it up
for
> > under $400. Is that low end enough for you? Besides we were talking
about
> > SOHO environment. Its not like I'm going to have a rats nest of
equipment
> > tied to each other to troubleshoot.
>
>
> The discussion revolved around the inclusion of a switch
> (rather than hub) into a firewall product which sells for
> under $500 all-inclusive. Hardly a candidate to include
> $400 worth of switch hardware.
>
> Low-end switches are less than $100 these days, and it
> would seem fairly clear that these are the class of
> devices (with no manual port-configuration facility) that
> would likely be included in a firewall which has a retail
> price of <$500, if at all.
>
> Then of course you'll have all the people kvetching that
> the firewall only has a 4-port hub or switch, but they
> have 5 workstations.
>
> I'd think people would rather have the flexibility to buy
> whatever hub or switch they want, not have to pay for
> something built into the firewall that may not meet their
> needs, and then get whatever does suit their needs. (ie
> a $99 5-port switch if they so desire)
>
> If people think that two small boxes are "too complicated",
> and those people are trying to put together a network,
> their problems have only just begun.
>
>
> > Besides, where do you come off acting like you are the only person in
the
> > world with any networking experience and that only your opinion matters?
> >
> > Chris
>
>
> Only in the face of belligerent people who attempt to
> authoritatively spew misinformation.
>
>
Philip J. Koenig
spamf...@navasgrp.dublin.ca.us (John Navas) writes...
> [POSTED TO ba.internet]
> See_email_@ddress_below.This_one_is.invalid (Philip J. Koenig) wrote:
>
> >That whole thing really opens up a whole train of thought
> >in regards to computer technology: that in many cases,
> >vendors can claim all sorts of things and *most* customers
> >would never be the wiser if their product didn't do exactly
> >what they claim.. yet they get the marketing capital by
> >making the claim anyway.
>
> Interesting coincidence -- I was making this same point to my wife
> yesterday. I had asked her why she hadn't told me that her software
> wouldn't do something that it was supposed to do. She said she had
> thought it must have been something that she was doing wrong.
Yep, common presumption -- and the corollary to that is, people
are too embarrassed that they might be stupid or technically
ignorant to complain to anyone, for fear that other people will
"find out they're stupid".
> I think this is a major reason that we have to put up with such crappy
> computer product quality -- now that the majority of the market is made up
> of people who are in awe of and intimidated by computers, vendors know
> that customers will tend to blame themselves rather than the vendor. It's
> also a factor in the ISP business, where ISP's routinely blame "The
> Internet" rather than admit responsibility for problems that are really
> under their control. I think it also accounts for the popularity of AOL
> -- AOL users I know have no idea how slow and limiting it is.
Yep. My feeling is that this is the fundamental reason behind
why most PC products these days have near-useless tech support.
Since the vast majority of PC users are so tickled that *anything*
works, the vendors see no need to staff their support lines with
reps with any in-depth knowledge.. most of the time they can just
spew some gobbledygook, tell the caller to go reinstall the OS,
hang up and go on to the next victim.
> This all seems to have reached a new fever pitch, with computer product
> marketing claims worthy of snake oil purveyors. What makes it work is
> that much of the press has degenerated into little more than a recycling
> mechanism for "infomercials." The sorry state of anti-virus protection is
> but one of many examples -- for an interesting rant, see "CERT missed
> Melissa's ultimate lesson" at
> <http://kumite.com/myths/opinion/certmisd.htm>.
Interesting article (I only read part), but I disagree. I think
that Melissa's ultimate lesson was "Don't use Microsoft email
client software". :-)
BTW, I thought that the "heuristic" facility in many antivirus
products today was exactly what the author of the above article
was claiming didn't exist any more: ie "profile" based antivirus
detection, as opposed to "signature" based.
Interesting when you look at the big picture though -- what
engine for continuing revenue-generation would exist if these
antivirus-purveyors didn't have an excuse to keep you on the
hook forever for virus pattern updates?
> My fear is that firewall software will go the same way, given the way the
> press has been dazzled with wildly exaggerated claims like these:
>
> BlackICE Defender offers INSTANT INSTALLATION, no configuration
> necessary, making it a product for the masses. BlackICE Defender
Nah, just low-end and "personal" firewall products, and it's
the same reason everything on the low-end in any market does
it: "Selling the sizzle and not the steak" is a time-honored
tradition when it comes to mass-marketing.. and "personal
firewalls" are about to become a mass-market product. (thanks
to DSL, cable internet, a PC in every pot, and an upswing in
hacker fear)
Philip J. Koenig
(Christopher Painter) writes...
> That is not entirely true. The conversation involved you making the claim
> that 1) A switch would be harder to support because of autoconfiguration
> problems and that 2) a switch wouldn't really by any additional performance
> with just a few hosts.
>
> Since then you have come back to say
>
> > If people think that two small boxes are "too complicated",
> > and those people are trying to put together a network,
> > their problems have only just begun.
>
> Sounds like you have turned around on the first point.
Hardly. The argument is "put a switch in there because
it's "better, simpler, and it improves performance".
I still disagree. I still think it's going to cause way
more support calls for autoconfig problems than it's worth
for a company who's really trying to sell a firewall.
I stand by my contention that the "performance improvement"
for most small networks is likely to be unnoticeable if
not nonexistent.
Expanding on my other sentence: If people think that having
a separate firewall and a switch/hub is "too complicated"
(the argument being putting it inside the firewall somehow
makes it "easier") then I really don't think those people
that find it so difficult to connect the 2 boxes together
with a patch cable are competent enough to deal with the
variety of (far more complex) issues that will come up with
setting up this network, defining security policies, monitoring
for intrusions, etc.
John Navas
See_email_@ddress_below.This_one_is.invalid (Philip J. Koenig) wrote:
>In article <8d9x4.6222$sR2.1...@news.swbell.net>,
>spamf...@navasgrp.dublin.ca.us (John Navas) writes...
>> ... The sorry state of anti-virus protection is
>> but one of many examples -- for an interesting rant, see "CERT missed
>> Melissa's ultimate lesson" at
>> <http://kumite.com/myths/opinion/certmisd.htm>.
>
>BTW, I thought that the "heuristic" facility in many antivirus
>products today was exactly what the author of the above article
>was claiming didn't exist any more: ie "profile" based antivirus
>detection, as opposed to "signature" based.
Most of the "heuristic" facilities are simply variations on signature
checking. ( They really aren't "heuristic" in any event since there is no
self-learning in the system, but then the marketing folks don't seem to be
terribly concerned with accuracy. ;)
For insight on what the article really had in mind, see Integrity Master
<http://www.stiller.com/>, and in particular
<http://www.stiller.com/AVSW.htm>.
>Interesting when you look at the big picture though -- what
>engine for continuing revenue-generation would exist if these
>antivirus-purveyors didn't have an excuse to keep you on the
>hook forever for virus pattern updates?
"Ay, there's the rub."
Christopher Painter
InstallShield work for a major DoD medical logistics project when on one
particular build ( Powerbuilder P-Code ) the IBM Antivirus on our clients
computers suddenly decided to think our program was a virus. Of course I
have had the same feelings about that old project myself but thats another
story. The funny thing was Norton had allready purchased IBM Antivirus,
and DISA only provided site licenses for Norton. Yet we had another prime
contractor ( IBM ) at our customers site and they had decided to go with IBM
Antivirus. They just kept insisting that the Install for the our client was
broken. They wouldn't even believe white papers from InstallShield
Corporation mentioning what virus programs can do to installs. At our other
sites running McAfee and Norton everything was fine. Fortunatly the next
build the code changed and the virus scanner stopped thinking it was a
virus.
Chris
"John Navas" <spamf...@navasgrp.dublin.ca.us> wrote in message
news:emBx4.40$yd5....@news.swbell.net...