The OTHER problem with Netgear WGT624 (and probably others)

On Sat, 12 Aug 2006 09:15:16 -0700 Jeff Liebermann wrote: | snipped-for-privacy@ipal.net hath wroth: | |>Good example. However, I don't see how this example says that every |>small business needs to consider routers to be worth the expense when |>it means everything else has to go a bit further down in financing. | | Oh, that's easy. Calculate the maintenance costs of buying bottom of | the line equipment versus the capital equipment expenditure required | to prevent such maintenance costs. Also, if you look at the pricing | on commercial wireless installations, the cost of the hardware is | fairly small compared to the cost of the labor involved. If an | increase in capital equipment expenses can offset a labor charge, then | it might be worth the effort.

You're assuming that the maintenance cost is inevitable.

| Anyway, you've apparently missed my point. This is not about initial | hardware expenses. It's about the overall cost of ownership, of which | the initial hardware costs are only a small part. If better hardware | causes fewer problems, then it's a good investment.

Do you figure in the percentage probability of an expense and prorate it across all expenditures to calculate an average cost of ownership? Or are you assuming the worst case where everything will fail and drive costs to a maximum?

| Incidentally, not all businesses run on the bitter edge of bankruptcy, | where nickels and dimes are counted. I don't see too many of these as | they cannot afford my exorbitant labor rates.

The ones I know of that run in the bitter edge of bankruptcy are the ones that overspent to begin with.

|>So, do you have a list of routers that will survive a firmware flash |>without power? Or did you just recommend buying a UPS? | | They have a UPS. The power glitch went right through it. Again, you | missed my point. This is not about hardware selection. It's about | the effects of wireless/internet downtime on a business, and how | wireless has gone from a frill to a necessity.

Sounds like someone bought a cheap UPS. I don't always put them in, depending on need, but where I do, I get the dual-conversion type that are always converting AC to DC, paralleling the battery, and converting DC back to AC, 100% of the time. I never see glitches with those.

|>A business actually providing internet access in relation to the business |>they are doing really does need to consider the business continuity impact |>of the decisions. But not all businesses are doing that. | | Gibberish. I think you're saying that only wireless ISP's need to | consider wireless reliability. Try again. I have a few corporate | customers who use wireless to offset the cost of wiring their | building. Wireless is most certainly "mission critical".

Wireless ISPs are only one example.

If there is mission critical work going on _inside+ a building, I most certainly _never_ recommend wireless be used. I recommend a wired LAN in all cases. Or fiber in certain cases.

But it depends on the nature of what the business is doing and how they use the internet. Most can survive a few hours to a couple days if the LAN goes down, or if just the internet access goes down. Some can't, and they would be advised for something better (with different choices given costs and other requirements).

|>| I think I have about 4 customers currently using Sonicwall TZ170w |>| wireless routers. These are not cheap, but very reliable and very |>| feature infested. See: |>

|>And how many of these customers operate a business in which network |>access is directly related to the operation of their business? | | Three of them. If the wireless goes down, the laptops with the days | mobile warehouse updates requires that the drivers remove the laptops | from the trucks and drag them inside. Something similar with the | other warehouse. The 3rd is a medical office where they could | probably live without wireless but it would be a noticeable | inconvenience. The 4th is a coffee shop, who wanted to isolate the | customer laptops from the internal network. Incidentally, the owner | is a former engineering manager who certainly knows his networking but | doesn't want to waste his time screwing with it.

All but the 3rd seem to be cases where wireless is essential. Of course I don't know the details of the medical office and how wireless would be used there. But it would tend to concern me if I was making use of their service and they were making use of wireless.

|>| Anyway, if you have problems evaluating equipment, try to estimate |>| what downtime or a failure will cost. Then compare this downtime cost |>| with the initial hardware costs. | |>I do that. | | I don't see any evidence of that. You don't appear to think like a | small business owner, where both the short and long term costs of | everything are considered in each decision. If businesses were | perpetually on the bitter edge of bankruptcy, and capital equipment | expenditures were such an agonizing experience, the SMB (small medium | business) sector would have collapsed long ago.

Just because I take a hard critical look at things and don't come to the same conclusion you do doesn't mean I don't consider such things. But I'm not running the client's businesses, either. I just know how such businesses often are run from past experience. The ones that are more successful tend to be the ones that avoid overspending on all the things that are not a direct function of what the business is. For example a delivery business can't afford to go cheap on their trucks, but that can afford to go cheap on copy machines if they are not limited to a single one. A TV station can't afford to go cheap on the transmitter or master control, and in most cases the studio equipment. But they can go cheap on many other things that would be just a short term inconvenience if they fail (a new operation can't do that with remote cameras).

| One of the questions I'm always asked is "How long with this thing | last before it's obsolete and needs to be replaced" (i.e. life cycle | costing). With SMB wireless and computing in general, that's a very | real problem, but one which you haven't bothered to even mention in | your wireless hardware decision making process.

Funny you bring that up. It is in fact a frequent reason to spend less on things. I generally do recommend going cheap on the computers based on giving them a certain number of years before they are obsolete. But in many cases there has been a workable "hand me down" scheme where people without the need for the latest computer power end up with the computer the manager once used. Computers can be stretched to usage over quite many years (which is going to kill a LOT of Vista sales for Microsoft ... you just watch).

| It's also not easy to estimate downtime costs. I worked for about 30 | hours trying to nail down a realistic estimate of what a large medical | office recovery will cost in downtime and recovery. I had to revise | the numbers about 5 times in as many years. It required that I | document various disaster recovery scenarios. One of these scenarios | actually happened, and I found myself having underestimated the MTTR | (mean time to recover) by an embarrassing margin. I've also done dry | run recoveries with the usual surprises. As I said, I don't think | you've ever estimated the cost of downtime or failure for a business.

Yes I have. But not in a medical setting.

Speaking of medical settings, it sounds like you are working pretty in depth in that field, not just the wireless/networking setting. If that does include their computers, I sure hope you are taking appropriate consideration for verifiably wiping out the contents of all computer storage devices leaving such offices because of replacements or being upgraded. If a hard drive is replaced because it's too small, do you wipe off its content and check that it is, or irreversibly destroy it?

|>But I also figure in how they utilize the network. Most of |>the small businesses won't see much of a short term (e.g. sufficient time |>to buy a replacement router or otherwise determine what happened) impact. | | Any small business that can afford my exorbitant service rates, can | certainly afford to spend the money on superior hardware to avoid my | presence. More simply, I don't see many businesses where wireless is | NOT an important part of the business. It's not unusual for me to | charge more than the cost of the hardware to fix a system. They pay | because they really need the system to work, but they also ask about | getting better hardware.

It sounds like their wireless/networking failures are on average exceeding the failures in other areas.

| It also applies to home users with wireless. Most can manage without | wireless for a while, but I usually get the call within hours of the | wireless failing because it's has usually become so much a part of the | daily home life, that the inconvenience of not having wireless | connectivity through the house is considered a serious problem. It's | not unusual for me to arrive and find a brand new wireless router | purchased in the hope that it would magically solve the problem. (Two | junk routers do not equal one good router). I know that if my | neighborhood wireless system goes down, I get phone calls within an | hour or two. There may be home users that can live with crappy | wireless range and performance, but I don't see those. | | A TZ170w is about $750. A cheapo wireless router is about $100. The | $650 difference is about 9 hours of my lab our. Ignoring interest | charges, if buying a better router can be balanced against my service | charges, downtime costs, recovery costs, and early replacement costs | over the life of the system, then it's a bargain.

If you are seeing 100% failure rates in the cheapo wireless gear, then I suppose the figures you give are right.

A business that is spending $10,000 on various things in a year, of which networking is just one part, has to consider the potential failure of all those things. Certainly if the cheapo wireless device has 100% failure rate, then going with something better is worth it. The thing is, the usual practice is statistical based. They do assume that some proportion of all that is spent will require more expenditure. Knowing a rough idea of the failure rates certainly helps. But I have a hard time believing that the failure rate of Netgear and Linksys stuff bought at an office supply store is anywhere close to 100%. Probably not above 15% within the first year. But I'm just beginning on wireless, so if you know the failure rate is higher, I guess I have to take your word for it.

Reply to
phil-news-nospam
Loading thread data ...

On 14 Aug 2006 04:26:09 GMT, snipped-for-privacy@ipal.net wrote in :

VPN is the means of access to the isolated server.

Do you really not understand that, or are you just being argumentative?

Way more sophisticated than that.

Mine doesn't.

Again, way more sophisticated than that. It's why I get paid the (not so) big bucks. :)

Reply to
John Navas

On Sat, 12 Aug 2006 16:59:37 GMT John Navas wrote: | On Sat, 12 Aug 2006 09:15:16 -0700, Jeff Liebermann | wrote in | : | |>A TZ170w is about $750. ... | | 25 node for $513.00, in stock: |

Just how many bytes is needed to store info for one node? Sheesh? 25?

Reply to
phil-news-nospam

On 14 Aug 2006 05:08:04 GMT, snipped-for-privacy@ipal.net wrote in :

Your point?

Reply to
John Navas

On 14 Aug 2006 05:05:31 GMT, snipped-for-privacy@ipal.net wrote in :

So you buy expensive UPS, but cheap out on wireless? Makes no sense.

Not my clients.

Reply to
John Navas

On Mon, 14 Aug 2006 05:07:18 GMT John Navas wrote: | On 14 Aug 2006 04:26:09 GMT, snipped-for-privacy@ipal.net wrote in | : | |>On Sat, 12 Aug 2006 00:16:54 GMT John Navas wrote: |>| On 11 Aug 2006 17:23:05 GMT, snipped-for-privacy@ipal.net wrote in |>| : |>| |>|>On Wed, 09 Aug 2006 14:31:39 GMT John Navas wrote: |>|>| On Wed, 09 Aug 2006 02:42:28 GMT, John Navas |>|>| wrote in |>|>| : |>|>| |>|>|>On 8 Aug 2006 21:41:27 GMT, snipped-for-privacy@ipal.net wrote in |>|>|>: |>|>| |>|>|>>So how would you connect a doctor's office to the internet to be in full |>|>|>>compliance with HIPAA? |>|>|>

|>|>|>Isolated network zone, enforced by router and firewall rules. |>|>| |>|>| More Complete Answer: My preferred solution is to put the server behind |>|>| a dedicated firewall and VPN endpoint that only allows connections over |>|>| authenticated VPN and blocks all outbound connections. The hardware is |>|>| secured in a locked cabinet which only the administrator can open. Even |>|>| if the LAN/WLAN were to be breached, the server would remain fully |>|>| secure behind the firewall, which logs activity and access attempts that |>|>| are regularly reviewed as part of the ongoing security process. When an |>|>| intrusion attempt is detected, the administrator is automatically paged. |>|>| All of this can easily be done on a modest budget -- all it really takes |>|>| is skill, business and technical. |>|>

|>|>How does VPN help an office connect to the internet? |>| |>| Isolates the server. |>

|>And how does isolating the server have anything to do with VPN? | | VPN is the means of access to the isolated server.

If it is accessible, it isn't isolated.

| Do you really not understand that, or are you just being argumentative?

It makes no sense to isolate a server you're trying to make accessible. Sure, protecting it from undesired access is appropriate. That is what a firewall is for. But you then configure the firewall to pass the kind of access ot it that you want. A VPN has nothing to do with it.

| |>|>BTW, I used to do intrusion attempts by pager. No longer. The noise level |>|>is way too high. Intrusion attempts are at such a high level these days |>|>that if you were to respond to each one, you would get very little else |>|>done. |>| |>| My filter and threshold results in relatively few alerts. |>

|>If you block the common points of hacker attacks and don't enable them |>for alerts, that would work. What is your threshhold? 10,000? | | Way more sophisticated than that.

When an attempt happens, you do something about it. At one extreme you just totally ignore the attempt. At the other extreme you report it. Somewhere in between you either randomly or intelligently decide to do different things with each different attempt. Maybe some attempts are more important than others because they are aiming at weaker points (do you have weaker points?).

But if you are going to beep the pager for each and every attempt, you're going to be majorly disturbed.

|>|>What you do is block the access in a sufficiently confident way that |>|>you don't have to concern yourself with attempts. |>| |>| That kind of confidence is almost always misplaced in my experience. |>

|>You have to make some sort of tradeoff. You can't be having your pager |>go off every couple minutes every time someone probes some port. | | Mine doesn't.

Which means you then have attempts for which your firewall chooses by some means to NOT notify you. You could do this with all attempts, or you could choose some attempts to notify you about as being particularly interesting for some reason.

|>|>So unless you have access |>|>to the means to track down and prosecute those who attempt to intrude (e.g. |>|>you run the FBI network, for example), don't waste your time. |>| |>| I don't waste my time. I do protect my clients. |>

|>But I wonder if you are getting paged for every intrusion attempt, or if |>you are just ignoring the bulk of them. With so many going on (rarely |>do I ever see an hour long period without an attempt), it has to be one |>or the other. Or are you now using knock-knock access schemes? | | Again, way more sophisticated than that. It's why I get paid the (not | so) big bucks. :)

But at least its not paging for every intrusion attempt. It seemed like when you first mentioned this that was what you meant.

Reply to
phil-news-nospam

On Mon, 14 Aug 2006 05:29:22 GMT John Navas wrote: | On 14 Aug 2006 05:08:04 GMT, snipped-for-privacy@ipal.net wrote in | : | |>On Sat, 12 Aug 2006 16:59:37 GMT John Navas wrote: |>| On Sat, 12 Aug 2006 09:15:16 -0700, Jeff Liebermann |>| wrote in |>| : |>| |>|>A TZ170w is about $750. ... |>| |>| 25 node for $513.00, in stock: |>|

|>

|>Just how many bytes is needed to store info for one node? Sheesh? 25? | | Your point?

It's an awfully low number. Too little RAM? Or bloated firmware? Or just an artificial limitation?

What hardware do you recommend which can be loaded with custom firmware (such as a Linksys WRT54GL) that is in your "reliable" category?

Reply to
phil-news-nospam

On Mon, 14 Aug 2006 05:32:15 GMT John Navas wrote: | On 14 Aug 2006 05:05:31 GMT, snipped-for-privacy@ipal.net wrote in | : | |>On Sat, 12 Aug 2006 09:15:16 -0700 Jeff Liebermann wrote: | |>| They have a UPS. The power glitch went right through it. Again, you |>| missed my point. This is not about hardware selection. It's about |>| the effects of wireless/internet downtime on a business, and how |>| wireless has gone from a frill to a necessity. |>

|>Sounds like someone bought a cheap UPS. I don't always put them in, |>depending on need, but where I do, I get the dual-conversion type that |>are always converting AC to DC, paralleling the battery, and converting |>DC back to AC, 100% of the time. I never see glitches with those. | | So you buy expensive UPS, but cheap out on wireless? Makes no sense.

No. Where reliability is critical, I go with the good equipment AND the good UPS to back it up. I'll take it that YOU did not specify the UPS for the place that had this problem, but only specified the router. It's just that this is NOT the case everywhere.

|>But it depends on the nature of what the business is doing and how they |>use the internet. Most can survive a few hours to a couple days if the |>LAN goes down, or if just the internet access goes down. ... | | Not my clients.

You have select clients. I guess you're in it only for the big bucks. If there are indeed enough such clients around, maybe the rest of us should try that :-)

Reply to
phil-news-nospam

On 15 Aug 2006 02:01:59 GMT, snipped-for-privacy@ipal.net wrote in :

It's logically isolated, not physically isolated.

We'll just have to agree to disagree.

Again, way more sophisticated than that.

Not in systems I set up.

Again, way more sophisticated than that.

That was your assumption, not something I said (wrote).

I just don't have the time or patience for more pointless wrangling on this, so I'm now giving you the last word.

Reply to
John Navas

On 15 Aug 2006 02:05:15 GMT, snipped-for-privacy@ipal.net wrote in :

It's a usage license, standard practice in the enterprise market.

I don't.

Reply to
John Navas

On 15 Aug 2006 02:09:13 GMT, snipped-for-privacy@ipal.net wrote in :

You take it wrong.

I do.

Reply to
John Navas

snipped-for-privacy@ipal.net hath wroth:

I'm in the maintenance business, so that's a fair assumption. I call to your attention the large number of software and firmware, updates and fixes.

Neither. My job was to calculate the cost of a crash and recovery, not to predict its incidence. That's the insurance company's problem. Please note that I suggested that you calculate the cost of a failure, not the probability of it happening.

There are those, I'm sure. If you look hard enough for a problem, you're sure to find it. Most of my customers have been with me for perhaps 15 years and have remained sufficiently profitable to pay my exhoribitant fees. My biggest headache is currently that many of them are retireing or selling their businesses.

Yep. APC BackUPS ES 350VA. Very cheap. Works well enough for power failures, but doesn't stop the small glitches. In this case, there's construction going on next door, which probably is causing the glitches.

Yep. SmartUPS series. No switching time, but there's a price. They suck power in standby. I recently measured an APC1400RH, which burned about 40 watts (70VA) doing nothing. At $0.15/kw-hr, that's about $50/year in electricity.

Chuckle. Many years ago, I tried to setup a 900MHz wireless link (all that was available at the time) across the freeway between a hospital and the medical office buildings. The hospital was "concerned" that I might be "irradiating" the patients and killed the plan. Years later,

2.4GHz wireless is all over the hospital, plus a cell site on the roof, but no wireless link across the freeway. The big antennas I proposed must be more dangerous than small antennas.

Most of the wireless is the medical offices is used for updating patient records in real time and thus eliminate medical transcription. In the hospital, it's also used for telemetry and monitoring.

I've never worked for a TV station, but I have worked for AM and FM stations. Let me assure you that cutting corners on the transmitter and studio equipment is very common. I've seen more money spent on furniture than on equipment. Perhaps there are stations that value their transmitters, but after 3 college and 3 broadcast stations, I haven't seen any.

The current philosophy is that two pieces of junk, one acting as a backup for the other, is better and cheaper than one allegedly "reliable" device. I have some rather "interesting" photos of some of transmitters and sites.

I'll take your word for it. After paying for my learning curve, I soon discovered that few IT people have even a clue what downtime would cost a company or even how long it would take to recover from a crash.

We smash the drives with a large hammer. However, I do break the rules on upgrades and use a "disk nuker" problem to wipe the drive. No sense in ruining a perfectly good drive. Keeping track of the drives is a major problem but is being handled.

Time's up.

Reply to
Jeff Liebermann

On Mon, 14 Aug 2006 22:36:58 -0700, Jeff Liebermann wrote in :

Standby type.

Line interactive type, not dual conversion; e.g.,

Reply to
John Navas

On Tue, 15 Aug 2006 03:59:20 GMT John Navas wrote: | On 15 Aug 2006 02:01:59 GMT, snipped-for-privacy@ipal.net wrote in | : | |>On Mon, 14 Aug 2006 05:07:18 GMT John Navas wrote: | |>| VPN is the means of access to the isolated server. |>

|>If it is accessible, it isn't isolated. | | It's logically isolated, not physically isolated.

Which is what a firewall does (as long as you don't have something else on the net that isn't isolated and leaks).

|>| Do you really not understand that, or are you just being argumentative? |>

|>It makes no sense to isolate a server you're trying to make accessible. |>Sure, protecting it from undesired access is appropriate. That is what |>a firewall is for. But you then configure the firewall to pass the |>kind of access ot it that you want. A VPN has nothing to do with it. | | We'll just have to agree to disagree.

In this case it's more than just disagreement.

|>|>If you block the common points of hacker attacks and don't enable them |>|>for alerts, that would work. What is your threshhold? 10,000? |>| |>| Way more sophisticated than that. |>

|>When an attempt happens, you do something about it. At one extreme |>you just totally ignore the attempt. At the other extreme you report |>it. Somewhere in between you either randomly or intelligently decide |>to do different things with each different attempt. Maybe some attempts |>are more important than others because they are aiming at weaker points |>(do you have weaker points?). | | Again, way more sophisticated than that.

You keep talking about sophistication, yet you don't even know what role a VPN plays?

|>But if you are going to beep the pager for each and every attempt, |>you're going to be majorly disturbed. | | Not in systems I set up.

You implied this in the first statement about it, that you would be notified of all intrusion attempts. I was hoping you just say something like you don't define certain kinds of attempts as sufficiently worthy to bother with.

| I just don't have the time or patience for more pointless wrangling on | this, so I'm now giving you the last word.

Well, at least I don't need to ask you about VPNs.

Reply to
phil-news-nospam

On Tue, 15 Aug 2006 04:00:26 GMT John Navas wrote: | On 15 Aug 2006 02:05:15 GMT, snipped-for-privacy@ipal.net wrote in | : | |>On Mon, 14 Aug 2006 05:29:22 GMT John Navas wrote: |>| On 14 Aug 2006 05:08:04 GMT, snipped-for-privacy@ipal.net wrote in |>| : |>| |>|>On Sat, 12 Aug 2006 16:59:37 GMT John Navas wrote: |>|>| On Sat, 12 Aug 2006 09:15:16 -0700, Jeff Liebermann |>|>| wrote in |>|>| : |>|>| |>|>|>A TZ170w is about $750. ... |>|>| |>|>| 25 node for $513.00, in stock: |>|>|

|>|>

|>|>Just how many bytes is needed to store info for one node? Sheesh? 25? |>| |>| Your point? |>

|>It's an awfully low number. Too little RAM? Or bloated firmware? Or just |>an artificial limitation? | | It's a usage license, standard practice in the enterprise market.

Doesn't make it right, especially if it is already overpriced for what hardware it has.

|>What hardware do you recommend which can be loaded with custom firmware |>(such as a Linksys WRT54GL) that is in your "reliable" category? | | I don't.

So basically you leave your customers at the mercy of pixie dust salesmen and restrictive licenses that don't allow growth?

Reply to
phil-news-nospam

On Tue, 15 Aug 2006 04:01:47 GMT John Navas wrote: | On 15 Aug 2006 02:09:13 GMT, snipped-for-privacy@ipal.net wrote in | : | |>On Mon, 14 Aug 2006 05:32:15 GMT John Navas wrote: | |>| So you buy expensive UPS, but cheap out on wireless? Makes no sense. |>

|>No. Where reliability is critical, I go with the good equipment AND the |>good UPS to back it up. I'll take it that YOU did not specify the UPS |>for the place that had this problem, but only specified the router. ... | | You take it wrong.

So you actually specified a low end UPS that can't protect your customer?

|>|>But it depends on the nature of what the business is doing and how they |>|>use the internet. Most can survive a few hours to a couple days if the |>|>LAN goes down, or if just the internet access goes down. ... |>| |>| Not my clients. |>

|>You have select clients. ... | | I do.

Leaves more for the rest of us.

Reply to
phil-news-nospam

On Mon, 14 Aug 2006 22:36:58 -0700 Jeff Liebermann wrote: | snipped-for-privacy@ipal.net hath wroth: | |>You're assuming that the maintenance cost is inevitable. | | I'm in the maintenance business, so that's a fair assumption. I call | to your attention the large number of software and firmware, updates | and fixes. | |>Do you figure in the percentage probability of an expense and prorate |>it across all expenditures to calculate an average cost of ownership? |>Or are you assuming the worst case where everything will fail and drive |>costs to a maximum? | | Neither. My job was to calculate the cost of a crash and recovery, | not to predict its incidence. That's the insurance company's problem. | Please note that I suggested that you calculate the cost of a failure, | not the probability of it happening.

When weighing in a wide range of expenditures, probability plays a very important role. Given 100 purchases where the failure rate over the usage period is 3% then you can expect approximately 3 failures. Plan for it. But you don't have to plan for 50 failures in this case.

Businesses that are not flush with cash have to figure this in so they can minimize the expenditure on as many things as possible.

|>Sounds like someone bought a cheap UPS. | | Yep. APC BackUPS ES 350VA. Very cheap. Works well enough for power | failures, but doesn't stop the small glitches. In this case, there's | construction going on next door, which probably is causing the | glitches.

Where glitches don't matter, but having some power during an outage does, then I suppose that is an OK choice for that power size. I would not use such a thing for mission critical functions.

|>I don't always put them in, |>depending on need, but where I do, I get the dual-conversion type that |>are always converting AC to DC, paralleling the battery, and converting |>DC back to AC, 100% of the time. I never see glitches with those. | | Yep. SmartUPS series. No switching time, but there's a price. They | suck power in standby. I recently measured an APC1400RH, which burned | about 40 watts (70VA) doing nothing. At $0.15/kw-hr, that's about | $50/year in electricity.

If you need glitch-free reliability, wouldn't that be worth it?

|>But it would tend to concern me if I was making use of their |>service and they were making use of wireless. | | Chuckle. Many years ago, I tried to setup a 900MHz wireless link (all | that was available at the time) across the freeway between a hospital | and the medical office buildings. The hospital was "concerned" that I | might be "irradiating" the patients and killed the plan. Years later, | 2.4GHz wireless is all over the hospital, plus a cell site on the | roof, but no wireless link across the freeway. The big antennas I | proposed must be more dangerous than small antennas.

The larger microwave ovens in hospitals are 900 MHz. That was probably a number that got them so concerned. A shame they totally misunderstood technology and you were unable to convince them of what was right.

That reminds of me a story where a hospital in Dallas was using a lot of monitoring devices in the 186-192 MHz range. Then a local TV station fired up their new digital TV transmitter on channel 9 that the FCC had licensed them with. The hospital's usage was all technically secondary. Ultimately they had to move, but the station voluntarily took the digital transmitter back off the air for a while to give them the time to do so. I wonder how that affected their technology planning or perspectives.

| Most of the wireless is the medical offices is used for updating | patient records in real time and thus eliminate medical transcription. | In the hospital, it's also used for telemetry and monitoring.

|>A TV station can't afford to go cheap on the transmitter |>or master control, and in most cases the studio equipment. | | I've never worked for a TV station, but I have worked for AM and FM | stations. Let me assure you that cutting corners on the transmitter | and studio equipment is very common. I've seen more money spent on | furniture than on equipment. Perhaps there are stations that value | their transmitters, but after 3 college and 3 broadcast stations, I | haven't seen any.

I've seen quite the opposite. I did volunteer work for a community FM station once. Everything was shoe-string, duct-tape, bailing wire :-) But when they finally got some major funding, it went for a new tower site, new antenna, new transmitter, and a license upgrade.

|>But they can |>go cheap on many other things that would be just a short term inconvenience |>if they fail (a new operation can't do that with remote cameras). | | The current philosophy is that two pieces of junk, one acting as a | backup for the other, is better and cheaper than one allegedly | "reliable" device. I have some rather "interesting" photos of some of | transmitters and sites.

That's not unlike the benefit of doubling the data speed when it increases the packet failure (requiring retransmission) rate by say 15%. It's still a win. OTOH, if latency is an issue, it might not.

|>| As I said, I don't think |>| you've ever estimated the cost of downtime or failure for a business. | |>Yes I have. But not in a medical setting. | | I'll take your word for it. After paying for my learning curve, I | soon discovered that few IT people have even a clue what downtime | would cost a company or even how long it would take to recover from a | crash.

When I talk to the managers about what to put in, I try to get these figures from them. I've actually had one manager (was a hired position) say "just make sure it never goes down". My reply was "is that a blank check?". Turns out he just assumed that "never go down" was a simple configuration choice. I finally convinced him _we_ had to work the numbers together and figure out how much to spend. I ended up putting in triple redundancy 7200 routers. A few months after I left, one of them did in fact fail (burned out power supply) and the guy who was responsible for the monitoring then figured it must be a misconfigured alarm since it was labeled "internet 1" and he could tell the internet was working just fine. So he turned the alarm off and they ran things that way for 3 months until the upstream ISP that router was connected to called and wondered why the circuit had stayed down for so long.

|>Speaking of medical settings, it sounds like you are working pretty in |>depth in that field, not just the wireless/networking setting. If that |>does include their computers, I sure hope you are taking appropriate |>consideration for verifiably wiping out the contents of all computer |>storage devices leaving such offices because of replacements or being |>upgraded. If a hard drive is replaced because it's too small, do you |>wipe off its content and check that it is, or irreversibly destroy it? | | We smash the drives with a large hammer. However, I do break the | rules on upgrades and use a "disk nuker" problem to wipe the drive. No | sense in ruining a perfectly good drive. Keeping track of the drives | is a major problem but is being handled.

I hope the hammer manages to at least warp the platters.

The disk nuker is probably adequate for the job. The one I wrote wiped the platters with random garbage just to be sure (since it cost no more time than just wiping with zeros). The problem is making sure people actually do it. At one place I worked at, I got a bunch of disk drives for a project that were manufacturer refurbs. On two of them I found medical database data, with medical info, SSNs, etc. I used that data to show the company's head auditor what risks we had in letting data out by recycled disks to get some policy changes.

Reply to
phil-news-nospam

snipped-for-privacy@ipal.net hath wroth:

None of the low end wireless hardware (Linksys, Dlink, Netgear, etc) specify an MTBF estimate and therefore has no track record or numbers with which to calculate a failure rate. Cisco 1300 series units are at 132,000 hrs (15 yrs). 400,000 hrs (45.5 yrs) for the PoE adapter. Sonicwall specifies the TZ150 at 79,000 hrs (9 yrs), and the TZ170 at

69,000 hrs (7.9 yrs).

I can make a fairly bad guess, based local circuit board heating, at the MTBF of some of the low end wireless hardware. It's not going to be very good. Ignorning infant mortality it's MTBF is generally not a concern unless it's less than the expected lifetime of the product. My guess(tm) is that low end hardware MTBF is at or below the expected lifetime.

Sure. Using what numbers for this calculation? Most businesses don't have the need, time, or expertise to estimate failure rates. They dive into consumer reports, read a few reviews, talk to consultants, assume that spending more money will get them a better product, and then buy exactly what their competitors are using because they know it works. Immitation is the safest way to buy.

Sure. Nothing is mission critical until it fails with a shop full of customers. Then, it magically becomes a crisis. I sometimes put a big fat capacitor across the 12VDC power input of various routers to minimize power related glitches. In the distant past, before cheap UPS's, I used to do the same thing inside PC's that were overly sensitive to power line glitches.

Sure, if it were disclosed by the manufactory. It seems that standby power consumption isn't commonly specified. I bought a "Kill-a-Watt" cheapo power meter for the purpose. Try selling a UPS to a commercial customer and then tell them that it will cost them $50/year to own. The switching time verus opertating cost discussion is always entertaining.

Not at the 3 local hospitals I'm familiar with. The heart monitors and crash carts all run on 900MHz wireless. 900MHz "industrial" were banned long ago. However, the real reason is that there is no international 915-928MHz frequency band. Therefore, the autoclave and oven manufacturers would need to make a different model for US and Europe. Too big a PITA so they settled on 2.4GHz, which also works somewhat better.

I'm not sure what they were thinking. The logic of the refusal was never adequately explained. My guess is that they took a look at the big 900MHz dish or yagi in the broshures and decided that it would be too ugly or something. Never mind the antenna farm and cell site on the roof.

Increasing the speed also reduces the range. Not much of a win if you can't communicate. Latency is affected by retransmissions, which are always bad. However, with adequate signal strength, the latency is hardly noticeable. If you're thinking of speed-o-light, at 10Km the increased latency is only 66 usec.

I can see it now.... Web based configuration page with a settable reliability figure. If I need some revenue, I just set the router to fail every few months. Nice feature.

Reply to
Jeff Liebermann

Intel saved you the effort of calculating the cost of downtime:

formatting link
Industry Sector Loss Revenue per Hour Energy $ 2.8 million Telecommunications $ 2.0 million Manufacturing $ 1.6 million Financial Institutions $ 1.4 million Information Technology $ 1.3 million Insurance $ 1.2 million Retail $ 1.1 million Pharmaceuticals $ 1.0 million Banking $ 996,000

That's just the revenue loss and does not include the cost of recovery, replacement hardware, overtime, bad press, etc. Obviously, this is for the loss of connectivity, server crash, or something major, not just a wireless router failure. Still, the numbers do suggest that failures are expensive.

Are you sure you still want businesses to buy the cheapest considering the cost of downtime?

Reply to
Jeff Liebermann

On Wed, 16 Aug 2006 09:54:30 -0700 Jeff Liebermann wrote: | snipped-for-privacy@ipal.net hath wroth: | |>When weighing in a wide range of expenditures, probability plays a very |>important role. Given 100 purchases where the failure rate over the usage |>period is 3% then you can expect approximately 3 failures. Plan for it. |>But you don't have to plan for 50 failures in this case. | | None of the low end wireless hardware (Linksys, Dlink, Netgear, etc) | specify an MTBF estimate and therefore has no track record or numbers | with which to calculate a failure rate. Cisco 1300 series units are | at 132,000 hrs (15 yrs). 400,000 hrs (45.5 yrs) for the PoE adapter. | Sonicwall specifies the TZ150 at 79,000 hrs (9 yrs), and the TZ170 at | 69,000 hrs (7.9 yrs).

Have those estimates proven accurate?

| I can make a fairly bad guess, based local circuit board heating, at | the MTBF of some of the low end wireless hardware. It's not going to | be very good. Ignorning infant mortality it's MTBF is generally not a | concern unless it's less than the expected lifetime of the product. My | guess(tm) is that low end hardware MTBF is at or below the expected | lifetime.

Then what is the "expected lifetime"?

|>Businesses that are not flush with cash have to figure this in so they |>can minimize the expenditure on as many things as possible. | | Sure. Using what numbers for this calculation? Most businesses don't | have the need, time, or expertise to estimate failure rates. They | dive into consumer reports, read a few reviews, talk to consultants, | assume that spending more money will get them a better product, and | then buy exactly what their competitors are using because they know it | works. Immitation is the safest way to buy.

It's not done very accurately. There is an assumption that a certain percentage of anything will fail, or otherwise need to be replaced, at a certain rate, usually around 5% per year. That is factored in to the overall cost, not of any one individual thing, but of everything overall. It can turn out that some items will fail early, and that such failure could have been predicted, had there been the resources to investigate it. But for that to be viable, _every_ purchase would have to be investigated, and that is not practical to do. So that leaves the practice of buying everything at commodity pricing, except for what one already knows to have a higher or earlier failure rate, and leave a percentage of financing available to cover a normal rate of failure overall.

|>Where glitches don't matter, but having some power during an outage does, |>then I suppose that is an OK choice for that power size. I would not use |>such a thing for mission critical functions. | | Sure. Nothing is mission critical until it fails with a shop full of | customers. Then, it magically becomes a crisis. I sometimes put a | big fat capacitor across the 12VDC power input of various routers to | minimize power related glitches. In the distant past, before cheap | UPS's, I used to do the same thing inside PC's that were overly | sensitive to power line glitches.

Or you could use a battery (on the 12 VDC stuff, not the PC).

|>| Yep. SmartUPS series. No switching time, but there's a price. They |>| suck power in standby. I recently measured an APC1400RH, which burned |>| about 40 watts (70VA) doing nothing. At $0.15/kw-hr, that's about |>| $50/year in electricity. |>

|>If you need glitch-free reliability, wouldn't that be worth it? | | Sure, if it were disclosed by the manufactory. It seems that standby | power consumption isn't commonly specified. I bought a "Kill-a-Watt" | cheapo power meter for the purpose. Try selling a UPS to a commercial | customer and then tell them that it will cost them $50/year to own. | The switching time verus opertating cost discussion is always | entertaining.

I bet it is an entertaining discussion.

|>The larger microwave ovens in hospitals are 900 MHz. | | Not at the 3 local hospitals I'm familiar with. The heart monitors | and crash carts all run on 900MHz wireless. 900MHz "industrial" were | banned long ago. However, the real reason is that there is no | international 915-928MHz frequency band. Therefore, the autoclave and | oven manufacturers would need to make a different model for US and | Europe. Too big a PITA so they settled on 2.4GHz, which also works | somewhat better.

So what frequency do the "industrial" large cavity microwave ovens use? They can't use 2.45 GHz in the larger sizes.

|>That was probably |>a number that got them so concerned. A shame they totally misunderstood |>technology and you were unable to convince them of what was right. | | I'm not sure what they were thinking. The logic of the refusal was | never adequately explained. My guess is that they took a look at the | big 900MHz dish or yagi in the broshures and decided that it would be | too ugly or something. Never mind the antenna farm and cell site on | the roof.

There are some nicer looking panel and rod antennas now days.

|>That's not unlike the benefit of doubling the data speed when it increases |>the packet failure (requiring retransmission) rate by say 15%. It's still |>a win. OTOH, if latency is an issue, it might not. | | Increasing the speed also reduces the range. Not much of a win if you | can't communicate. Latency is affected by retransmissions, which are | always bad. However, with adequate signal strength, the latency is | hardly noticeable. If you're thinking of speed-o-light, at 10Km the | increased latency is only 66 usec.

Increasing speed loses noise immunity and that is related to loss of range, since there is less signal at greater range to begin with to overcome the noise.

|>Turns out he just assumed that "never go down" was a simple |>configuration choice. | | I can see it now.... Web based configuration page with a settable | reliability figure. If I need some revenue, I just set the router to | fail every few months. Nice feature.

Interesting concept. But it's probably already patented.

Reply to
phil-news-nospam

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.