Windows XP SP3 Causing Router Crashes

KrispyBytes writes "Windows XP SP3 has been named as the culprit causing home routers to go into a crash and reboot cycle. One router maker has released firmware updates to fix the problem, but has not yet revealed what is actually different about XP SP3's networking stack or UPnP behaviour that causes the problem. Router maker Billion Managing Director Raaj Menon said "as Microsoft plans to make Windows XP SP3 an automatic upgrade this month, the number of affected routers may increase significantly.""

Comments

Maker?

[Anonymous Coward]

One router maker has I think you meant manufacture

Re:Maker?

[Neuropol]

One router maker has I think you meant manufacture i think YOU meant manufacturer

Re:Maker?

[ROMRIX]

i think YOU meant manufacturer

I think YOU meant "I"

Re:Maker?

[Anonymous Coward]

One router maker has I think you meant manufacture i think YOU meant manufacturer I do not think it means what you think it means.

Re:Maker?

[Anonymous Coward]

i think YOU meant manufacturer

I think YOU meant "I" I think YOU meant "I".

We could go on forever, couldn't we? :)

Re:Maker?

[Anonymous Coward]

I think you meant /"."

Re:Maker?

[Midnight Thunder]

I think YOU meant "I".

We could go on forever, couldn't we? :)

If you really wish to:

GOTO 23694689

Re:Maker?

[AHuxley]

"Manufacture" is the OEM shop floor.
Where almost the same products get a candy plastic cover with a different bezel or logo or 'brand'
The maker is the cubicle types with the huge FCC/ACMA guides and MS marketing guides.
They try to stay ahead of the MS pain.

Re:Maker?

[msim]

*boom*tish*

Not surprising

[noidentity]

Windows XP SP3 has been named as the culprit causing home routers to go into a crash and reboot cycle.

Not surprising Windows causes that when installed on a router, considering it also makes PCs go into a crash and reboot cycle when installed on them.

Re:Not surprising

[Anonymous Coward]

Lets not jump to blame this on Windows. It could be that Windows isn't doing anything wrong, just something the router should be able to handle, but can't. We can point fingers when we know what the actual issue causing the router problems is.

Re:Not surprising

[Anonymous Coward]

Wow, -1 Troll?

This is probably +2 funny, at least! Let's hope the mods come to their senses.

Re:Not surprising

[noidentity]

Well duh; it was just a lame attempt at a joke, not an actual claim that Windows was to blame. A properly-designed router will only reboot repeatedly if some client is logging in and causing it to reboot.

Re:Not surprising

[negRo_slim]

Lets not jump to blame this on Windows. It could be that Windows isn't doing anything wrong, just something the router should be able to handle, but can't. We can point fingers when we know what the actual issue causing the router problems is. Ya know I agree, as I've had SP3 installed in one form or another for some time now. With nary a problem... Can't even remember the last time I had to cycle my linksys befsx41. Besides I fail to see why a router should ever be affected to such a degree by a computer on it's network. Really does sound like flawed workmanship.

Re:Not surprising

[postbigbang]

There are a number of possible circumstances where a router might try to respond, then tie up resources. Example, if let's say that a routing, ARP, or layer-2 protocol arises inappropriately with distinct, yet smelly malformed garbage as a payload. Or a BGP dump sequence is done that confuses the router or blows its delicate cache tables. Spanning tree malformations can do it. Ooops. Better not say more. Isn't IOS invulnerable or something?

Re:Not surprising

[Malevolyn]

Let's hope the mods come to their senses. You must be new here.

Re:Not surprising

[thatskinnyguy]

Why in the hell was this modded troll?! What this man has described is exactly what happened to the first few machines I installed SP3 on. Granted it was a release candidate and I haven't had any problems with the go-live version. But still, troll? I think not.

Re:Not surprising

[thetoadwarrior]

While my router is working fine, my Windows laptop's shares fail to show up on my other machines and this happened straight after installing SP3 and all my settings imply nothing has changed. I just need to get another Windows machine in here to see if it's only being ignored by Linux machines.

Maybe it's a coincidence and maybe it's not. The only way to know for sure is if Microsoft honestly comments on it.

Re:Not surprising

[furbearntrout]

You must be new here.
The mods are on crack.
The mods are always on crack.
But there is something you can do about it.

Re:Not surprising

[peragrin]

I never use crack when i mod.

I much prefer weed and alcohol. far cheaper and it makes you spaced out enough that even the AC become funny.

Re:Not surprising

[hairyfeet]

Weird. Not having a problem here,but here are a few suggestions. Are you using the "simple file sharing" in XP or standard? Because I have seen simple file sharing get boned after an update. You can usually fix by unsharing the folders,rebooting and then resharing. If you aren't using the simple file sharing or are having trouble still,check the folder permissions. I have also seen updates bone the file and folder permissions. I'd say a good 80% on machines I've had to deal with resource sharing problems it would be one of those two. The rest usually come down to routing and firewall issues. Can you ping the machine? See it on the network? If so I would lean sharing and folder permissions issues.You also might want to try typing secpol.msc into the start/run dialog and then going to security options and scrolling down to LAN Manager Authentication level and making sure it says "Send LM & NTLM responses" and not "NTLMv2 responses only". Anyway I hope this helps and as always my 02c,YMMV

Re:Not surprising

[Paul Jakma]

The parent poster shouldn't be modded informative. Their post is a jumble of random network related terminology (several of which have 0 bearing on home routers) into information-less sentences. E.g. *BGP* or IOS on a home router? "Cache tables" (did the poster misremember hearing someone say "hash tables"?). The crowning glory though:

"Spanning tree malformations can do it".

The parent is either a wickedly funny troll, or an ignorant parrot. I just can't make up my mind..

Re:Not surprising

[RobertM1968]

Lets not jump to blame this on Windows. It could be that Windows isn't doing anything wrong, just something the router should be able to handle, but can't. We can point fingers when we know what the actual issue causing the router problems is.

Yes, lets not blame Windows XP SP3... after all, this has happened with other OS's and other versions of Windows, right?

Oh, wait, this has only happened with XP SP3 machines....

And other OS's (Linux, OS/2, eComStation, MacOSX) can handle more Internet traffic of more different types/protocols than Windows XP SP3 - so there technically shouldnt be anything "more" or "unhandlable" that an XP SP3 machine could possibly do that hasnt already been done on another OS without causing such problems.

So, again, what's the probability that it isn't Windows XP SP3 doing something wrong?

Re:Not surprising

[postbigbang]

You mean your tongue isn't in your cheek?

Re:Not surprising

[lyml]

Yes, lets not blame Windows XP SP3... after all, this has happened with other OS's and other versions of Windows, right?

Oh, wait, this has only happened with XP SP3 machines....

So, again, what's the probability that it isn't Windows XP SP3 doing something wrong?

I'd say 100% seeing as that it is the router that is crashing.

Think of it this way, if writing this post made your computer crash, would it be my computers fault, or would it be yours?

Re:Not surprising

[RobertM1968]

I'd say 100% seeing as that it is the router that is crashing.

Hmmm... what is the margin of error on that statistic?

Oh, nevermind... I came up with the same figure anyway. ;-)

Think of it this way, if writing this post made your computer crash, would it be my computers fault, or would it be yours?

Well, I'd like to blame it on your computer... and wonderfully, I can - and not be deemed insane. Things like that happen on /. all the time (like the guy I was responding to).

=)

Re:Not surprising

[deamonpainter33]

Isn't IOS invulnerable or something?

Ha yes it is invulnerable :) ...but just like anything, the router/switch is only as smart as the person who's programming it. Be it those at Cisco with there IOS revisions or the Network Tech that's setting it up.

Re:Not surprising

[RobertM1968]

See though, here's the thing... who do you blame?

In a way it is (caused by) SP3... (because) of something the router cannot handle.

So, it raises a few better questions than the ones being raised here (the blame game):

- (ROUTER'S FAULT) Why can't the router handle whatever type of traffic - and should it? At the very least, as a possible attack vector for routers, shouldn't it?

- (NOT NECESSARILY SP3's FAULT, BUT STILL AN ISSUE) Why is SP3 generating such traffic? What type of traffic is it generating? Could this traffic be considered (or detected elsewhere as) a DOS attack of some sort? (We do know that enough SYN packets will crash various routers - even high end ones). What is SP3 actually attempting to do (regardless of HOW, the more important questions are WHAT and WHY).

So, while the router may be at fault for the behavior due to the type of traffic, SP3 is at fault for generating traffic of a nature that is not needed (in any way I can think of) to utilize the Internet... and considering some of the new ad and update and spyware and DRM technologies that MS is trying to bring over to XP (see previous /. articles, various MS patents and more regarding their search plans, "Live" product plans and more)... is this traffic not just flawed, but totally unwanted and intrusive? Or is it simply a screw-up on MS's part that happened to indicate vulnerabilities in various routers?

See the thing is, the reasons MS has such code creating such traffic may be important (or simply a screw-up)... but regardless of that, it showed vulnerabilities in various routers... but regardless of that, it also showed some sort of traffic that SP3 generates that may also be the cause of other routers (that arent affected adversely by such traffic) detecting as an attack of some sort, causing all sorts of other issues (for instance, a subnet or port being shut down to block the traffic).

Think how wonderful that would be if it was at a large company, medical institution, school, EMS station, etc... where all their machines were on a NAT network, and one of them that got upgraded to SP3 suddenly got their single shared IP blocked from the Internet.

So, I think there may be plenty of blame to point at both MS and the router manufacturers...

But the sad thing is, (and I am loathe to say this on /. where I am expected to make judgements based off little or no facts), until enough facts come out (showing what type of traffic, why the traffic is being generated, and what unaffected routers do when they receive the traffic), the only blame so far is:
- MS for doing something (traffic wise) that no other device or OS manufacturer seems to have ever done before.
- The router manufacturers in question for having an implementation that is not robust enough to survive such traffic without crashing.

Re:Not surprising

[Maple Syrup]

Gotta be a troll. Here's the giveaway:

layer-2 protocol arises inappropriately with distinct, yet smelly malformed garbage as a payload

Re:Not surprising

[willyhill]

Darn, I already flashed my MotoPacket(TM) router with this information. I stab at the evil monkeys who can't get their spanning tree formations right!

Re:Not surprising

[nawcom]

Not everyone goes for a Cisco router when they are looking for a cheap reliable one for home. Hell, I've never seen one for sale at a BestBuy. So your bitch session that has to do with IOS is worthless. What are you gonna whine about now? How you still have your DEC alpha and we all have x86s so we should go fuck ourselves again?

Re:Not surprising

[davolfman]

In all seriousness though, I've had more than enough experience with consumer routers crashing. Personally I blame VXWorks as my one linux-based router is much more stable than any other I've ever had.

Re:Not surprising

[jibjibjib]

In fact, it's definitely the case that it's something the router should be able to handle. If a router receives a packet that causes it to crash, it's a flaw in the router's software, no matter whether the packet is malformed or not.

Re:Not surprising

[eneville]

> Well duh; it was just a lame attempt at a joke, not an actual claim
> that Windows was to blame. A properly-designed router will only reboot
> repeatedly if some client is logging in and causing it to reboot.

You have a good point here. Everyone will soon know what these routers
are and it might act as a bit of publicity for the company. Might be
negative in some ways, as a home networking product it could help
promote their brand name.

It's an interesting idea.

On the other hand, a network router should maintain uptime in excess of
years, it's not a difficult task to route packets after all, just a few
bit comparisons and a protocol stack. I work in an environment where
some gateways and proxies have uptimes in excess of 8 years! So you're
quite correct in the reboot principle.

Home NAT/routers are not enterprise equipment though, their intended
uptime is in the region of weeks and months, not years.

Re:Not surprising

[Geonagger]

I had a similar problem,pre SP3 installation though, on a Xp + Win 2000 networks - the network computers could be pinged and data shared via shortcuts etc but shared folders could not be not viewed. I ended up changing the network name from WORKGROUP to something else and it worked! Later I renamed the network back to WORKGROUP and it was fine. Regards, Geonagger

Re:Not surprising

[mpeskett]

My POS router (at home, not a "serious" piece of hardware) seems to need to be rebooted at regular intervals.

After x amount of time it just stops working properly... seems to either stop responding to, or stop carrying out DNS lookups, since stuff that's already connected (like IM or torrents) will carry on for a while, but websites fail to load, and the IM goes down too if I disconnect and try to reconnect.

Cycle the power and it comes straight back up, 'tis quite annoying.

Re:Not surprising

[eionmac]

1. It may be combination of router and MS software , but router just locked up and ceased functioning atthe 'prepare for download , when 100% prepare reached. Aborted download 3 times.
  Downloaded other matter, other XP pro upgrades, from MS after these aborts , so it is this specific download/router combination.
(I use Thomson router issued by ISP BT in UK)

Re:Not surprising

[yuna49]

This is a common affliction in home routers when people run torrents. The problem is that the torrent client establishes dozens of TCP/IP connections to peers. Eventually the router's NAT table runs out of memory, and traffic comes to a halt.

The solution is to limit the [i]number of connections[/i] (not bandwidth) so as not to overtax the router. Try an upper limit of 100 connections or so and see if the problem goes away. If not, bring the limit down until things work smoothly again.

Re:Not surprising

[Feanturi]

I've also had some sharing issues since SP3, but not general network problems. I can drag files from a local drive to a network drive just fine, but I cannot drag files from a network drive to anywhere, whether network or local. The drag operation does nothing if it's a left-click, and if it's a right-click drag the Move/Copy menu does not come up when releasing. I have to select the files and then hit Cut or Copy, and then I can Paste to the other drive. It's really annoying, and happens from both machines in my little home network. This started when I installed the SP3 release candidate much earlier, and did not get fixed when the actual SP3 got installed.

Re:Not surprising

[Anonymous Coward]

Lets not jump to blame this on Windows. It could be ... You must be new here.

Sorry, had to be said.

Re:Not surprising

[zzottt]

well said.

Re:Not surprising

[eneville]

> This is a common affliction in home routers when people run
> torrents. The problem is that the torrent client establishes
> dozens of TCP/IP connections to peers. Eventually the router's
> NAT table runs out of memory, and traffic comes to a halt.

> The solution is to limit the [i]number of connections[/i] (not
> bandwidth) so as not to overtax the router. Try an upper limit
> of 100 connections or so and see if the problem goes away. If
> not, bring the limit down until things work smoothly again.

That's interesting.

100 x ( 4+4+2+2 [socket data] ) * 2 = 2400bytes for 100 NAT
sockets

I would have hoped that there would be more memory dedicated for NAT
states.

Re:Not surprising

[thetoadwarrior]

I can ping the machine and my selpoc.msc settings are set properly.

I will have to try changing my workgroup and unsharing / sharing my folders again and see if that sorts it. Thanks for the tips!

Re:Not surprising

[hairyfeet]

Try unchecking inheritance on the folder permissions. Are you sharing a folder on the network drive,or the drive itself? I have seen problems with Windows file sharing when it comes to sharing drives. While it can be fixed with a lot of tweaking,time,and hassle,I find it easier to simply share a folder on the drive rather than the drive itself. Also due to inheritance the closer the folder is to root the better. You don't want it inheriting a bunch of permissions from the folders above it on the tree.

If you are still having trouble after that feel free to shoot me an email and I'll see if I can't walk you through some of the more complex fixes. I used to run a chat channel for Windows errors so I'm used to troubleshooting from a distance. And as always this is my 02c,YMMV and use the advice given at your own discretion.

Oh brother...

[Anonymous Coward]

Honestly... I'm sure that this has nothing to do with XP SP3 and is a coincidence, or just completely made up. And even if SP3 is doing something differently, if a router crashes due to a machine connected to it, it's a problem with the router.

Re:Oh brother...

[spazdor]

is a coincidence, or just completely made up

Unlikely, given that the OP mentions that at least one manufacturer has fixed the problem with a firmware update. You can't really write software to fix a problem until you've figured out what the problem is.

You're right though, a properly hardened router will keep ticking regardless of what's plugged into it. Mostly.

Before anyone goes on a MS rant

[Gregb05]

A computer on the network should not be able to crash the router. This is a problem with the manufacturing of the routers, not anything in particular with SP3. This problem could have arisen in any OS. The fact that it appeared with SP3 is irrelevant. I return you to your MS bashing.

Re:Before anyone goes on a MS rant

[RiotingPacifist]

True but it takes two to tango, just like a website should never crash a browser it is relevent that the OS/website is abusing the specifications soo badly to cause the crashes.
It is the manufactures fault that thier crashing, but this bug wouldnt be seen if xp was behaving correctly.

Re:Before anyone goes on a MS rant

[Wavebreak]

How do you know it's not? Until we see some actual analysis of this, there's no way to know if XP is at fault or if it's just shitty routers doing what they do. The latter seems far more likely to me, considering how absolutely shit most of the routers out there are, especially with stock firmware.

Re:Before anyone goes on a MS rant

[Jugalator]

True but it takes two to tango Yes, it wouldn't have happened without the "help" of SP3 in this case. That being said, with the relevant information not released here, it's not certain SP3 is doing anything inherently wrong according to the networking standards. Testing SP3 on all hardware configs is additionally nothing one can expect Microsoft of doing.

Re:Before anyone goes on a MS rant

[John Hasler]

> It is the manufactures fault that thier crashing, but this bug wouldnt be seen if xp was
> behaving correctly.

Nonsense. Any router that can be crashed by anything that a computer connected to it does has a critical bug and should be recalled immediately.

Re:Before anyone goes on a MS rant

[jellomizer]

However... If they find out what causes the router to crash with SP3 then all it will take is someone to duplicate the information sent and crash the router again and again. If the router crashes is has to be the fault of the router not of the OS, as other routers don't crash. As well as a poorly designed website. If your web browser crashes from a badly made website then it is the web browsers faults. Your argument only really holds true in cases of custom designed software where the sender of the data will need to agree to send the data in the correct format as well the receiver agrees to get the data in the correct format. And still even in that case a good program will be able to atleast say something is wrong, vs. it crashing.

Re:Before anyone goes on a MS rant

[Kjella]

Between having a security hole that allows denial-of-service attacks and sending some slightly mangled packets, I'd go about 98-2% to the router manufacturers on this one. That is assuming the packets are actually mangled, which isn't proven. I wouldn't care how broken a web page was, if Firefox crashes then it's Firefox's fault (or extensions/plugins, but that's a different issue). If you came and said "While it's true that the browser shouldn't crash in any circumstances, Apache would certainly deserve bashing if it's a result of them violating the specs and sending out mangled web pages." I think people would laugh. Assume the input data is crap, that applies equally everywhere and any software that can't handle that is poor software.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

Lets make this easier and flat out assume XP's UPnP implementation was intentionally designed to crash the largest number of routers on the planet in a clever bid to raise Vista sales.

Even with this unlikely assumption in play it would still be 100% the fault of the router for crashing.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

Are you fucking stupid or something ??

if a router crashes , its the ROUTERS fault.

Doesn't matter whats on the network. No need to analyze anything.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

Wrong.

When the network stack on a particular operating system changes the algorithms used for flow control and backoff, it can quite easily break hardware which was designed for a different load, for example, not designed to handle sliding window sizes.

It's always a tradeoff for changes like this; better performance for new equipment using the new design, some broken old equipment; both of which are good for selling the latest new thing, so it's not going to invoke much guilt in the manufacturers when they make changes and break old stuff. Installed base is a liability; if what you have works too well, there is no reason for you to buy vN+1. Since computer technology has basically been "good enough" for basic tasks like web, office documents, and email for the past 8 years, the old drive to make each revision of the products better with each release is pretty much dead. Think about it; with the exception of more advanced games, does your computer do *anything* it could not do eight years ago? Nope. It might do some things a little faster, do some things a little prettier, but there really aren't any new killer apps or breakthroughs.

Oh, and commercial computer equipment/software is *always* duct tape and bubblegum. It's just a matter of how many coats of paint the manufacturer(s) bothered to slap on to hide the realities of the technology business from you. It's about selling stuff now, not pushing the envelope of technology.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

So has no one heard of DoS attacks? It's basically the same thing here. If Windows XP is creating too many concurrent connections (or even going awry) it could shut down the router. That's what makes the difference between a SOHO router and an enterprise router. The more processing power the router has, the more it's able to "keep track" of all of the packets. The SOHO router will just give up and shut down.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

We all agree that computers should not be able to crash routers. However, SP3 includes their Next Gen TCP/IP stack that "complies" with several RFCs noted here:
http://technet2.microsoft.com/windowsserver2008/en/library/60f9e0c6-dfb3-4ead-aa12-3ba7653664fd1033.mspx?mfr=true
Of course most routers are not protocol-aware of these, and due to the unexpected use of such protocols, the effects on routers could manifest the reboots that we see. So SP3 IS relevant since it is the ONLY thing that has changed. What this implies to me is that, again, someone dropped the ball in the testing department at either MS or the router(s) companies. These things are going to happen whether its MS, Linux, MAC, Solaris, etc. if updates are deployed without fully testing.

Re:Before anyone goes on a MS rant

[BitterOak]

Nonsense. Any router that can be crashed by anything that a computer connected to it does has a critical bug and should be recalled immediately. Really? So if I have an especially nasty computer that sends a 100kV "signal" down the twisted pair cable to the router, it's the router's fault if it gets fried?

Re:Before anyone goes on a MS rant

[Anonymous Coward]

One of the first routers I had would occasionally get a packed of death. I eventually traced it to folks hitting my home webserver with a correct, but odd, http request.

Crap home routers.

Re:Before anyone goes on a MS rant

[Sleepy]

>just like a website should never crash a browser it is relevent that the OS/website is abusing the specifications soo badly to cause the crashes.

All browsers MUST expect garbage input, and pretty much anyone who expects otherwise has their head in the sand, and shouldn't go NEAR code. Document formats don't run in local memory and do not have system access - they're interpreted structures. If a browser crashes, it's the browser's (or it's plugins) fault, 100% of the blame 100% of the time. At worst, you should expect degraded-looking content and that's that.

I don't mean to be snarky, but your argument couldn't be more wrong or inappropriate. With networking, you're pretty close to the physical layer (not a great analogy, but browser code is far removed from traffic and is just a local representation, a user application).

We don't know yet what this is caused by. If it is affecting a lot of routers, it might very well be a "DOS". Or it could be something that holds too many connections open, or IP6 traffic that doesn't go anywhere and ties up the router table till it times out.

This could happen to Linux also, but it's less probable -- it's be code put out in the wild, and the distro's would do their own QA process, and may hold back. Most distros don't run kernel.org kernels, but their own patched tree.

Re:Before anyone goes on a MS rant

[tshawkins]

Really? Do you know it's the router's fault and not XP SP3's fault? Sure, you can say "an OS should never crash a router" but that's simply not true. If the OS abuses the router in ways that it simply can't handle and shouldn't have to, the OS is at fault. Errrh, a router should NEVER crash, you should be able to stream absolute crap at it for years on end without it crashing. part of a routers role is to reject malformed packets and request resends, not go tits up at the first sign of trouble. Christ if routers behaved like this the whole net would be down most of the time, do you realised how much damaged data flows around the net.

Re:Before anyone goes on a MS rant

[postbigbang]

Everyone wants a bulletproof, but quickly reacting app. Sometimes you can't have both. You can build parsers that vet web pages for sanity sake (or just look for malware as some plugins do), but they'll slow down even the fastest clocking machines; the insertion loss of the parser will be like putting a foot on the garden hose.

Routers and layer 2/3 bridges have to react at wire speed, and therefore have lean, racing engine code with only the barest of exception handlers. Inside the code are lots of routines that have to react to protocol changes related to table building. Screw up those tables even legally (according to the obscurities of even well-known protocols) and the routing/bridging device will behave badly, even to the point of apparently not working. It's happened before, and will happen again. Is it XP3? No one knows yet.

The next update of will likely fix the problem; likely it would arrive before a Microsoft fix, and it would be more effective to fix the crashing device than go after all possible XP SP3 users. Sadly, once in the 'wild', it's the router vendor's problem rather than Microsoft's, no matter who is to blame for the original mistake.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

> So if I have an especially nasty computer that sends a 100kV "signal" down the twisted pair cable to the router, it's the router's fault if it gets fried?

It's certainly not the protocol layer at fault. Last I looked, Microsoft did not manufacture networking hardware.

Re:Before anyone goes on a MS rant

[statusbar]

No, not when you consider that Microsoft was the primary designer of the fundamentally flawed UPnP specification...

--jeffk++

Re:Before anyone goes on a MS rant

[Ravon Rodriguez]

Routers and layer 2/3 bridges have to react at wire speed, and therefore have lean, racing engine code with only the barest of exception handlers. Inside the code are lots of routines that have to react to protocol changes related to table building. Screw up those tables even legally (according to the obscurities of even well-known protocols) and the routing/bridging device will behave badly, even to the point of apparently not working. It's happened before, and will happen again.

This may be true, but there are routers out there that manage not to crash and don't have any reaction delays. I would bet that this is due to routers not having the complete protocol rules in their firmware, however obscure said rules may be. If you buy a fifteen dollar router, you get a fifteen dollar router.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

"Any router that can be crashed by anything that a computer connected to it..."

This describes all routers.

I can connect just about any computer to any router and crash the router.

Also, if XP SP3 installed = router crash AND uninstall XP SP3 = router not crash, the problem is XP SP3, not the routers.

Are you trying to say that if suddenly, without warning to auto makers, I decide to add dirt and sugar to the most widely used formula of gasoline now, that cars breaking down as a result is the auto manufacturer's fault and not the idiot that added dirt and sugar to gas without telling manufacturers?

Re:Before anyone goes on a MS rant

[Tony Hoyle]

Exactly. If these routers can crash being sent normal packets - even if they get a lot of them - then they're vulnerable to a DOS attack and should be recalled until they're fixed.

Re:Before anyone goes on a MS rant

[Ravon Rodriguez]

We all agree that computers should not be able to crash routers. However, SP3 includes their Next Gen TCP/IP stack that "complies" with several RFCs noted here:
http://technet2.microsoft.com/windowsserver2008/en/library/60f9e0c6-dfb3-4ead-aa12-3ba7653664fd1033.mspx?mfr=true Of course most routers are not protocol-aware of these, and due to the unexpected use of such protocols, the effects on routers could manifest the reboots that we see. So SP3 IS relevant since it is the ONLY thing that has changed. What this implies to me is that, again, someone dropped the ball in the testing department at either MS or the router(s) companies. These things are going to happen whether its MS, Linux, MAC, Solaris, etc. if updates are deployed without fully testing.

If this is indeed the case, then I'd go so far as to say that its not anyones fault. While I agree that the router shouldn't crash, router manufacturers can't be held liable for not following specifications that weren't available when the router was manufactured, and Microsoft shouldn't be held liable for using said specifications.

Re:Before anyone goes on a MS rant

[Fieryphoenix]

packed of death.

I think you meant packet. Or /. Or something.

Re:Before anyone goes on a MS rant

[Fweeky]

Apache isn't responsible for the well formedness of web pages, it's responsible for HTTP. If Apache's sending out mangled HTTP headers which causes a browser to crash, sure, the browser vendor is at fault for not failing gracefully or recovering, but Apache sure as hell should take a lot of blame for not catching protocol violations during QA. Potential security issues are very serious, yes, but so are potential interoperability problems and Microsoft are the *last* people who should be let off lightly for those.

Ultimately both are indicative of poor development practices. Sure, the router developers are completely responsible for a DoS hole, but Microsoft are equally culpable for failing to follow spec, iff that is indeed the case.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

>> A computer on the network should not be able to crash the router

This is not true. In fact, without first knowing the root cause, the suggestion of the router vendor(s) being at fault is as flawed as blaming Microsoft. Most routers, whether domestic or otherwise, also act as an end-point for various protocols... e.g. uPnP, ssh, SMTP, etc. Just like any other end-point, they can be vulnerable to defects in the logic of any protocol that they implement. A prime example of broken logic in a protocol is the case of TCP SYN flooding vulnerability.

The router *could* be well programed with respect to faithfully following the relevant protocol spec, and yet still be vulnerable to entering an unforeseen state.

So, before laying blame, perhaps we should take the time to discover the root cause.

Re:Before anyone goes on a MS rant

[toddestan]

That's obviously not the case here, as how could Microsoft get your computer to send down 100kV down the ethernet port with just a service pack?

Re:Before anyone goes on a MS rant

[ultranova]

Nonsense. Any router that can be crashed by anything that a computer connected to it does has a critical bug and should be recalled immediately.

Most home routers are configured through a web interface. That web interface has a reboot button. In other words, a home router can be made to reboot by sending it a proper HTTP request. This is not a bug, it's a feature :). It is also likely indistinguishable from a crash-and-reboot, since the reboot clears the crashed state. For all we know, Windows could be flooding the router with reboot commands...

Re:Before anyone goes on a MS rant

[hairyfeet]

While I agree with you to some extent,I think we also need to look at the bigger picture. If you RTFA(I know..what the hell is wrong with me,reading TFA like that) you'll see that this isn't some Cisco or Sonicwall piece of quality gear,but one of those ubercheap Chinese junk routers. If you will look here you will see we are talking about a $32 USD router. Now we all know when selling gear for that cheap that corners are going to be cut,otherwise it just wouldn't be profitable. And we have no idea what kind of corners were cut in its packet handling code. So I think I'll save the XP SP3 condemnation until I see more than a single $32 router model crash. But that is my 02c,YMMV

Re:Before anyone goes on a MS rant

[marcansoft]

Interestingly, we just released a new version of The Homebrew Channel for the Wii (beta8), and someone reported that it causes their router to crash. We've switched to an older version of the Wii system firmware (from IOS35 to IOS21 - no, not the Cisco kind of IOS) in older to support users without the newest updates (this applies to all users though, since all versions / branches of the firmware are kept on the Wii), and I have a feeling that some kind of bug in the WiFi networking in this version is causing it. Wonder if it has anything to do with the Windows problems....

Re:Before anyone goes on a MS rant

[John Hasler]

> Most home routers are configured through a web interface. That web interface has a reboot
> button. In other words, a home router can be made to reboot by sending it a proper HTTP
> request. This is not a bug, it's a feature :).

If the router can be reconfigured and/or rebooted without a password it is buggy.

Re:Before anyone goes on a MS rant

[thegrassyknowl]

custom designed software where the sender of the data will need to agree to send the data in the correct format as well the receiver agrees to get the data in the correct format. I thought that was the point of standards? The router conforms to acceptable network standards. The OS is meant to conform as well. MS releases a new patch, it makes the OS start sending packets outside the standard. The router gets confused and crashes. Sure, the router needs to be more robust, but I'd say that is Microsoft's fault for sending invalid packets.

I have a fairly large network here with a lot of varied machines on it and my router never crashes. Add one SP3 box and bam, down it goes. I think it's fairly safe to say that MS have done something dodgy in SP3.

Re:Before anyone goes on a MS rant

[ConceptJunkie]

If the router can be reconfigured and/or rebooted without a password it is buggy.

I can do that by bouncing the power. In fact, my router has a power switch.

Re:Before anyone goes on a MS rant

[GigaplexNZ]

Except to figure out why it is crashing so the problem can be fixed...

Re:Before anyone goes on a MS rant

[Slashdot Suxxors]

On all the home routers I've used you still need a password and username to access any of the administrative controls of the router. It's not like anyone can send a packet to 192.168.1.1 and reboot your router.

Re:Before anyone goes on a MS rant

[nawcom]

heh, let's remember that the next time some major DoS is in the news. It's obviously not the source of the DoS attack's fault, its the server.

"Doesn't matter what[']s on the network. No need to analyze anything." You sir, are stupid. If EVERY OTHER NOS SENDS LEGIT PACKETS THROUGH THE ROUTER BUT XP SP3 IT DOESN'T MEAN THAT ROUTER SHOULD BE FIXED TO ALLOW IT. It's called standards; MS is known to have pouting fits about following them.

if a router crashes , its the ROUTERS fault. We'll see if you change your mind when I unload my packets onto your router #2 sloppy style.

Re:Before anyone goes on a MS rant

[Anonymous Coward]

Apple Macs also cause a similar issue on BT/Thomson speedtouch routers with iPlayer in quite a few cases.

Once again, it is the router manufacturer's fault.

Re:Before anyone goes on a MS rant

[ultranova]

If the router can be reconfigured and/or rebooted without a password it is buggy.

Username and password are usually "username" and "password", so it's not like this is much of a peed bumb.

Re:Before anyone goes on a MS rant

[RaceProUK]

SP3 isn't the only software capable of crashing a router: the PS3's OS crashes my router regularly.

Re:Before anyone goes on a MS rant

[goldn_64]

http requests cannot be "odd" to a router. http is an application layer protocol, routers only go up to the network layer.

Re:Before anyone goes on a MS rant

[prisoner-of-enigma]

I thought that was the point of standards? The router conforms to acceptable network standards. The OS is meant to conform as well. MS releases a new patch, it makes the OS start sending packets outside the standard. Perhaps you should stop and consider that it's the router that's unable to handle an obscure -- yet standards-compliant -- packet being sent by XP SP3? For evidence, I would put forth that no other router we yet know of has any problems whatsoever with SP3. Further, I'd posit that any router that can be crashed via a packet from any wired or wireless client is defective.

You have no proof at all that SP3 is sending invalid packets. You have merely assumed that, while giving the router manufacturer a complete pass on its own "dodgy" firmware. After all, if it was a Windows-based router that was crashing after you upgraded to Fedora Core 9, would you blame Fedora? Or would you instead gripe that Microsoft made a faulty router? I don't have to guess to hard to figure out which way /. would write that article. Hint: Fedora wouldn't get the blame.

Perhaps you should put your bias down next time and analyze the situation objectively before rendering a conclusion that is so obviously flawed.

Re:Before anyone goes on a MS rant

[yuna49]

Thank you for having read TFA and reminding people here we're talking about one particular model of router. The Slashdot headline about "router crashes" would lead you to think the problem is widespread and occurring across a variety of brands. As far as I can tell, it is only this router that's having problems.

Personally, I'd something have to do with the implementation of uPNP is at fault, but that's just a wild guess.

Re:Before anyone goes on a MS rant

[hairyfeet]

The router could also be choking on the packet stream,as I do a lot of work on the consumer side and have seen it happen. What happens is you try to transfer a large file on the LAN side and the thing starts mangling the packets before finally choking and croaking. And while I haven't worked on this model I have worked with plenty of under $50 routers. The FA really doesn't tell much,such as whether it happens while using the WAN or LAN and seems more like a sensational headline designed to grab page views.

Speaking of cheap routers,if anyone needs an uber cheap router either to have a spare or just to set up a network on the cheap,may I suggest this one. I picked this one up a couple of weeks ago and surprisingly it purrs like a kitten. Doesn't choke on file transfers, has a nice interface for router adjustments, and stays very cool to the touch. Just don't expect it to look like the picture,as apparently that was the old model. The model you actually get is a tiny square that you can fit in your pocket and lays flat,as opposed to on its side in the picture. And with an after rebate price of $12 you really can't beat it. Hell,I couldn't even find a hub for that price! I figured being that cheap I'd have to use my Win2K pro box as a gateway and let my boys strictly use this for their LAN games,but no matter what I throw at it she just purrs like a kitten. But anyway that is my 02c,YMMV

Re:Before anyone goes on a MS rant

[fbjon]

Really? So if I have an especially nasty computer that sends a 100kV "signal" down the twisted pair cable to the router, it's the router's fault if it gets fried? The Etherkiller qualifies as neither NIC nor computer, FYI.

Re:Before anyone goes on a MS rant

[FictionPimp]

Its actually both the fault of the router makers and windows. Microsoft is at fault for 'changing' the standard, and the router makers are at fault for having a bug in their software that causes the router to crash when it gets something like that. BOTH of them should fix it.

Re:Before anyone goes on a MS rant

[jellomizer]

The point of standards is so we all have an agreed way of dealing with the information. However it is the implementers fault if it incorrectly deals with non-standards. Computer Science Intro classes. While Garbage in = Garbage out you need to make sure your program can handle a monkey sending in inputs. So even if SP3 is broken TCP/IP and all. It is still all the blame on the router for crashing, vs. Rejecting the bad packets or what ever default action is.

Re:Before anyone goes on a MS rant

[halcyon1234]

it takes two to tango

And only one to charlie foxtrot

Blaming the wrong programmers

[Jeffrey Baker]

Shouldn't the title of this post be "Shitty router programming causing router crashes"? It should matter what type of garbage come off the wire, the router must be able to handle it all without error.

Re:Blaming the wrong programmers

[Jugalator]

Agreed. If SP3 can do this unintentionally, imagine what the series of communicated data with the routers can do if a malicious writer now reverse-engineer whatever SP3 is doing, and would spread a time-triggered virus, for example. These kind of hardware issues are never good.

Re:Blaming the wrong programmers

[alx5000]

Otherwise, imagine the havoc an external attacker could create just by imitating this supposedly XP SP3's new behaviour (unless it only affects the internal LAN ports somehow and, still, a router should know better than letting others take it down...). Besides, were it to be XP's fault, I wonder how secure any network using this routers could be, since a random virus infection could render all its neighbours connectionless, just by imitating Windows's "strategy".

Re:Blaming the wrong programmers

[thermian]

Maybe, SP3 appears to have been responsible for screwing up my machine, causing it to crash constantly within a day of being installed.
I re-installed and just got the updates singly from windows update. It might just be that my machine is four years old however, I'm not definitely sure it was SP3, it was just the timing that makes it seem that way.

Re:Blaming the wrong programmers

[nurb432]

Or even " Windows XPSP3 exposes errors in routers, manufacturers scramble to fix possible DoS vulnerability'

i hate windows as much as the next guy but i agree, its not really Microsoft's fault.

Re:Blaming the wrong programmers

[yomegaman]

The "article" is just a reprinted press release from Billion. Of course they blame SP3, since the alternative is admitting their products are buggy pieces of junk.

Re:Blaming the wrong programmers

[KUHurdler]

imagine what the series of communicated data with the routers can do if a malicious writer now reverse-engineer whatever SP3 is doing, and would spread a time-triggered virus

you're right, then all the routers would reboot at once... and the world would be at a standstill for an entire 30 seconds.

Re:Blaming the wrong programmers

[Ravon Rodriguez]

Obviously not all routers are affected, and I'm inclined to believe that its a small percentage of them that are. Does anyone have any data to confirm or deny?

Re:Blaming the wrong programmers

[Jugalator]

LOL. But seriously speaking, the thing is it could happen again and again until those computers were cleaned. ;) And many XP computers don't even run antivirus.

Re:Blaming the wrong programmers

[GumphMaster]

If my Billion box is anything to go by then this is right on the money. Too many connections on a flat out BitTorrent session and *boom* the ADSL bridge hangs. No warning, no logged errors, no response except to the serial interface, no recourse except to power-off and throttle BT.

Other Glitches?

[Illbay]

This seems like a good place to mention this.

I have had some screwed-up problems since installing SP3 myself, not related to this "Router" issue.

First, Firefox no longer functions correctly. Specifically, field data that is retained (you can see it right there already filled-in), does not register. For instance, if I have a user id and password already entered and saved from a previous session, even though it shows up in the field, the server system doesn't "see" it.

Second, the USB ports on my HP Port Replicator xb2000 (I believe) no longer function.

I don't believe this is coincidence, but is directly attributable to SP3.

Re:Other Glitches?

[oz_paulb]

I have to agree.

I updated to SP3 yesterday, and now my microwave stopped working.

Coincidence? I think not!

Re:Other Glitches?

[couchslug]

"Second, the USB ports on my HP Port Replicator xb2000 (I believe) no longer function."

Boot a live Linux CD such as Knoppix and see if it works. It's a handy way to swap OS for testing.

Re:Other Glitches?

[Enderandrew]

For what it is worth, I have no problems with USB or field data with SP3.

What version of Firefox are you running?

Re:Other Glitches?

[Anonymous Coward]

Congrats on making the only comment so far that isn't totally idiotic.

Re:Other Glitches?

[punissuer]

Funny, but IMHO Windows' support for USB still sucks. Sometimes my XP Home machine boots up and totally ignores my USB keyboard. Unplugging the kb and plugging it back in fixes the issue temporarily, but why should I have to do that? The motherboard setup program has no such issues, and neither do any Linux distros I've tried.

Re:Other Glitches?

[Hankapobe]

I have to agree. I updated to SP3 yesterday, and now my microwave stopped working. Coincidence? I think not!

And I have heart burn....Hmmmm, I think you've discovered something here!

Re:Other Glitches?

[Linker3000]

I sense a Farkism here:

"Oil prices reach $138/barrel on news that.." ...Shakes Magic 8 ball...

"...Stock Exchange installs Windows XP SP3"

Re:Other Glitches?

[Krilomir]

Yeah, let us blame Microsoft for all the crappy windows drivers out there instead of the hardware manufacturers. The fact that it works in Linux is just another indication that it is the drivers that are the issue.

Re:Other Glitches?

[Anonymous Coward]

I updated to SP3 yesterday, and now my microwave stopped working.

What's wrong with you!? Don't you remember that update that made Sony batteries to explode? There is no such thing as terrorism anymore, it's just Windows Updates!

Re:Other Glitches?

[green_flyboy]

It's called a feature not a bug. I do believe this feature is part of Microsoft's new security policy.

Re:Other Glitches?

[Anonymous Coward]

Just tried updating my toaster to XP3, but all I got was Blue Smoke of Death. Didn't even have time to fry the Router. Next toaster is going to be NetBSD again.

Re:Other Glitches?

[RulerOf]

Yeah, let us blame Microsoft for all the crappy windows drivers out there instead of the hardware manufacturers.
I'm going to go a step further and blame the shitty motherboard.

Re:Other Glitches?

[Kjella]

I updated to SP3 yesterday, and now my microwave stopped working. Well doh, your microwave (and your toaster) runs Linux so you can't upgrade them to Windows XP SP3, it probably died from shame.

Re:Other Glitches?

[SpaceLifeForm]

Yeah! No more botnets and spam.

Why didn't Microsoft think of this long ago?

Re:Other Glitches?

[6Yankee]

"My CD burner's USB, you insensitive clod!"

Re:Other Glitches?

[punissuer]

Yeah, let us blame Microsoft for all the crappy windows drivers out there instead of the hardware manufacturers. The fact that it works in Linux is just another indication that it is the drivers that are the issue. I agree, it probably is the drivers. But this is a USB keyboard, a generic Human Interface Device, and the Device Manager assures me that every driver file for every device in my system that has "USB" in its name was written by...drum roll please...Microsoft.

What if I was the other way around?

[ccguy]

If an upgrade to a router caused Windows to enter a reboot cycle would we be blaming the router manufacturer or Microsoft?

Re:What if I was the other way around?

[vux984]

If an upgrade to a router caused Windows to enter a reboot cycle would we be blaming the router manufacturer or Microsoft?

Would anyone notice?

Kidding aside, my first thought was this is CLEARLY a router problem. Even if SP3 is completely defective and sending out complete garbage to the router, the router should cope better than going into a 'crash and reboot cycle'.

Re:What if I was the other way around?

[ccguy]

Kidding aside, my first thought was this is CLEARLY a router problem.

Obviously. Anyway they'd better fix it soon because a customer with no internet connection and no way to fix the problem online is a pissed off customer.

Re:What if I was the other way around?

[Opportunist]

Problem is, though, he would (as usually) be pissed at the wrong party. That MS is the wrong party this time might be divine justice, but it still means that the wrong side gets heat for something someone completely different fubar'ed.

Re:What if I was the other way around?

[Jesus_666]

Both. Windows sends out funky data; that's not optimal and should be fixed (unless it turns out that Microsoft uses an obscure but valid part of some standard). The router can be crashed by sending it funky data; that's a major screwup and shouldn't have happened at all.

Microsoft (maybe) gets a slap on the wrist for not paying attention; the router manufacturers get a kick in the balls for producing junk.

Re:What if I was the other way around?

[Anonymous Coward]

I'd probably blame the router manufacturer and the company that made my network card since it was probably them that wrote the driver.

Re:What if I was the other way around?

[Frankie70]

If an upgrade to a router caused Windows to enter a reboot cycle would we be blaming the router manufacturer or Microsoft?

We always blame Microsoft.

Re:What if I was the other way around?

[Anonymous Coward]

yes

Re:What if I was the other way around?

[Anonymous Coward]

Microsoft, obviously. If your network device crashes because it doesn't like something on the network, you screwed up.

No lies, just a bad track record.

[hikaricloud]

As for the service pack not crashing routers, they actually do have the ability. So no MS bashing for me, just truth. Microsoft just has a bad run with service packs for XP, huh? SP3 has also been the culprit for a lot of machines just up and crashing. At both of our shops, we've gotten scads of machines, all with the same issues, all caused by an SP3 update. It's insane. First ME, then XP SP2, then Vista, now SP3. Microsoft really wants to be the evil empire, don't they?

Re:No lies, just a bad track record.

[hikaricloud]

Hey damn. I forgot the br codes. Great.

Re:No lies, just a bad track record.

[Anonymous Coward]

At both of our shops, we've gotten scads of machines, all with the same issues, all caused by an SP3 update. It's insane. God damn. If you're running a shop with "scads of" computers, try TESTING the damned service pack before rolling it out sitewide. You deserve the hassle for such a stupid oversight.

Re:No lies, just a bad track record.

[biolysis]

"Apparently sarcasm and humor is flamebait nowadays."

It is when there's no sarcasm or humor. Like in your post.

You came off as just another anti-MS troll with no idea what they're talking about. If it was intended to be sarcastic or humorous, you missed badly on both.

Re:No lies, just a bad track record.

[Anonymous Coward]

Like people might say when discussing Linux and Mac updates, are you sure it's SP3 not some strange third party program that's causing issues?

Thats funny

[kcbanner]

My OpenBSD router is fine...oh wait, I don't run Windows either.

"Billion" routers?

[Anonymous Coward]

Uhhh... never heard of them.

Tinfoil hat time!

[larry bagina]

We all know that most routers use Linux, usually an old kernel that won't/can't be upgraded. I'm wondering if they are intentionally exploiting a bug in the linux TCP/IP stack to cause crashes/instability/etc. With MS trying to get in on the phone/embedded device market, they could license NT/IIS for router usage and make some serious cash.

Re:Tinfoil hat time!

[Ethanol-fueled]

No, it's just the Linux overlords making their stuff not work--on purpose--with Microsoft products. It's a conspiracy to propagate desktop linux since 99% of broadband customers access the net through some kind of cheap router.

...And you all think that Microsoft is anti-competitive? :)

Re:Tinfoil hat time!

[Anonymous Coward]

congratulations, that's the dumbest thing i've read today.

Re:Tinfoil hat time!

[John Hasler]

The problem is much more likely to be in code added by the manufacturer, either in their custom device drivers or their application code.

Re:Tinfoil hat time!

[SanityInAnarchy]

We all know that most routers use Linux Do we? I'd guess a fair number don't.

Re:Tinfoil hat time!

[anotherone]

Even if that were true (it isn't) it would STILL be the router's fault.

Re:Tinfoil hat time!

[spazdor]

Mods, please note the "Tinfoil Hat Time!" in the parent's title. This is clearly intended as tongue-in-cheek, rather than trolling.

So...

[laurent420]

Windows is a DOS?

Re:So...

[Enderandrew]

Windows 3.1 developed an ice cream flavor called HogginDOS.

Re:So...

[marquis111]

Well, yeah. Windows can be classified as a DOS, or Denial Of Service, in many different ways. Never thought about it that way, but I guess it is. Good catch.

Re:So...

[thegrassyknowl]

It's all a DOS even if they try and hide the DOS with that fancy bitmap screen at startup.

*tongue out of cheek* I actually had a customer return a computer when I was working at a computer store years ago. She was insistent she paid for Windows95 and none of this DOS crap. Any text at all on the console (BIOS, "Starting Windows", etc) was interpreted by the customer as DOS and she wanted a refund.

Didn't you wonder why they all have those pretty manufacturer logos full screen now until the OS removes them and puts up its own screen?

And many users actually ask why I run DOS on my brand new quad-core beast. They are bewildered that someone would use "DOS"... something about the high resolution text console scares them.

Happened to me...

[flar2]

This happened to me. I booted into Windows XP for the first time in months just to check out SP3 and that same night my router went crazy, lights blinking on and off. It's a cheap no-name router, I'll have to find out what chipset it is and whether I can upgrade the firmware just in case I ever boot into Windows again.

Re:Happened to me...

[unclenate]

Not only have I see this happen with an Xbox 360, but I used the following methodology to stop the router crash... http://forums.xbox-scene.com/lofiversion/index.php/t502544.html Unfortunatly it required making the directly wired unit a DMZ host. I wonder if anyone has tried this with an affected PC/Win unit.

Well that explains everything.

[Anonymous Coward]

PC gaming has been a real drag for the last few weeks with the constant crashing of my router... I wonder if there's anything I can do about it.

Re:Well that explains everything.

[Darkness404]

Ummmm... Switch to Linux/Mac/BSD/Plan 9/Solaris/Hurd/Syllable/FreeDOS/Haiku/Windows 95?

Crashes Netgear MR814v3

[ROMRIX]

I've been wondering what the hell has been going on with my conne

Re:Crashes Netgear MR814v3

[Anonymous Coward]

I've been wondering what the hell has been going on with my conne Those routers have crashed and been general POS's long before SP3...

DOS attack...

[Anonymous Coward]

So billion routers are prone to a DOS attack. Misbehaving nodes on the network should not be able to take the router down.

To bad the article doesn't mention what the culprit is. Is it malformed DHCP requests? IPv6 trafick, UPNP.... Why is this article on slashdot?

Works for me, and probably for you

[Ethan Allison]

It only affects the "Billion BiPAC 5200" series.

I've never used one, never seen one, never heard of one, and you haven't either. Odd how the summary fails to mention that the problem is only with this obscure model...

Re:Works for me, and probably for you

[Anonymous Coward]

In my experience this also affects every Linksys model I had come in contact with that are flashed with the stock firmware or DD-WRT (any version).

Re:Works for me, and probably for you

[oldspewey]

DD-WRT (any version) The dd-wrt forums have nothing on XP SP3 problems ... have you reported the issue?

Re:Works for me, and probably for you

[Anonymous Coward]

I call bullshit on this one, dd-wrt is basically a small linux distro tailored to run on broadcom chipset routers. If there was any way XP SP3 could crash linux, it would be all over lkml and dd-wrt fora by now. So, again, utter and complete tripe.

Re:Works for me, and probably for you

[FJR1300 Rider]

I have a Linksys WRT54GL running DD-WRT v24-rc7

DD-WRT v24 std (c) 2008 NewMedia-NET GmbH
Release: 04/24/08 (SVN revision: 9433)

and I have 2 XP SP3 machines on my network, both a laptop and a desktop, and it all runs smoothly.

root@router:~# uptime
21:48:31 up 15 days, 13:46, load average: 0.00, 0.00, 0.00

Re:Works for me, and probably for you

[ColdWetDog]

and I have 2 XP SP3 machines on my network, both a laptop and a desktop, and it all runs smoothly.

-- snip --

load average: 0.00, 0.00, 0.00

Of course it's running smoothly, you're not doing anything!

Re:Works for me, and probably for you

[Ravon Rodriguez]

Ditto, I have a WRT54G running factory firmware (v1.00.9 May 12, 2006), and I haven't noticed any issues.

Re:Works for me, and probably for you

[Colonel Korn]

I have a Linksys WRT54GL running DD-WRT v24-rc7

DD-WRT v24 std (c) 2008 NewMedia-NET GmbH

Release: 04/24/08 (SVN revision: 9433)



Same here, with 2 XP SP3s and a Vista 64 SP1, zero problems with WRT54GL running DD-WRT. I think the GP was grasping for straws.

and I have 2 XP SP3 machines on my network, both a laptop and a desktop, and it all runs smoothly.


root@router:~# uptime

  21:48:31 up 15 days, 13:46, load average: 0.00, 0.00, 0.00

Re:Works for me, and probably for you

[cibyr]

I have a Billion 7404-VGPM. They're a pretty popular brand here in Australia (at least among tech-savvy customers) as they offer a lot of high-end features (VoIP, QoS, etc) for budget prices. It's actually cheaper in most cases to pick up a Billion than to grab a Linksys and run DD-WRT on it (which was my last setup).

Re:Works for me, and probably for you

[Anonymous Coward]

In Australia, Billion is one of the most common brands sold...

Why do US people think we always copy them in preference.

Oblig comment

[Anonymous Coward]

You haven't been on slashdot for long?

Re:Oblig comment

[Anonymous Coward]

Correction:
You must be new here.

Re:Oblig comment

[Anonymous Coward]

What do you mean? He's been shilling here since Microsoft hired him.

Re:Oblig comment

[Anonymous Coward]

Long enough to know to post anonymously if I am going to say anything (no matter if its true or not) in defense of Microsoft.

Same as Vista

[Enderandrew]

SP3 borrows a Vista feature (presumably the same code) to detect "Router Black Holes".

From http://www.winsupersite.com/faq/xp_sp3.asp

"Black hole" router detection algorithm. XP gains the ability to ignore network routers that incorrectly drop certain kinds of network packets. This, too, is a feature of Windows Vista.

Re:Same as Vista

[Anonymous Coward]

SP3 borrows a Vista feature (presumably the same code) to detect "Router Black Holes".

That has been around for years, it's called path MTU discovery.

And frankly, if unusual packets crash a router, the problem is with the router.

Re:Same as Vista

[Enderandrew]

I agree. Quite frankly, given that only one small manufacturer is reporting issues, and that they were able to resolve the issue on their end, makes me believe the issue is more with the router.

Re:Same as Vista

[Anonymous Coward]

Black hole detection was already in the network stack for Windows XP and predecessors.

It was not enabled by default.

Re:Same as Vista

[Anonymous Coward]

Exactly, Vista did that for me as well, only my router just hung up and didn't reboot.
A firmware update fixed that so this is not such a big deal as people make it out to be.

Re:Same as Vista

[Anonymous Coward]

It's so old it's described in Stevens. It used to be common when your link might hop over a SLIP connection, or some over bizarro-world protocol.

Re:Same as Vista

[Free the Cowards]

Path MTU discovery relies on getting ICMP packets from the router in question to inform the sender that the packet has been dropped due to MTU limits. Some routers don't send this notification, and some firewalls black it, which breaks path MTU discovery. Vista's black hole router detection allows it to discover the path MTU in this scenario, where the traditional technique simply fails. It's not the same thing at all.

Re:Same as Vista

[Anonymous Coward]

I guess MS doesn't like networks that drop packets on TCP ports 135:139 like a hot rock. Good luck MS ignoring a network like that.

Re:Same as Vista

[bwcbwc]

WTF? Does this mean if I disable uPNP on my router for security reasons, Windows is going to ignore the router? And what do they mean by "ignore"? Will it refuse to recognize the network connection at all?

What if the router ran Linux?

[HockeyPuck]

What would we do if these routers ran a linux based firmware? What would we do? Can we flame linux or do we continue to flame msft for abusing specs?

I'm looking for guidance from the /. flamers.

Re:What if the router ran Linux?

[pembo13]

There hardly any comments yet. Most are defending Microsoft. Who is this "we" that are flaming msft?

Re:What if the router ran Linux?

[SanityInAnarchy]

You can flame anything. Maybe it was Microsoft specifically targeting a Linux exploit? Maybe it was a driver, or a modification that is entirely the router manufacturer's fault?

I think the most telling thing about this is the danger of monoculture. When routers are only tested against specific versions of Windows... But that's not a flame, as there's no one entity you can blame for this. Good routers wouldn't have this problem.

Re:What if the router ran Linux?

[MrMr]

The poor MSerables are so used to being laughed at around here we don't even have to press submit anymore to make them feel it.

Re:What if the router ran Linux?

[maxume]

You know what I hate, I hate people that eat cabbage.

Except when it is broccoli.

Re:What if the router ran Linux?

[LarsG]

I`d say the router is worthy of flame no matter what OS it is running. A router must be stable, it should not crash even if you send complete random gibberish at it.

If it turns out that XPSP3 is sending broken UPnP traffic to the router, then MS is a valid flame target for not following the spec properly. That does however not absolve the router.

Re:What if the router ran Linux?

[Akaihiryuu]

Hey, I take any chance I possibly can to flame Microsoft, since I literally depise them. But I can't find anything to flame here...it's not Microsoft's issue, it is *clearly* the router manufacturer's issue.

Re:What if the router ran Linux?

[Anonymous Coward]

Really because it seems like you're looking for attention.

Re:What if the router ran Linux?

[Anonymous Coward]

I'm looking for guidance from the /. flamers.

Well, ok.

Go fuck yourself ya stupid git.

You're welcome.

Buggy Routers

[John Hasler]

Any router that can be crashed by anything that any of the computers connected to it do is seriously buggy. This is not Microsoft's fault.

Re:Buggy Routers

[Lehk228]

not quite, any router that can be crashed by bad packets is buggy, i don't see the need for routers to be designed to anticipate other bad behavior, such as dumping 110 V AC down all 8 pins

/pedantic

Re:Buggy Routers

[John Hasler]

> ...such as dumping 110 V AC down all 8 pins

Interfaces that comply with the Ethernet standard are transformer isolated (except for the brain-damaged idea of POE, but only the most idiotic router designers would implement that (and even POE should be fused)).

Re:Buggy Routers

[Hierarch]

Sure, it isn't MSFT's fault. However, the fact that SP3 triggers the problem is relevant due to the sizable install base that's about to develop problems. That is, if anybody out there actually has this router — I've never even heard ofit!

Re:Buggy Routers

[amorsen]

except for the brain-damaged idea of POE, but only the most idiotic router designers would implement that What's wrong with PoE? Personally I think it's brilliant, avoiding lots of little inefficient wall warts everywhere.

Re:Buggy Routers

[confused one]

If there's 110V AC on the data pins, those isolation transformers are toast. The router proc. may not crash, but it will most assuredly destroy the port.

Re:Buggy Routers

[Akaihiryuu]

Even though I am normally severely anti-Microsoft in every way possible, I have to bite the bullet and agree with you here. NO computer connected to a router should ever be able to crash it, no matter what happens. Even if Windows is severely broken and sends all manner of bad packets out, the router should be able to handle it without error (even if it just involves dropping the packets). If it crashes, it's the router's fault. A similar issue has been happening with WOW lately, something about the packets that WOW is sending out is causing certain routers and modems to crash. Blizzard is actually being quite generous in working with the router manufacturers to help find the problem, even though it is not their problem at all. Though the number of idiots on the forums insisting that it is a WOW problem is rather annoying. Myself, I don't have this problem, because I don't use consumer routers...I built my own out of some mini-ITX hardware and Gentoo.

Re:Buggy Routers

[Ravon Rodriguez]

Agreed, that would be akin to a MB manufacturer having onboard LAN that sends an obscene amount of voltage down the wire. Obviously it violates specs, and it's not going to happen. I'm of the mind that Microsoft didn't violate any specs here, its just the router manufacturer not having a current firmware and poor error handling (obviously it has some error handling, or else it would just lock up and not reboot).

Re:Buggy Routers

[Ethan Allison]

There's a difference between crashing something and physically melting it.

Re:Buggy Routers

[DrJimbo]

Oh please. Didn't those go out of style along with buggy whips?

Re:Buggy Routers

[prisoner-of-enigma]

not quite, any router that can be crashed by bad packets is buggy, i don't see the need for routers to be designed to anticipate other bad behavior, such as dumping 110 V AC down all 8 pins Well, duh! Routers shouldn't typically be designed to withstand a point blank 12-gauge shotgun blast either, but that's not the point of the post.

Routers should be designed to gracefully handle anything that comes down the wire that resembles an Ethernet frame, regardless of the content of that frame. You're arguing Layer 1 issues when the rest of us are discussing Layer 2/3 (or above) issues.

It's a cardinal rule of programming that you should always assume any input is complete and utter malicious garbage, designed specifically to destroy your precious code. Or it least it was when I was in college a few decades ago.

Re:Buggy Routers

[Lehk228]

that was a few decades ago, now the goal is to get it out the door so it at least still works when the customer gets it home, then blame improper setup or other equipment for any problems.

here's an idea for ya?

[arbiter1]

How about you buy a good name brand router and not one them cheap no name brand's. i got a linksys and a dlink, both work just the same before and after sp3 was install on both my desktop and laptop.

Re:here's an idea for ya?

[amnezick]

you bought a separate router for your laptop?

Since I use Vista

[Clockwurk]

I don't have this problem.

Re:Since I use Vista

[Lennie]

You however have other problems I'm guessing. ;-)

Re:Since I use Vista

[Colonel Korn]

You however have other problems I'm guessing. ;-) Having just installed Vista 64 after trying Ubuntu for a month and giving up because of crashes, and having had 0 crashes, even program crashes, in my first 4 days, I'm no longer believing the Slashdot meme that Vista is such a piece of crap. It was the smoothest OS install I've ever done and it's a much snappier OS on my system than XP or Ubuntu were. Maybe 4 gigs of ram is the trick, and maybe I'm just lucky to have well supported hardware, but on my system it's outstanding.

Re:Since I use Vista

[Lennie]

Well, yes I've heared people say one week, everything is fine and weeks later, Vista is crap.

And I've read horror stories on fora like this about Vista just dying and never getting up.

And I'll stop right there, that sounds to much like FUD. That was just hearsay.

Obligatory quote

[timbck2]

All your base (stations) are belong to us!

Router Trouble.

[fuzzyfuzzyfungus]

As some have said, if a machine on the network can crash the router(short of violating physical specs for things like ethernet voltage and polarity), then the router has Issues.

What I don't understand is why so many of your basic 4 ports lan, one port wan, and an antenna type routers have such lousy firmware. I understand that the hardware is built right down to price, and isn't going to be exciting; but software is a different matter. There are really only a few chipset variations in general use, OpenWRT supports most of them and provides a solid and extensible foundation. ddWRT is less extensible and flashier, still solid. Tomatoe is out there as well. In a world where people are literally giving high quality router firmware away, how can anybody ship a router with bad firmware?

Re:Router Trouble.

[John Hasler]

> As some have said, if a machine on the network can crash the router(short of violating
> physical specs for things like ethernet voltage and polarity), then the router has Issues.

Unless you are applying 10KV or something miswiring should not crash a router.

Re:Router Trouble.

[Hierarch]

"Don't blame me, I'm an interpreter, I'm not supposed to know the difference between a power socket and a computer terminal!"

Re:Router Trouble.

[Anonymous Coward]

The firmware projects you mentioned all require a certain (amazingly small) amount of RAM/Flash to work. Believe it or not, router manufacturers make devices which have even less memory, and then they need smaller, more "efficient" firmwares. If you're wondering how much money can be saved by designing a device with 2 to 8 megabytes of RAM instead of 16 or more, even though skimping on RAM means you need a proprietary firmware and $10 buy half a gigabyte of DDR2 RAM retail, join the club.

Re:Router Trouble.

[kybred]

Unless you are applying 10KV or something miswiring should not crash a router. Like this?

Well, OK, it's not 10KV but I think it would do the job.

Re:Router Trouble.

[ScrewMaster]

No kidding. Take Linksys, for example ... they're a division of Cisco for chrissakes, and you'd think they could devote some resources to providing top-notch firmware. Sometimes I think Cisco doesn't want the Linksys stuff to become too good. Wouldn't want it to compare too well to their higher-end corporate products, I suppose.

Personally, I use Tomato (simply because it's solid, I like the GUI, and it has the specific features that I happen to need) and haven't had any problems with it with XP SP3. The author is also very helpful and responsive.

Re:Router Trouble.

[John Hasler]

Any NIC that is damaged by this does not meet the spec (which is, unfortunately, almost all of them).

Re:Router Trouble.

[gad_zuki!]

Ive wondered that too. I imagine the reason why has to do with software engineers protecting their jobs and a corporate Not Invented Here attitude.

Re:Router Trouble.

[gad_zuki!]

>Wouldn't want it to compare too well to their higher-end corporate products, I suppose.

That's a pretty safe assumption. It would be stupid for cisco to improve the linksys line. My dd-wrt linksys can probably replace a cisco 800 in many situations. It costs perhaps 1/10th the price.

I wouldnt be surprised if cisco bought them so they could keep them in the residential market and out of the corporate market.

Crappy router.

[fluffy99]

Billion makes crappy knock-off routers, that were crashing or not working long before XP SP3 was released. Perhaps XPSP3 does do something different with uPNP, but that's not where the blame needs to be assigned. As an aside, uPNP is a crappy idea. Do you really want your OS and any programs (malware included) to have the ability to change your external firewall?

Re:Crappy router.

[Kjella]

As an aside, uPNP is a crappy idea. Do you really want your OS and any programs (malware included) to have the ability to change your external firewall? If they've compromised my machine and my personal firewall, I imagine all the interesting stuff like passwords and credit card numbers and whatnot have left through a HTTP post already. Remember, the external firewall on the router doesn't know what process is doing it. so unless you have a real lockdown it won't do much good. Sure it won't be able to open incoming ports and make it into a rooted server, but honestly that's the lesser of my worries. For most people uPNP is the alternative to finding the big old "OFF" button. There's a reason many troubleshooting guides say "try disabling the firewall" even though it's horrible security advice, and it's because people muck it up. Having each application update the firewall on a need-to-have basis is in fact the most competent firewall administrator you can have on most home networks. If anything effort should go into reducing the problem of malware installation.

Re:Crappy router.

[Anonymous Coward]

uPNP was designed for the stupid consumer.

99.997% of all consumers are too fucking stupid to "open a firewall port". They go "huh? my windows is broken, fix please!"

So in response to that manufactureres came up with uPNP to allow the computer to open up all the holes it wants without effort. This makes it great for games, viruses and trojans to have unfetterd access through that firewall.

EVERY SINGLE PROBLEM WITH COMPUTER SECURITY IS BECAUSE USERS ARE STUPID.

Re:Crappy router.

[Tony Hoyle]

If you've downloaded some malware, say via a flash advert for example, your machine hasn't been compromised 'as such' but it's trivial to tell a upnp router to open all ports. Getting credit card numbers is harder as they're stored in different places.

Many upnp routers can be configured to reroute outbound traffic to an external proxy, too. So you dial up your bank and suddenly it's doing a MIM attack on your secure connection (and don't think users care about security popups.. they simply don't).

There's a reason why professional routers don't support such nonsense.

Re:Crappy router.

[Anonymous Coward]

You do when you're at home and using a program that randomizes the ports every time (i.e. bittorrent). Or if a user is playing a game but doesn't know how to forward the ports on his router.

UPNP is very much a technology targeted for the home-user, not the enterprise. And for the very paranoid, you can always turn UPNP off on the router.

Re:Crappy router.

[Anonymous Coward]

There's a reason why professional routers don't support such nonsense. Economy of scale? Elevated risk?

Professionals wouldn't use gmail either. Does that make it inappropriate for home users?

Re:Crappy router.

[VENONA]

"Does that make it inappropriate for home users?"

Yes. A home user has just as much to lose a corporation, *in relative terms*. Completely hosed is completely hosed. The average home/SOHO user is just operating at a greater disadvantage, because few have the resources required to understand the threats, and make good decisions. People who read technical sites of any sort are a minority, in every survey I've seen, and have been for years.

In a post above, I think Tony Hoyle might have been referring to the .swf-based attacks that were being widely used back in January. Hopefully most people here turned off upnp on their router back then. Anyone that didn't should probably read https://www.kb.cert.org/vuls/id/347812
It has links to MS Knowledge Base articles, and lots more info.

Thinking that, "I'm just a home user," is basically what's given us botnets. I see no way that problem can be fixed, but sometimes I have to put my Don Quixote hat on, over my Security Guy hat, and post something.

Re:Glad I disabled auto-updates

[Anonymous Coward]

I always like to check what is foisted on my machine. It's been trying for weeks to get me to install WGA before I'm allowed near Office upgrades so I eventually gave up on that box and switched to OOo.

Except for label printing. What labels doesn't OpenOffice print?

Correlation != causation

[catscan2000]

It sounds like the Billion router's firmware had a really bad bug that happened to be poked by Windows XP SP3. Unless if this was in a third party library or some external code that they were using, I wouldn't be surprised if this was limited to just Billion routers.

XP SP3 didn't _cause_ the bug; it merely happened to recreate a condition that triggered a bug inside the router to crash itself. :-)

I installed SP3

[LM741N]

and now women won't go out on dates with me anymore. ....ok, they wouldn't with Service Pack 1 or 2 either, so I'm now trying Vista.

Re:I installed SP3

[DavidD_CA]

Judging by most of the people here, you won't have any better success with Ubuntu or SuSe either.

You might try Mac OS. Or at the least, get an iPhone and use it liberally while in public.

Re:I installed SP3

[Kjella]

I installed SP3 and now women won't go out on dates with me anymore. ....ok, they wouldn't with Service Pack 1 or 2 either, so I'm now trying Vista. Since you're already overqualified, may I suggest Linux? No, it will not help with women but I promise your computer experience will be better...

Re:I installed SP3

[Anonymous Coward]

so I'm now trying Vista

Enjoy dying alone.

Re:I installed SP3

[DeadInSpace]

Judging by most of the people here, you won't have any better success with Ubuntu or SuSe either. With Debian, on the other hand...

Re:I installed SP3

[bh_doc]

With the money I saved installing free software I could afford hookers!

Thanks, free software! *two thumbs up*

Here's the technical reason

[Anonymous Coward]

Quote from their website:

"After detail analysis, we found that Windows XP SP3 sent out the DHCP packet with the Option 43 data (include Microsoft's 'Vendor Specific Information'), but Windows XP SP2 sent out the DHCP packet without the Option 43 data. However, the Option 43 data is not compatible with Billion's original definition, so it will cause this problem. The affected firmware versions of BiPAC 5200 series are 2.9.8.x and 2.11.0.x~2.11.33.x. There is no impact to BiPAC 5200 series if the firmware is 2.10.x.x. Please check Appendix A for checking your current firmware version."

http://au.billion.com/downloads/Notice-Billion-5200-series-via-Windows-SP3.pdf

Re:Here's the technical reason

[Richard_at_work]

Just to clarify the parent posters information, Option 43 (Vendor Specific Information) is a valid but optional part of the DHCP spec, covered in RFC 2132, part 8.4. A server not equipped to handle the Vendor Specific Information must ignore it.

Re:Here's the technical reason

[Todd Knarr]

That doesn't make sense, though. Option 43 is sent by the DHCP server to the client. In this case the SP3 machine is the client and the router is the server. It's the server's responsibility to correctly format the vendor-specific data according to the vendor's spec, but the SP3 box shouldn't be the server in the scenario in question.

Re:Here's the technical reason

[Todd Knarr]

Aha. But it still doesn't make sense. If SP3 is sending option 43 as a client, then the data will be in Microsoft's vendor-specific format, not Billion's. The only way Billion would have a problem is if Microsoft changed the format without telling anyone. And why would a router need to look at Microsoft's vendor-specific information in the first place?

Re:Here's the technical reason

[A5un]

Not necessarily. A client (or an app on the client machine) can issue a DHCP UPDATE in order to find out specific DHCP Option response. The DHCP server in this case will answer with DHCP ACK containing the answer to the Option or simply ignore it (depends on the configuration). At least that's how dhcpd works.

Re:Here's the technical reason

[Anonymous Coward]

I guess the ignoring part was a little too hard to implement, so they opted for the crash and reboot feature instead.

Streams crossed

[Anonymous Coward]

Slashdot corporately supporting Microsoft? I must be new here.

Exactly

[biolysis]

Your point is the relevant one. Regardless of what OS did this, the fact is that no computer should ever be able to crash a router period. The incessant MS bashing reaches absurd levels sometimes.

Flamers and newbie programmers

[Anonymous Coward]

This is a problem with the router's firmware. It has nothing to do with XP, or any other Operating System.

First off:

  - Who the fuck didn't isolate the rebooting mechanism and interrupts from the main loop that relays packets?

  - Who the fuck didn't include ring-0 protection or anything like it on the critical interrupts like for example a reboot or a complete shutdown?

And finally, what kind of a moron faults an OS for a violation of firmware design? /. trolls, appearently.

so is it microsofts fault also?

[atarione]

that if you go to their web page and look on the left hand side it says URGENT NOTICE blah blah blah XP SP3 then "click here" if you are over "click" it takes you to one of their products (not the affected one btw) page if you are over "here" it takes you to the firmware. kinda funny really.. I would like my vote to the numerous others that suggest this should not have been subbed as a XP SP3 problem but a cock up on Billion's part.

Title is correct

[Anonymous Coward]

Obviously, Windows XP SP3 is crashing a router. It doesn't say that is Windows XP SP3 fault.

I'm sorry, broadband router manufacturer who?

[urbanriot]

Billion? Never heard of'em. My Linksys router isn't complaining...

Re:I'm sorry, broadband router manufacturer who?

[Anonymous Coward]

hmm.. i have had issues with random lockups on SP3 PC's on some routers, including:- Netgear 54G series, Netgear Rangemax, also with Dynamode.

I thought it was my router at first, but upon testing by isolating all PC's and leaving the router on overnight for several days, found that it worked fine. As soon as an SP3 PC was online, within about 6-15 hours the router would lockup solid, requiring a reboot.

If this is indeed SP3's fault, how would I go about starting a class action lawsuit against MS?

XP3 or the router's fault?

[93 Escort Wagon]

While I agree that a router shouldn't crash no matter what an attached computer does, I completely disagree with the Windows fanboys who claim this is a non-story. It's obvious SP3 is doing something different network-wise than any other OS, be it Linux, Mac, or even Windows XP SP2. It's important that people with the effected brands of routers have this information, so they can choose whether or not to delay upgrading. It's also important for network admins to find out exactly what SP3 is doing, and get Microsoft to fix it (or work with the router manufacturers, depending on what exactly is going on). Microsoft's recent track record vis-a-vis security and interoperability is definitely better than it once was; but we can't forget that they've done some very stupid and/or insular stuff in the not too distant past.

Re:XP3 or the router's fault?

[mapleneckblues]

So Microsoft is supposed to fix something that it might be doing right but the router's shitty firmware cannot support?

Re:XP3 or the router's fault?

[John Hasler]

> While I agree that a router shouldn't crash no matter what an attached computer does,
> I completely disagree with the Windows fanboys who claim this is a non-story.

I claim it's a non-story and I sure as hell am not a "Windows fanboy".

Re:XP3 or the router's fault?

[Tony Hoyle]

Windows fanboy? Show me a copy of vista and watch me foam at the mouth.

This is a non-story.

Re:XP3 or the router's fault?

[Shados]

Its implementing an optional part of the specs, but its still part of the specs, and the packets its sending are valid in every ways, shape and form. The router is buggy, period. The router's makers figured that since no one or almost used that part of the spec, that they didn't need to test it as a valid use case, and it bit them. Don't blame MS for that.

What the /. reads this as..

[HockeyPuck]

I think for April Fools we should rename this article since we know how great the editors' grammar is.

Billion's routers crash after upgrade to XP SP3.

OMG a Billion Routers crash....

maybe, maybe not

[frovingslosh]

I agree that no router should crash based just on packets it passes. But there are a few issues here. If SP3 is causing something akin to a DOS, and a router's tables are filling up due to bad packets, it might very reasonably decide that things are so bad that the best thing for it to do is a reset. We don't blame the router maker when an external DOS attack interrupts Internet access, why blame it if the DOS is from Microsoft software on the inside?

And there is also the potential issue of this being UPNP related. UPNP is a completely bogus thing, but Microsoft strong armed the industry to support it and it's in most routers and many users don't know to disable it. UPNP could certainly give ways to cause this issue, and I only hold the router itself responsible to the extent that it supports this blasphemy.

Re:maybe, maybe not

[PhrostyMcByte]

Vista had problems too. They changed it to do automatic receive window scaling and it affected some of the cheaper routers. I have not RTFA or anything, but maybe this is something similar.

Re:maybe, maybe not

[Anonymous Coward]

Internet access, computers, and ones time and sanity are expensive enough that to have such low expectations for basic network infrastructure is absurd.

If the NAT table fills up its still not acceptable to crash or reboot. I would demand the device have enough sense to expire the oldest or LRU or I would take it back.

Re:maybe, maybe not

[lazybeam]

Where I used to live there was an expensive-looking Netgear router that would always crash when I logged into my VOIP provider. This provider had no problems with any other (cheap or otherwise) router (including the ~$60 Billion 5100 modem/router I'm using now). It didn't matter which softphone I used, on both Linux and Windows, the Netgear would lock up hard: even with the latest firmware. It would only come back after shutting down the softphone and waiting 3-5 minutes.

Perhaps the funniest thing was when I tried a different VOIP provider and had no problems, even though they both used Asterisk. I dropped the second one when I moved house...

Odd, or Convenient?

[FurtiveGlancer]

Odd how the summary fails to mention that the problem is only with this obscure model... Maybe specious or suspicious would better describe the article's failure to mention this rather key piece of information.

Not MS to blame

[Opportunist]

As much as I hate defending the Redmond Computer Virus (tm), that's the router's fault.

Now, if SP3 created nonstandard packets that most routers still swallow but a router drops because they don't work to spec, blame MS. If the router replied with a bogus message to said nonstandard packet that locked up XP, blame MS. But a router HAS TO be able to accept a bogus packet. It may drop it, report it or if it feels like it send it on a roundtrip in hope that some machine can figure out what it's about, but it may NEVER crash due to it.

I hope I don't have to mention the security implications of this.

Re:Not MS to blame

[Anonymous Coward]

If the router replied with a bogus message to said nonstandard packet

or if it feels like it send it on a roundtrip in hope that some machine can figure out what it's about If a router receives an invalid packet it should be DROPPED. The responses you have listed are the makings of a good DoS attack or broadcast storm.

uPnP should be off in router anyway!!

[keithy]

If theres not yet any malware around that uses uPnP to 'drop' the defences of the routers firewall, there will be one day!!

Re:uPnP should be off in router anyway!!

[^_^x]

http://it.slashdot.org/article.pl?sid=08/01/14/1319256&from=rss

Speaking Of SP3...

[xanadu-xtroot.com]

Last week I updated my GF's XP_Pro machine to SP3 (she insists on having her machine using Windows despite having a better time using my KDE/Gentoo machine...). Since then, I've noticed that it's sending out SMB keep-alives about every 5 seconds to my machine (which is the Samba "PDC" also). SP2 wasn't doing this (or WireShark didn't pick up on it, anyway).

Could this be something that would hose a router as well? A ton of useless keep-alives?

notebook & desktop

[Max_W]

My notebook ASUS was crashed by SP3. I had to reinstall the OS after 2 days of trying to recover.

On the desktop I installed SP3 without ay problem.

My freee advice:

--**++ Back up your data before even thinking of SP3 install! ++**--

NAPT != Firewall

[Luke-Jr]

uPNP configures port forwarding for a NAPT (aka NAT) router. NAPT/NAT is *not* a firewall, and should not be treated like one. Its sole purpose is to translate addresses and ports (Network Address and Port Translating) between the internal and external networks. It is not meant to protect computers on either end from each other. uPNP facilitates the NAPT job by giving applications an easy way to automate the needed port forwarding for the WAN->LAN direction. If you want a firewall, get a real firewall.

Re:Glad I disabled auto-updates

[LVSlushdat]

Thats why I use Windiz instead.. You just turn off automatic updates, and set yourself a reminder in Thunderbird to remind you every month to login as admin (you DO run as a limited user, I hope!) and visit Windiz update, and get ONLY the updates you approve.. Though, I gotta funny feeling MS will eventually pull the same b.s. on Windiz they pulled on Autopatcher..

apples to oranges

[spazdor]

Did the hardware manufacturers all just write flawless Linux drivers and buggy Windows ones?

Or did Linux developers just go a step further than Windows did, and take it upon themselves to make sure that hardware works properly on their OS?

So in other words

[Sycraft-fu]

Don't buy Billion routers, because not only do they not listen to specs, when something does what they don't expect they just up and crash.

Billion?

[Anonymous Coward]

Sorry, When you said "router" I thought you meant something from a real company. I've never heard of Billion or Raaj. How many GPL violations does his router have? I'll just continue using my OpenWRT setup, thanks.

What's worse

[Sycraft-fu]

It sounds like the packet isn't bogus. MS chose to implement an optional part of the DHCP spec (vendor information). As per the DHCP spec if your device doesn't implement the optional parts, you just ignore them, not crash.

So this isn't MS sending a bogus packet, or even doing a "Windows own spec," thing. They are properly following the DHCP spec, and this POS can't handle it. I mean I'll give someone a tiny bit of credit if the problem is due to bad data. Not a lot, it's still a bug that needs to be fixed, but at least it was something unexpected. However when you are crashing because you didn't deal with part of a spec, well then you get zero sympathy.

Re:Glad I disabled auto-updates

[Anonymous Coward]

Open the updates menu, click "Advanced", untick the WGA notification, tick the "Do not notify me about these updates again", click next and click "OK" on the box that pops up. That stopped the WGA updaet for me..

Re:Nah...

[cnettel]

If you can crash the router, you have a possible DDoS attack. If you can do it on the WAN port, it would certainly be a flaw in the device. Depending on the crashing behavior, it is also possible that this is actually an exploitable path that could be used to permanently reflash the router for malevolent purposes.

another SP3 bug

[Anonymous Coward]

I would have to says it's the router's firmware that is the problem. the reasons have already been stated by others. but SP3 is buggy. case in point: updated laptop with sp3 tried to access my NFS got blue screened (bad header pool) probably driver issue with my intel pro/100. before the updated no such problem.

linksys wrt54g

[Anonymous Coward]

Hmm, I just replaced my linksys wrt54g w/vonage because it was doing this very thing. I hope I didn't just throw away an otherwise good router....doh

Re:Glad I disabled auto-updates

[afidel]

I'm more glad I disabled uPnP, it's a very poorly designed spec with even crappier real world implementations. It's about the most bug-prone technology I know of.

one model from one manufacturer

[confused one]

This only affects one model (BiPAC 5200) wireless broadband router, from one manufacturer (Billion), who's firmware has a bug. The model in question is found in Australia and Europe. A firmware update is available for download. End of story.

Re:one model from one manufacturer

[Anonymous Coward]

never heard of this 'billion' in australia sorry

Re:one model from one manufacturer

[Anonymous Coward]

They are one of Australia's biggest router manufacturers, and are particularly popular with geeks for their excellent feature support and VoIP performance.

And they have a deserved reputation for having occasionally buggy products too.

Is there a way to rate an article as utter crap?

[mrmeval]

It would be nice to watch misleading, uninformative crap be forced off the main page.

Mod parent up

[zlogic]

Please mod parent up, UPNP is ideal for home networks when you have one external IP but several PCs have servers running (non-passive FTP transfers, VOIP apps, p2p software). Setting a DMZ is not an option, and manual configuration is usually horrible on consumer-grade routers (D-link hardware reboots after adding a new firewall rule, and these rules can be added only one at a time), and you still have to assign incoming ports in applications. UPNP makes port forwarding configuration as easy as checking a "use upnp" option in applications that support it.

Re:Mod parent up

[Tony Hoyle]

If you have several PCs with servers running get multiple IPs or run ipv6 or something. Don't rely on port forwarding hacks. You're *already* out of the home user demographic by running servers in the first place.

VOIP does't need port forwarding at all SIP pairs with STUN and Skype has its own methods. Any client protocol that does is broken by design... but they're few and far between - my firewall is locked down tight and I won't use port forwarding on principle - and it's pretty rare for something to have an issue.

Re:Mod parent up

[zlogic]

If you have several PCs with servers running get multiple IPs or run ipv6 or something. Servers are integrated in a lot of applications, for example the eDonkey2000 p2p protocol won't connect two users who have a "low-id" (no port forwarding). A low-id and a high-id user can connect because the low-id can establish a TCP connection to the high-id user because the low-id client knows which port to connect to.
And file transfers in IM clients usually only work when at least one user has proper port forwarding, because the receiving side acts as a server. ICQ is especially bad, they've even recently implemented a file-transferring proxy that is really slow but at least works most of the time.

Re:Mod parent up

[jrumney]

VOIP does't need port forwarding at all SIP pairs with STUN and Skype has its own methods.

You're right, VoIP has workarounds, but there are significant performance gains from eliminating the extra TCP connection to an external server, not to mention privacy gains from having your VoIP or Skype traffic come directly to you via UDP. You may not notice it too much for low bitrate voice only, but throw video in the mix, and the difference is immediately obvious.

Re:Mod parent up

[fluffy99]

So you're saying most home users don't want to learn how to properly setup a firewall, and like having a setup where the software can punch its own holes and/or port forwarding. I can certainly see the user friendly aspect. It sounds great until you find out that uPNP conflicts happen when two PCs are convinced they need a specific port forwarded to them all the time (btdt with a few clients). The current uPNP implementations are still not very well done as evidenced by the Billion routers blowing up when the Windows uPNP client changed. Better routers have dynamic port forwarding which works pretty well. As for "consumer grade" routers - you definitely get what you pay for.

I totally agree- i had this happen 3 weeks ago

[Anonymous Coward]

It was my WRT330N Wireless-N Gigabit Gaming Router, it started going into almost a constant reboot whenever there was heavy traffic. I did NOT have SP3 installed. I was pissed because the router cost 199.00 at the time and only lasted a year. It was definitely a hardware defect on the router. I have a feeling that after a while linksys routers just start rebooting. Also wireless performance for very very poor after the reboots started happening.

DO NOT BUY LINKSYS.

captcha: thanks

Re:I totally agree- i had this happen 3 weeks ago

[silent_artichoke]

My Linksys WRT54G has been plugging along for 5 years without a single hiccup. The only time it is not working is when the power goes out. You had a hardware defect, it could happen to anyone. No reason to blame a company's entire product line for a single bad unit.

Re:Glad I disabled auto-updates

[Ihmhi]

Right, until a "critical security update" turns that option right back on. Better to just turn off Automatic Updates and disable Security Center in Administrative Tools > Services so it stops whining about your computer "not being protected".

Not only

[RocketSoft84]

It's not only the router that makes the endless reboot, apparently there are a few other things that can crash you. Some, or maybe all, have something to do with gdi32.dll. Either way, it was a pain in the butt to fix when it first happened.

Re:Glad I disabled auto-updates

[Phil John]

Meh, works fine here (on a Draytek 2900) for the odd app that can make us of it. I think a lot may be down to less than stellar implementations in router firmware.

Sequential ones

[cheros]

If I have a source, say a spreadsheet, with a column with bunch of numbers in, there is no easy way to quickly do a run of labels with one column entry per label, and that's not just because you have to "register a datasource" before you can even use field references.

Letters for a mail merge, yeah, no problem. A sheet of labels (all different)? Nope.

Just try it, you'll see what I mean.

WGA complaints on virtual boxes

[Anonymous Coward]

I installed SP3, and did not have those problems.

However WGA complained on all Qemu boxes and I had to re-register XP. Registration was a simple connection to MS, and no key or data was needed.

RTFRFC

[Krunch]

The client sends a DHCPREQUEST packet that contains options it is requesting from the server. The server is then supposed to craft a DHCPACK packet containing values for the options requested by the client.

http://www.ietf.org/rfc/rfc2132.txt

Re:RTFRFC

[Krunch]

Sorry I gave the wrong URL. DHCP is described in RFC 2131. RFC 2132 only lists valid options. http://www.ietf.org/rfc/rfc2131.txt

Re:RTFRFC

[Todd Knarr]

Yes, but that request should only have the request for option 43, which wouldn't be vendor-specific. Only the response should have vendor-specific data in it and the XP client machine shouldn't be responding to DHCPREQUESTs, it should be the one sending them.

Re:Glad I disabled auto-updates

[fluffy99]

It's a little scary when that "odd app" includes visiting a webpage with a malicious flash script.

http://www.gnucitizen.org/blog/flash-upnp-attack-faq/
http://blogs.techrepublic.com.com/tech-news/?p=1902

I have seen P2P programs make routers go wacky.

[antdude]

I have seen programs make routers go nuts like my old Netgear RT311 router from 2001 if I set the connections too high witout using a low limit with eMule. Same on a neighbor's D-Link router from a few years ago. Please remember these are consumer routers, not high-end fancy ones.

Re:I have seen P2P programs make routers go wacky.

[John Hasler]

> Please remember these are consumer routers, not high-end fancy ones.

And therefor they have to have buggy firmware else why would anyone buy the fancy high-end ones? Besides, proper design and debugging costs tens of thousands of dollars: an intolerable expense in a product you are going to sell a few million of.

Re:I have seen P2P programs make routers go wacky.

[antdude]

It seems like all consumer routers are weak to me regardless of their ages and updates. :(

Standards Shell Game?

[Zero__Kelvin]

The consensus seems to be that the router is at fault if SP3 crashes it. There is a major factor that people are overlooking here. uPNP is a M$ "standard." So here is the possible, and I would even assert likely, scenario:

M$ creates the uPNP standard, then revises it, then revises it again. To the extent that it is a standard, different versions of the "standard" are made available to different router designers, based on how close they come to touching their palms to the floor when bending over for M$. Now, those who handed over their first born have the newer tweaked standard available, and if they comply their router doesn't crash. In the meantime, other router companies have a different/older standard, to which they comply fully. Of course, SP3 makes use of the newer, less widely disseminated standard. Doing so causes implementations that haven't "paid up" to crash.

Yes, this definitely sounds like a scenario imagined by a guy who wears a tin-foil hat to those who don't know the M$ history, haven't read the M$ internal documents known as the Halloween Documents, etc. To people who know the history and understand how M$ works, this is a likely though unproven scenario.

Re:Standards Shell Game?

[John Hasler]

It makes no difference what Microsoft does. If software running on a connected computer crashes a router the router is buggy.

Re:Standards Shell Game?

[Kalriath]

You're almost as bad as Twitter. Try RTFA, it explains exactly why it happened, and exactly why it's the router manufacturers fault.

Re:Standards Shell Game?

[Zero__Kelvin]

Read what I wrote again. I didn't say the router isn't buggy. I said the router may be following the particular version of uPNP, which may prescribe the buggy behavior. Ever heard the expression Garbage in, Garbage out? If you follow a "standard" and the standard is flawed, intentionally or otherwise, then the implementation will be flawed.

Re:Standards Shell Game?

[Kalriath]

Ack. Apparently TFA doesn't. But Billion themselves do.

Re:Standards Shell Game?

[Zero__Kelvin]

Thanks for the Link!

"After detail analysis, we found that Windows XP SP3 sent out the DHCP packet with the Option 43 data (include Microsoftâ(TM)s âVendor Specific Informationâ(TM)), but Windows XP SP2 sent out the DHCP packet without the Option 43 data. However, the Option 43 data is not compatible with Billionâ(TM)s original definition, so it will cause this problem."

So the question becomes, where did Billion get the original definition? Were they accurately following an older uPNP spec? Did they have access to the newer one, if any, prior to the release of the firmware? As I originally conceded in my original post, this is not evidence that what I propose is true. However , it certainly would be consistent with my example scenario in my original post.

Re:Standards Shell Game?

[Kalriath]

WTF? You didn't even read it! It specifically says DHCP, not fucking uPNP (which is a crappy idea anyway!)

It's not consistent with your contrived scenario in any way shape or form. Nor is it evidence that what you propose is true.

Re:Standards Shell Game?

[Zero__Kelvin]

Yes, I stand corrected. I did see DHCP, but did not look into it enough to see if option 43 is part of an rfc (it is in rfc 2132), or if uPNP has its own "DHCP" with M$ specific extensions, which is one of the ways M$ uses the embrace/extend/extinguish method. So yes, it is almost certainly Billions fault, assuming of course that M$ is implementing rfc 2132 option 43 in a manner consistent with that standard.

Also, I owe you a debt of gratitude. It's really great of you to point out that it is not evidence that what I proposed is true, just in case people didn't notice that I explicitly pointed that out in each of my posts. Thanks for your help there ...

lol. Open source owns, the resources are awesome.

[joocemann]

I just began installing ubuntu on my laptop with plans to install it on my next PC. I use my current PC for all things, but will keep it for music production. I can't wait to use ubuntu on a new powerful pc! I've seen some things about ubuntu beryl that look phenomenal and make vista's visual interface look like childsplay.

I look forward to the day microsoft is #2.

SP3 apparently helped brick an NVIDIA card

[Anonymous Coward]

Although I'm used to do firmware updates from alternative bootable media, I've also updated firmware using official utilities designed to work in Windows.

Recently, I apparently managed to help Windows brick an NVIDIA card (FX 1700), using NVIDIA's official flash utility.

I contacted HP about this, because the update utility from NVIDIA shipped on the driver CD from HP, and the graphics card was officialy supported on this workstation; also, the firmware update was marked Recommended by HP (if it wasn't, my company would apparently have had to pay for the replacement card). The support person I talked with wanted to blaim SP3.

SP3 is basically all patches since SP2 in one package, plus a few minor updates, so what's the big deal? Obviously, I didn't try my luck, so I abandoned the update, even from other bootable media, since I didn't feel I got a very scientific/accurate answer, like it usually is in the IT profession.

More accurate title for this article:

[Anonymous Coward]

Crappy Router No One Has Heard Of Is Crappy

How many ...

[PPH]

... more service packs full of bugs does Microsoft have to release before you people give up and switch to Vista?!!!

anomalies in networking behaviour between SP3-SP2

[Anonymous Coward]

there are definitely differences between how SP3 handles networking and SP2 that "shouldn't be there".

I can CONSISTENTLY force a reboot (kernel dump!) simply by connecting alternately to the same machine with an old RDC client vs the new one. (denial-of-service, anyone? but then you'd need an account on the machine so it's not THAT critical a vulnerability)

Re:Glad I disabled auto-updates

[Kalriath]

WindizUpdate does not have any updates of any kind released after November 2007, so it's more of a security risk than Windows Update.

Re:Glad I disabled auto-updates

[psued0ch]

It could just be due to the fact that UPnP is ubiqitous and users experience difficulties with it when you consider the thousands of available devices and peripherals that are used with the technology. Similar to the topic issue.

OMFG News!..

[Anonymous Coward]

Standard MS operating procedure..nothing new..move along...nothing to see here...

Oh fuck almost forgot...."I'll be modded as troll/flamebait"

Ah, if only mod points were directly related to the amount of pussy you got. Wait, then this would't be slashdot, silly me.

Happened to me - DIR-655

[Anonymous Coward]

This recently happened to me and my next door neighbor. Router just rebooted constantly, even after reconfig, reflash, changing channels everything. I finally pull the plug on the DLink DIR-655 and bought an Apple Airport extreme. Hasn't happened since. I am using XP Pro / SP3.

A rather minor bug like this gets attention..

[thejynxed]

While issues such as SP3 causing Office XP and 2k7 service packs appearing as updates in Microsoft Update on a machine that has never had any version of Office installed gets no mention... Automatic Updates keeps on trying to force them down :)

Also, anyone using USB 1.1/2.0 devices under SP3 have any issues? On my laptop, I have USB 2.0 devices and USB 1.1 devices running on a USB 1.1 connection (yes, the laptop is that old). When I reboot the machine, XP refuses to boot to the desktop (just hangs after the Windows loading screen) unless I unplug all USB devices from the laptop. Then it continues booting. This behavior did not occur under SP2. I wonder what changed. I've read of similar things happening with some file named Verclsid.exe hanging while trying to authenticate something, but the solution of renaming it with the .old extension didn't seem to work.

I would give up and go back to SP2, but meh :P

Re: Commercial or otherwise

[Douglas Goodall]

I agree that a router should never crash from being send garbage.

I think it is fair after reading these posts to point out that there is a difference between a commerical router that is engineered to closely adhere to standards and is expected to be reliable and secure, and the alternative.

In this case the alternative is some piece of hardware that seems to execute a Linux kernel that may or may not be standard. If a Linux kernel on some hobby hardware (or inexpensive home stuff) is being used as a router and it seems to be working, that is great and you just saved ten thousand dollars buying a real router. But never forget there are highly engineered hardened routers that are expensive because it took lots of time and trouble to assure their reliability, perhaps many years in the case of Cisco.

I know people are quick to grab some PC hardware, load up unix, and declare themselves to have some network infrastructure, but these are toys in comparison to the real stuff. Businesses with important work to do and important data to secure must use the right stuff.

I mean no disrespect to Linux, but it is just not the same. Now there are router companies that make minimum hardware and load up a Linux kernel, and declare this a router. You get what you pay for.

Re: I beg to differ with you

[Douglas Goodall]

I believe that a router should have robust enough protocol handlers to survive a malformed packet. If it can be crashed it is not merchantable.

Do you think you could reasonably sell a firewall with an ad that says, "This firewall will stop any bad stuff that is coming at it, as long as what is coming at it is perfectly correct protocol code that meets the RFCs in all respects and breaks no rules. Ha ha ha.

could it just be the router is too slow?

[zmollusc]

Maybe the guy should let it keep rebooting? _Some_ of the EvilGovConpiracyProtocol packets will be getting through, and once all the results of scanning all the hardware and data on his network have been passed to the TotalitaranDatabase servers there will be less data sent and the router will stop acting up and will function ok ( until the daily update is sent out. But that should be a much shorter burst of messages).

Once upon a time...

[bgraw3]

I had an old Linksys wireless router, a Toshiba laptop with the original (no service packs) Windows XP, several wired desktops, and a Mac Powerbook.

The "new" Toshiba laptop usually booted Linux. But, there were still a few things I had to do with Windoze, so it still had dual boot capability.

Usually, within 15 minutes of booting Windoze, I had to physically reboot the router.

The solution: First, don't boot to Windoze. Second, get a new router.

The problem did not occur with the wired machines. I always figured it was something wonky in the new XP wireless driver. I tried firmware upgrades, new drivers on the laptop, etc.

At some point, I realized that a new router was far less expensive than the time I was wasting on this Windoze/Linksys incompatibility.

Re:Glad I disabled auto-updates

[nosfucious]

Good point.

Don't just disable it, remove it from your System. It's just another networking service and it can be un-installed.

Although, as the parent poster mentioned, it's not beyond Microsoft to re-install it as part of a Service Pack/Security Update. (See Windows Messenger).

Re:NAPT != Firewall == Bullshit!

[Anonymous Coward]

It is not meant to protect computers on either end from each other.

Well, you can argue that if you wish. But, there is no arguing that NAT is the single most effective firewall to date. Without port forwarding rules and uPNP, NAT COMPLETELY blocks inbound session initiation. It also quite effectively hides internal hosts making attacks against an internal host even harder.

Re:Man-Denis!

[Anonymous Coward]

Patrick?

Billion is wrong here...

[Anonymous Coward]

This article I think is miss titled. As you can see if you read billions notes about the upgraded firmware, Billion not XP3 is to blame. From their document "After detail analysis, we found that Windows XP SP3 sent out the DHCP packet with the
Option 43 data (include Microsoftâ(TM)s âVendor Specific Informationâ(TM)), but Windows XP SP2 sent
out the DHCP packet without the Option 43 data. However, the Option 43 data is not
compatible with Billionâ(TM)s original definition, so it will cause this problem. The affected firmware
versions of BiPAC 5200 series are 2.9.8.x and 2.11.0.x~2.11.33.x. There is no impact to BiPAC
5200 series if the firmware is 2.10.x.x. Please check Appendix A for checking your current
firmware version."
Option 43 data is optional based on the standard (http://www.faqs.org/rfcs/rfc2132.html) it should not be causing the router to crash. Based on what I'm seeing I would think any other operating system and/or setup that sent option 43 data to this router would cause the crash.

I'll bite

[FoamingToad]

SP3 sets a system restore point before installing.

I know this as I trialled the beta version on my games machine - and it bluescreened due to a wonky driver (I believe the chipset driver for my mainboard).

Booted to safemode and rolled back. Easiest recovery in ages.

I'm interested to know what caused the hassle with the laptop to the extent that it took you two days to recover.

But all in all, sensible advice. Backups are a good thing (tm).

F_T

Re:I'll bite

[Max_W]

Difficult to tell. I tried your approach but it did not work. I raked the Internet with search engines and tried about everything related to SP3 blue screen of death. Ten times each.

I have the 160 GB back up hard disk. So it was not a complete catastrophy for me. I lost only my private e-mails. Work e-mail is IMAP. All work related files were backed up.

Still I fear that if MS makes it automatic update it may knock out my notebook again.

I reinstalled the Windows XP SP2 and the ASUS notebook works normally.

Re:In defence of linksys

[FoamingToad]

I've had a different experience with their kit. Eight years with a BEFSR41 (which I sold to a friend last November, and is still going strong) which I only passed on because I wanted additional ports. I've now got a BEFSR81 which is a nice bit of kit. Only six months, but it's happily working on. Well worth the £35 I paid for it. Only disadvantage to me is that they're quite hard to get hold of in the UK, and they don't support Gigabit ethernet (I therefore keep a cat6e xover cable for big transfers with the server - thankfully this is fairly rare now). I've also had a WAP54G which is OK for a couple of years now. Again, quite happy with its capabilities. For home use, I'm fairly happy with the Linksys kit. Seems stable and I've recommended them to a few people without having them bite me in the a$$. By contrast, the couple of no-name brand routers I've picked up from fairs (because I've been pushed for time) seem to be a lot more prone to wonkiness.

SP3 is *NOT* ready for Primetime

[mujadaddy]

SP3 has, in my brief experience with it:

*Prevented upgrading from Windows Media Player 10 to WMP 11 (Shut up -- Netflix streaming needs '11)
*Tricked a game I was trying to install into thinking there's *NO* DirectX installed.

I rolled back to SP2, and both of these issues were nonexistent.

why Billion router crashes with XP SP3

[hukketto]

XP SP3 sends out DHCP packets with Option 43 data (including Microsoft's "Vendor Specific Information"), which Service Pack 2 (SP2) did not. However, Option 43 data is not compatible with 5200 series router's original definition, which causes the problem (as reported from a notice from Billion)

RFC2132 says: "Servers not equipped to interpret the vendor-specific information sent by a client MUST ignore it (although it may be reported)."

It seems to be a Billion's fault...not Microsoft's

Re:NAPT != Firewall == Bullshit!

[Luke-Jr]

That it blocks anything has nothing to do with it being NAPT. It would be perfectly valid for a NAPT to treat unknown inbound connections specially and broadcast the attempt to all known internal IPs-- connecting it through to whichever responds positively first (if any).