Slashdot Mirror
Vista Bug Costs Users In Swedish Town Their Internet
Lund, Sweden refuses to work around a Vista bug, so people who live there must choose between Vista and internet access. It's nice to see the right people being held accountable for a change.
Is if the city offered free Ubuntu CDs as "Windows Upgrades."
Ubuntu is an upgrade from XP and Vista.
Keep that shoe on the other foot for just a little longer. Imagine them having "support scripts" that travel through a KDE interface instead of Outlook Express or IE. Imagine them requiring Ubunto to install your access. In short, imagine all of the "standardization" Windoze enjoys being flipped on you.
In the free software world, users can edit a few well annotated text files to get the job done if they are given the proper information. That task is harder in Windoze because you must dig through several GUIs that don't tell you what to ask for in advance or ever.
It's a shame that ACs can post with more points and more frequently than Twitter.
Their internet is b0rked?
Blearf. Blearf, I say.
Wouldn't using a router to connect to the internet bypass the bug?
Fucking tricky one, eh?
Like choosing between an anal probe and a cream bun.
Lundis Energi should have been testing Vista back in its early alpha release stages to ensure compatibility with their Linux based server system
Why? If their existing system follows the appropriate standards, why should they have to test someone else's future product to check compatibility?
If this happened in my town--and if I were using Vista--I'd be pretty damn unhappy. Usually a story is funny because someone got what they deserved in a particularly humorous way, or because someone subjectively considered evil takes it in the pants. Here I see a bunch of people getting shafted by two corporations that don't want to play nice, and this perhaps for the crime of simply owning a new computer.
The university I work for in Sweden began testing with Vista when it was the called Longhorn. We discovered some bugs with the communication between Vista and some of our servers (running Solaris) back in 2004. The bug repports were submited to MS back then and the thing was fixed on the next Longhorn beta "release". It seems it's easier for some not to test and cry out like a baby when it's too late.
It's time to realise that Abble's products are the biggest abomination these days. Just say NO to the dumb iAbble way!!
Between the DRM, the DirectX 10 incompatibilities, the high cost, the extreme hardware requirements, the inability to play multimedia and download at the same time, and now this, I just don't understand why anyone would still be using Windows. And it's not like there aren't alternatives. Between even just OS X and Ubuntu, there are systems that don't have such inherent flaws.
Granted, its not on Linux but I've been using the Squid proxy server for a long time now and Vista never had any issues with it. It can be a little tricky since Vista won't recognize the Internet connection, only the local connection to the proxyserver. But apart from that its not rocket science to get it to work. So guys; better get someone who understands what he's doing.
The problem as reported is that the Vista DHCP client fails to obtain an address from Linux servers running (I'd presume) ISC dhcpd.
When I bought a laptop recently it came with Vista. When I connected it to my network it failed to obtain an address. I assumed there was some misconfiguration problem I was missing, Turns out it's a fundamental difference between the DHCP client in Vista and the one in prior versions of Windows. See this item from Microsoft: http://support.microsoft.com/kb/928233/en-us.
The version of dhcpd I'm using is an old one (2.0). I thought about upgrading it to see if that would solve the problem, but since I wasn't planning on keeping Vista on the laptop, I didn't bother upgrading. All our other machines run Linux and don't have this problem.
I wonder what decision will be made in enterprises running Linux DHCP servers that introduce Vista into the workplace. Will they follow the Microsoft KB item above and "fix" the problem on every new Vista box they buy? Or will the replace the Linux DHCP box with Windows Server?
.. actually the city in sweden with most students per capita, since lund university is located there. If anyone is willing to adopt to linux or just bash windows it's young people. This is probably a big issue down there but so far I haven't heard anything about this over here, and I'm about 150km away.
Vista sets the DHCP BROADCAST flag.3 3
http://support.microsoft.com/default.aspx/kb/9282
This is in compliance with DHCP standards.
Ofcourse the incompetent Admins will blame Vista and not fix the router software.
Now, I'm all for requiring microsoft to fix the bug rather than mutilating a linux server setup (I was once asked to cripple the network of a EUR250K linux computer cluster because one of the professors in the university in question preferred Windows XP, which just wouldn't talk IPv6 properly - fortunately he eventually listened to reason, though it took a demonstration of the problems his choice would cause for other users (most particularly, CompSci grads doing IPv6 research projects ?!) to make it happen), but the admins for the town should have said "we have advised microsoft of the problem", not let a microsoft rep say "Tut tut. If they had only contacted us...". Microsoft employees are quite often technically truly incompetent (n.b. my bar for "competence" is quite high, but I think its fair - as a pretty much linux-only geek since ~ 2000, I STILL know more about windows internals than most microsofties I encounter), but are masters of spin.
It's nice to see the right people being held accountable for a change.
/. , so the majority of folks will be going "HAHA ST00P1D MICRO$OFT LUSERS PWNED LOLZ!1!!eleventyone!"
Nice bit of flamebait there.
Yeah, I know it's
Once everybody gets that out of their system, IMHO Lundis Energi is really being a bunch of assholes, and I have no sympathy for them, as it makes them seem like a company run by a bunch of 15-year-olds who've just discovered Ubuntu.
They find a bug (or rather, the users did) in newly-released software that doesn't play nice with their Linux-based server. Rather than you know, cooporate with Microsoft to help diagnose the problem, they're essentially saying to their users "We think you're a bunch of losers (LUSERS HAHA!!1!), so ya'll use the OS we want and tell you to use! If you don't like it, kiss our asses! And Micro$oft can kiss our ass too until they fix the bug!"
Because, after all, bugs never, ever happen on any software ever, and developers psychically know what exactly a bug does without any reporting by end-users whatsoever.
Now if they have a policy of "NOT Windows Vista compatible right now" clearly stated to their users, then that's understandable. But eventually, most folks will move to Vista (like it or not), so this bug needs to be squashed on whomever's end.
I'll end my rant with this:
how in the hell is Lundis Energi so sure it's not a bug on their software?
(sarcasm) Oh right, it's Microsoft, so it MUST be them. (/sarcasm)
I do tech support and when Internet Explorer 7 came out we noticed that it didn't really get along with the NAT routers we send out to our customers (they sometimes need to do a very very small amount of configuring), I'm not entirely certain of what the problem is but there is no problem with IE5/6, FF, Safari, Opera or even links, but IE7 is a no-go. It took the manufacturer a good three months to come up with a new firmware that addressed the problem, and until then we had to teach hundreds/thousands of customers how to use telnet (and how to install it if they were running Vista, the telnet client is disabled by default). Good times...
Oh well, at least it's not Windows 9x, I have to give MS some credit for eventually killing off all support for that branch as our superiors decided that since MS no longer supported 95/98/ME in any way then neither should we. :-)
/Mikael
Greylisting is to SMTP as NAT is to IPv4
"It doesn't work" has never been a useful comment.
Also, I don't see why an ISP should test every OS version to check if it's compatible with their network. I thought we all used the TCP/IP standard for internet stuff. And if Vista had a broken TCP/IP implementation, then why is this the first report about this? What makes this ISPs infrastructure so different?
One, relatively strong Monopoly (Microsoft) gets screwed in a small town by another absolute monopoly. Think about it, they are the only ISP, so they are a monopoly and they are using that to damage a company in a completely separate market. Anyone who makes out that this is good is completely wrong, just because someone isn't Microsoft doesn't mean its a good thing that they screw over the consumers.
The article seems very light on details. Does anyone know what the actual problem is?
"Welcome to our world. We are the wasted youth. And we are the future too." Yes, I know these are stupid lyrics.
Windows XP is the world's de facto standard for O.S. Not Vista. How come users in sweden can access the internet but those with Vista don't?
The answer here is that Microsoft probably took the decision to break the TCP standard on purpose, hoping the admins would work around the bug er... new standard.
Personally I applaud the decision of the sweden admins. Microsoft must not be allowed to gain control of the market by breaking even more standards.
I just read above that the DHCP flag is part of the standard.
Nevermind then. (blush)
I'm using a couple of Vista boxes on my local LAN with a home brewed CentOS router running ISC dhcpd 3.0.5. No problems with obtaining an IP address at all. Sounds like the flaming is misguided this time for a change. Perhaps Lund is using an ancient version of dhcpd?
Both of the english articles listed in this slashdot-post says that Lundis Energi has no desire to do anything. However, in a Swedish newspaper http://www.metro.se/se/article/2007/08/28/14/2423- 48/index.xml they say:
"Our technicians are looking in the matter to see what we can do but it is mainly up to Microsoft to fix this issue" /Åsa Holmander, product manager at Lundis Energi (rough translation)
This is another example of how Vista has better security than previous Windows releases. It won't let you connect to the internet, by design. Another problem solved!
Glad to see an ISP do the right thing and not pander to buggy software.
I don't think this is an accident. I don't know whether Microsoft deliberately created this "bug", but you can bet that they deliberately didn't spend a whole lot of money on compatibility testing with Linux.
Furthermore, there's a good chance that many people will blame the Linux servers and the company that deployed them, rather than Windows Vista. And, in a sense, the company is responsible: they should have tested Vista clients long before the Vista release (that's why Microsoft has beta programs). And the resolution to this problem may well be that they switch to Vista servers, because they are, in the end, a business with customers.
So, people, you may be gleeful about this, but this sort of thing ends up hurting Linux, because the Vista clients are a given and they are not going away, and Linux will be remembered as the "OS that caused all the trouble".
It's a feature.
So, for the densest concentration of Swedish students in the country, their perception of Vista is "broken/doesn't work/incompatible." That's a serious problem! There's no telling how many future sales this will prevent both among current consumers and future sales, when these students will be out of university and in corporations where they'll influence (and eventually become) decisionmakers. Besides, how much time/money was (and is being) spent to deal with this bug?
Hey MSFT, what's the Total Cost of Ownership now?
It's a feature!
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
Mod Parent Up!
Bad news guys; Microsoft isn't the one with a bug causing the problem. Poor implementation yes, bug no.
For some bizarre reason Vista expects the address returned from the DHCP server to be broadcast, instead of sent via unicast packet. This is permitted in the specs and supporting the broadcast flag on the server is suggested. ("SHOULD", not "MUST" in the spec.).
When researching this I found 2 network types which required this, Infinibad and 1394 (Firewire). It looks to me like Microsoft picked the one which would (theoretically atleast) work on all network types, instead of only on a few.
Of course, this is a typical bad decision as it means that responses from a DHCP server with a lot of Vista clients will flood the network with broadcast responses, but hey, they arent know for making good decisions.
This implies that it's Vista refusing to interoperate with Linux, which obviously would play into story submitter twitter's frequently-espoused odd Microsoft conspiracy theories. In actual fact, it's a recognised bug acknowledged by Microsoft as being due to old routers or DHCP servers which do not support the DHCP broadcast flag (a formal part of the DHCP RFC standard).
Solution: upgrade your damn DNS server. This ISP are just acting like petulant teenagers, unwilling to do anything to fix something which is, at the end of the day, their problem (i.e. their non-standard DHCP implementation).
By summer it was all gone...now shesmovedon. --
RESOLUTION
Warning Serious problems might occur if you modify the registry incorrectly by using Registry Editor or by using another method. These problems might require that you reinstall your operating system. Microsoft cannot guarantee that these problems can be solved. Modify the registry at your own risk.
To resolve this issue, disable the DHCP BROADCAST flag in Windows Vista. To do this, follow these steps:
1. Click StartStart button, type regedit in the Start Search box, and then click regedit in the Programs list.
User Account Control permission If you are prompted for an administrator password or for confirmation, type your password, or click Continue.
2. Locate and then click the following registry subkey:c es\Tcpip\Parameters\Interfaces\{GUID}
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Servi
In this registry path, click the (GUID) subkey that corresponds to the network adapter that is connected to the network.
3. On the Edit menu, point to New, and then click DWORD (32-bit) Value.
4. In the New Value #1 box, type DhcpConnDisableBcastFlagToggle, and then press ENTER.
5. Right-click DhcpConnDisableBcastFlagToggle, and then click Modify.
6. In the Value data box, type 1, and then click OK.
7. Close Registry Editor.
So Vista isn't (formally) going counter to protocol, it's just going counter to a 15-year old custom. Nonetheless, Vista *can* cooperate, it just needs to be told not to raise the DHCP BROADCAST flag. And yes, that route goes via a registry modification.
In summary: a tropical storm in a teacup.
From the article, even in Swedish, it makes it clear that the town doesn't want to cooperate with Microsoft on providing data for the bugfix. The accountable party here, then, is the town internet provider and not Microsoft.
[Town]: Our internets doesn't work with Vista
[Microsoft]: Okay, do you have any data on why not?
[Town]: no but it's your fault, fix it!?!?
[Microsoft]: Well, what's even a short description of the problem? Side effects? Can your Linux server be changed to alleviate it in the meantime?
[Town]: THE INTERNETS IS BROKEN, FIX IT THOUGH OKAY!!!!????
Yeah, all Microsoft's fault. If this was on Mozilla or Novell or Linux bugzillas it would have been closed as "irrelevant".
And counter to Microsoft's last 4 operating systems.
They got it right back in 1995 (12 years ago)
Nope. Just another example of how Microsoft does not care about published standards. Their DHCP services can handle it so why should they spend any time understanding the standard that the rest of the world follows?
After all, everyone else will probably change to support Microsoft's weird implementation. Who cares about the problems that the users have in the meantime? If Microsoft is lucky, no one will be able to explain the problem in terms those users could understand and the rest of the world will be blamed for the problems when it is Microsoft who is not following the published standard.
Windows Vista correctly implements the DHCP standard BROADCAST flag,
which some poor-quality software mishandles.
The RFC recommends, but doesn't require, implementation of the BROADCAST flag, so Vista should do something reasonable when the server doesn't support it.
Furthermore, an RFC isn't a standard, it's a "request for comments".
Even if it were a standard, so what? That doesn't change the fact that Microsoft made a change that they must have known would break things. Standards-conformance can be used for deliberately screwing users and engaging in monopolistic practices just as much as standards non-conformance (i.e., Office XML won't be any less monopolistic even if it becomes a "standard").
Finally, there is no problem with Linux software quality; it implements this feature just fine. The Lundis servers just seem to have disabled it, or maybe they are using an old version. Or, heck, maybe they are running NT 3.51 and simply lying about it.
"it's a recognised bug acknowledged by Microsoft as being due to old routers or DHCP servers which do not support the DHCP broadcast flag"
..
...
Trust Microsoft to make Microsoft DHCP incompatible with everyone elses
"Windows Vista cannot obtain an IP address from certain routers or from certain non-Microsoft DHCP servers"
Rest of ad hominem ignored
was: Re:Nice going, twitter.
davecb5620@gmail.com
If nobody there has internet... How the hell has this ever reached the front page?
Never thought my little obscure home town would get on Slashdot :)
Regarding the matter at hand, I actually live like 200 meters from the Lunds Energi company mentioned in the article but haven't heard anything about this issue until now. Something tells me that this has been exaggerated. The whole city is blanketed with free wifi anyway, so it's not like we're dying or anything.
Oh, and if you need to know anything about blondes, beer or vikings, feel free to ask ;)
It's not a bug in Vista... they just turned on a flag that is already defined in the standard. The problem is that many many servers didn't implement the standard fully. Who's fault is that? I'd say it's the software that doesn't fully support the standard. Guess what...that's not Vista in this case. Vista has some other network problems, but this isn't one of them.
Cisco's IOS has had the option of turning on the DHCP broadcast flag on its router client since v12.2 and I'm guessing (can't find a specific reference) that the Cisco IOS DHCP Server probably has also supported it since then... so it's been supported by the major network router manufacturer for many many years.
The D-Links and the Linksys' (yeah, I know they're Cisco now) routers don't support it either, so it isn't just Linux DHCP servers
Karma: Excellent. 15 moderator points expire sometime.
I think the whole point of the post you replied it is that it actually *does* follow published standards.
Did the author already forget that Vista was almost a different OS until Beta 1, and especially Beta 2...? Testing it under the Longhorn Alpha would hardly have gave them anything interesting. It was very, very different, and very very bugged. That was behind the delay of Vista too -- Microsoft scrapped large parts of the Longhorn client code when they moved to beta. And the time period between beta and RTM for Vista was probably shorter than necessary, again likely rushed since Microsoft had messed up during the development so much that they badly needed the OS out at some point soon. Which we're now paying for in the form of incompatibilities and poor performance until maybe SP1 the next year at the earliest.
Beware: In C++, your friends can see your privates!
I have a hangover, a terrible one. I read both articles and have no clue what the bug is. Really.
I run a linux based network and have a vista machine accessing the net just fine, with transparent proxym dhcp, nat and everything coming from a linux sever.
If my ISP refused to fix a presumably widespread problem preventing me and a significant number of other customers from obtaining a lease from their DHCP servers on the grounds of principle, I would find another ISP. Deal with your principles on your own time. When I'm paying you for a service, I expect you to fix a problem if it's realistically doable and economically viable. Which this obviously is.
That the Vista DHCP client actually does follow standards just makes it all the more absurd, and if I was affected, I'd promptly find a new ISP.
loser
Apparently it is geek paradise. Liberals too.
Read radical news here
You forgot WinME. :)
That's what a love about windows. No need to deal with messy config files. It's all right there in the registry.
People keep posting about the fact that M$ broke the TCP/IP standard (not even close!) or the dhcp standards. The broadcast bit that the MS KB article talks about is part of the RFC. People need to read more before they start talking about things they don't fully understand.
http://faqs.org/rfcs/rfc2131.html - Section 4.1 (Broadcast Bit)
My $0.02 CDN
"Well you're not Fiona Apple, and if you're not Fionna Apple, I don't give a rat's ass."
Lunds Energi is a small ISP operator in Lund (they are the power company). They have a small fraction of the market and most of the city is still online. By far the largest provider is Telia (former state incumbant), followed by a host of other players (tele2, perkspektiv bredband (~30 EUR/month for 100mb ethernet), bredbandsbolaget, etc.).
To say that the entire city is offline is almost comical. Even assuming Vista has a 20% market penetration, and that Lunds energi has a 5% market share in Lund, it is at most 1% of the city.
Wow,
when you have to drop to a command-line under Unix the Windows-fanboys cry fowl, but then they put up with this quick and easy solution???
Hypocrites as bad as they get..
For starters theres a bunch of windows guys on here going "oh, vista is performing as per the rfc" - which really makes me laugh and think "MS did something on a standard for once? unbe-f**king-lievable" and how the ISP is the problem and on the other side you have linux guys going "oh here's another broken MS implementation" without even reading what people have actually said about the issue.
You are ALL, everyone of you (who responded with anti or pro windows sentiments), retards. For starters the article is so light-on with details that any conclusions you draw can only end with "if we knew more..". Any of you that came up with "Microsoft are to blame" or "Lund are to blame" should never be allowed to post - EVER AGAIN!
Has the community become so divided here in the pro and anti MS camps that you cant see the forest for the tree's? Get a clue. Half of you didn't even bother reading prior posts before your own replies - for that you should also be ashamed of yourselves.
You're saying that a business, a school, and a grandmother of 74 having to go in and edit the registry on their new PCs in order to access the Internet because Microsoft did not want to follow protocol is a tropical storm in a teacup? Are you kidding?
From the KB article you cite (proving Microsoft is aware of this problem and not fixing it for its users):
Warning Serious problems might occur if you modify the registry incorrectly by using Registry Editor or by using another method. These problems might require that you reinstall your operating system. Microsoft cannot guarantee that these problems can be solved. Modify the registry at your own risk.
This has been done right for more than a decade. What the hell is Microsoft's problem? Why is it broken now? And why won't they fix the flaw in their operating system?
Looking at the date of that KB article, Microsoft has known about this since March 15, 2007. So where is the patch? Do we have to wait for SP1?
This place is getting worse by the week.
It almost sounds like a Linux "A feature not a bug"(tm) :-).
/etc/thisdamnconfigfile.conf with vi.
:q!? What's the? Stop Editing!
:forcequit :DearMisterViIreallyWantToLeaveYou :letmeoutyoupervert! :helpVItrappedme man vi reboot Emacs"
IE:
Of course there is no bug! You just have to open
change the "DearGodPleaseMakeSureIWillBreakNothing" flag to 0
Close the file. Kill the daemon and restart it.
In the real world:
Oh my God how does this text editor work? Insert not Delete! How do I save eh?
kill thedaemon
daemon restart
Error line 26458: : unrecognized command ":q!
I've had no problem using Vista to obtain a DHCP IP address on my Linksys WRT54G v6.
From rfc-1542 that regulates the use of the broadcast flag:
And from the same rfc, a section that defines certain terms, including SHOULD
(also later defined in rfc-2119)
These two documents show that although the use of the BROADCAST flag is standard compliant (and we could argue that the DHCP server involved SHOULD try to answer them), the implementation of a modern DHCP client on a system capable of receiving unicast responses using the BROADCAST flag is not (at least without a very good reason to do so).
Considering that Microsoft has already made DHCP software capable of receiving unicast DHCP requests, and seeing that the fix proposed by Microsoft themselves is a simple change in a registry entry (meaning the software is in fact capable of receiving unicast responses), I'd say the software is not compliant.
I consider that behavior non compliant unless Microsoft had a very good explanation (the possibility of Vista running on a pc not capable of receiving unicast responses not a good one, since the flag should have been set the other way). It is at the very least gross imcompetence, if not outright malicious behaviour.
GPG 0x1B479C78
Lund is/was HQ for SonyEricsson... wonder how they're doing?
It just wasted a post instead.
It's dead on two points...
1 hosts file blocks it
http://www.mvps.org/winhelp2002/hosts.htm
2 DNS blocks it. DNS 1 67.138.54.100 DNS 2 207.225.209.66
http://www.scrubit.com/
This makes many of the troll posts safe for work.
The truth shall set you free!
A Troll is similar to Flamebait, but slightly more refined. This is a prank comment intended to provoke indignant (or just confused) responses.
This one is a pretty low bar, I'll grant you that. One can be indignant of a number of things. I'll leave whether or not it confuses the individual with the mod points up to that individual, but again, this isn't particularly hard to do.
A Troll might mix up vital facts or otherwise distort reality, to make other readers react with helpful "corrections."
The OP is on topic and does anyone here think that MS being used by a large percentage of the population is incorrect? The last paragraph is opinion, neither more or less inflammatory that 95.398% of the posts here, including this one..
Trolling is the online equivalent of intentionally dialing wrong numbers just to waste other people's time.
I really don't get that impression from the OP. Maybe the word "dickish" upset some of the more highly refined members of the forum. I suppose we could consider adding the tag "Adult Language" for these folks so as not to upset their delicate sensibilities. I find it really ironic that anyone could mod a comment troll given the submitter of the thread.
There, I feel better. Much more satisfying than metamoderating.
Faster! Faster! Faster would be better!
It's ironic how often people use words like "accountable" and "responsible" when they're trying to pass the buck.
"Screw Sun, cross-platform will never work. Let's move on and steal the Java language." - Visual J++ Product Manager
``Who's to blame.
Well, both actually. Lundis Energi should have been testing Vista back in its early alpha release stages to ensure compatibility with their Linux based server system. On the flip side, Microsoft has to get into gear with increased interoperability between Vista, Linux, XP, OSX, FreeBSD and every other operating system choice available on the market today.''
I don't agree with that assessment. It is not up to Lundis Energi to do compatibility with upcoming operating systems that they themselves do not produce. It is up to them to implement standards, and it is up to Microsoft to do the same. If either side breaks the standard, they are to blame.
Unfortunately, none of what I have managed to dig up with Google actually explains what the problem is. It seems, however, that this is because the people from Lundis Energi refuse to provide details. No matter on whose side the bug is (and, honestly, it could be on both sides), refusing to help resolve the issue is decidedly unhelpful (not to say boneheaded). If you discover a bug, the right thing to do is to report it, and perhaps suggest a workaround. You don't go screaming bloody murder and demand that the bug be fixed, while simultaneously refusing to provide help in fixing it (which Microsoft has asked for in so many words).
So, in as far as I can judge from my couch over here, Lundis Energi is definitely doing something wrong. Microsoft may or may not be doing something wrong here, but without further details, I can't tell.
Please correct me if I got my facts wrong.
Whenever I see a story like this I'm always curious as to who is really at fault.
The question is, what does the broadcast flag *do* and should servers be required to implement it?
From some web page:
The "broadcast flag": DHCP includes a way in which client implementations unable to receive a packet with a specific IP address can ask the server or relay agent to use the broadcast IP address in the replies (a "flag" set by the client in the requests). The definition of DHCP states that implementations "should" honor this flag, but it doesn't say they "must". Some Microsoft TCP/IP implementations used this flag, which meant in practical terms, relay agents and servers had to implement it. A number of BOOTP-relay-agent implementations (e.g. in routers) handled DHCP just fine except for the need for this feature, thus they announced new versions stated to handle DHCP.
So servers "should" implement it but are not required to. But now there are a few obvious questions:
1. What are the benifits of using broadcast over unicast? If the client can accept unicast and unicast is somehow "better" then why are MS client's sending the broadcast flag in the first place?
2. If servers are not required to support the broadcast flag, is there a alternative protocol sequence that will allow clients that desire the broadcast behavior to successfully obtain an IP with the server? If the purpose of the broadcast flag is for clients that incapible of receiving unicast responses, then it seems that client could not possible negotiate successfully with a server that does not support the broadcast flag.
3. Ultimately, the question will be, should DHCP server implement the broadcast flag?
So people complain when Microsoft doesn't follow RFCs to the letter, but when they're fully compliant, people.. still complain?
Stop moaning. The DHCP client SHOULD not do this, but then again, the DHCP server MUST accept these requests anyway.
Any RFC standard where the client SHOULD have a certain implementation dictates that the server MUST accept anything within the standard, including clients not doing what they SHOULD do, while still adhering to the standard.
The issue is with servers not adhering to the standard. Trying to excuse server non-compliance with esoteric, standard compliant client communication is laughable. Especially coming from people historically obsessed with standards compliance.
Oh, OK... maybe it's older linksys models.. I saw D-Link and Linksys mentioned on various google searches... I'm all Cisco ("real" Cisco) at home so I couldn't try it
Karma: Excellent. 15 moderator points expire sometime.
That's not a bad thing here. Microsoft is generally a dick without principle.
Microsoft, however, does. And the only way to get through to Microsoft is through their end-users -- or maybe their actual customers.
Actually, no.
Generally, when it's the other way around -- that is, when some open-source project can't communicate with something standard-compliant -- well, first off, pigs are flying; this just generally doesn't happen.
But also, we fix it. We don't run around screaming and blaming others unless there is a reason to.
Example: If it's actually a bug in, say, Firefox rendering, we fix Firefox. However, if someone deliberately sends the wrong page, or even just an "access denied" page, to Firefox users based on nothing more than a user-agent string, then we pull out our user-agent switchers and pretend to be IE -- and we also bitch loudly.
Take a look at the shit the Wine project has to do, on pretty much a daily basis, just to get Windows programs to run. They can't even write to Microsoft's standard, because Microsoft doesn't, and application developers don't -- Microsoft writes whatever they felt like that day, developers work around that, and Wine gets stuck having to reproduce "bug for bug" compatibility.
So in general, no, the community does not usually have an attitude of "obey the standards or we won't cooperate." Perhaps we should. I know I often have a mind to block users on IE6, at least, and maybe IE7 also, so I don't have to do extra work to support them.
Don't thank God, thank a doctor!
This kind of thing is the attitude that gives Linux users a very bad name. Who'd run Linux on the desktop or even servers at work if it refused to interoperate (despite the specs saying it should) with everything else on a pathetic tecnhicality that's not even incorrect!
If anyone wants Linux adoption to continue, they need to lose the "/us l33t, j00 sux0rs" attitude.
Chances are that you are just ignorant, but we have formal standards for a reason. Not following the standard sounds very much like their famous EEE strategy. Something I think most of us don't want at all. So I am actually quite happy with this because it is M$ that should fix this.
Fresh from the RFC 2131 (DHCP):
...and thats exactly what happens....
:-)
A server or relay agent sending or relaying a DHCP message directly
to a DHCP client (i.e., not to a relay agent specified in the
'giaddr' field) SHOULD examine the BROADCAST bit in the 'flags'
field. If this bit is set to 1, the DHCP message SHOULD be sent as
an IP broadcast using an IP broadcast address (preferably 0xffffffff)
as the IP destination address and the link-layer broadcast address as
the link-layer destination address.
So the linux dhcp server SHOULD reply on the broadcast address but he does not. That behavior ist STILL adhering to the standard (it is NOT a MUST in the rfc)....
Out of the same RFC (talking about the client):
The TCP/IP software SHOULD accept and
forward to the IP layer any IP packets delivered to the client's
hardware address before the IP address is configured; DHCP servers
and BOOTP relay agents may not be able to deliver DHCP messages to
clients that cannot accept hardware unicast datagrams before the
TCP/IP software is configured.
So Vista is not accepting unicast datagrams as they SHOULD do. So they are adhering to the standard as well. But here the rfc warns that if you not accept unicasts you may not be able to receive ip-addresses.
Summary: Both sides adhering to the standard. The linux server is missing a feature that should be honnored, the vista clients relying on a feature where they have been warned that it might not work. Or to say it shorter: Vista gets what it deserves: No IP-Address
Right people here are discussing RCFs and wonder what is going on, well I live in Lund and here is my take on what has happened:
a)Per the RFC servers do not need to implement the broadcast flag, but it is a good idea if you want to support systems that use it.
b)Per RFC Vista doesn't need to clear the broadcast bit, but it is strongly recommended and setting it is intended for legacy clients only.
c)Lund's energi's network doesn't support the broadcast and thus Vista machines do not get an IP over DHCP since they set the broadcast bit.
d)For reasons we don't yet know, Lund energi won't implement a workaround on their server. I don't know enough about DHCP or their systems to tell why, so I guess there might be a technical issue or perhaps they are just being jerks.
e)The fix is to set a registry key, which is easy for technical users, but a pain for those who don't know about it.
My judgement is that Lund's energi has a shitty DHCP server and Vista is a shitty DHCP client. Since the fix is so simple ( adding a registry key ) this really ought to be a non-issue, but because Microsoft and Lund's energi are both incompetent crappy companies the end user is left with a problem that would actually be rather easy to resolve. Those in the know can work around it, but non-technical users are left without service while those responsible point the finger at one another. The sad thing is that this really isn't particularly surprising. Hmm, did I forget something? Oh yea, the article summary is wrong since there are scores of ISPs in Lund, and this only affects one of them. So yea, I'm not very surprised at all...
The official problem seems to be described here:r mation/2007-04-19%20Problem%20med%20Microsoft%20Vi sta
//fatal
http://www.lundsenergi.se/oppetstadsnat/driftinfo
For those of you who don't understand swedish, here's my translation for you:
Customers with Microsoft Vista installed has problems getting an IP-number assigned by the city network. This means it's not possible to access Internet or other services. We are working as hard as we can to solve the problem.
_If_ anyone has heard from an official source that Lund Energi is *not* working on solving the problem and blaming Microsoft - Please post about where you got this information from. (I'm guessing the whole thing probably originated from some angry Vista-using customer who made up false accusations and the rumor got started...)
Discouraged doesn't mean OPTIONAL. The DHCP server doesn't meet the specification, therefore is broken.
One single city's servers trigger a "bug" (if it really is one).
The one(s) responsible for those servers don't disclose (even to MS, apparently) the details about what the bug is or how it conflicts with their internet access.
At the same time they also refuse to fix the problem at their side - something that should be easy, given that the problem doesn't manifest itself anywhere else in the world.
If you ask me, something stinks in Lund.
It does remind me of a problem I had in early XP betas, that turned out to be my ISP making activation attempts fail (in this case the problem was fixed before XP went gold).
I never found out what the exact reason was, but I suspect it was their blocking of certain ports at the router level, including port 80 to enforce use of their proxy, several other outgoing ports, and all incoming ports below 1024 to withhold people from running servers on their broadband connections).
To make things worse, their proxy was so misconfigured that half of the online stores and a bunch of other sites would only work if you used https.
It wouldn't in the least surprise me if it turns out that something similar is going on here.
Or even worse, that someone in Lund found a real bug (in DHCP for example), changed the server's behavior to trigger it, and is now singing "ne-na-ne-ne-nah, I fscked microsoft" (that would explain why it's only happening there, why they're not too keen about letting out the details, and why it took more than half a year before we heard of it - if it had been like this from the start we ought to have read about it last December).
And finally: IMO it's simply not done anymore to hook up a desktop computer directly to an internet connection. At the very least you use a small NAT router to keep out incoming connection attempts (at least those you don't expect). And with such a router, the server wouldn't see a Vista machine and the connection would work.
Hmm, and I thought getting stuff to work by having to edit a text file or copy&pasting something into the commandline was annoying.
Actually, this is typical. GUI config things are almost always much more complex and difficult to use than just editing a text file. But there has been a massive propaganda effort to convince people otherwise, and people fall for it.
Back in the 70s, when unix first came out, one of its widely-touted improvements over other computer systems was that you could configure everything with a single tool, a text editor. All the commercial systems had been moving over to fancy full-screen interactive stuff, which was done on character-based terminals, but were otherwise similar to the current GUI stuff. Every package had its own config tool, every such tool was different from every other, and you had to spend a lot of time learning each of them.
If you had to make a change 6 months later, you wouldn't remember that config tool, so you'd have to spend days relearning it just to change a couple of numbers that were hidden somewhere in a confusing mess of screens linked together in mysterious ways. And if you upgraded a package, the config tool would change, so you'd have to spend days relearning it. But unix-based packages mostly had a plain-text config file, editable with any editor. Conventionally, config files had lots of comments and examples of how to do common things. You could quickly scan a config file, searching for keywords, make the changes, write the file, restart the daemon if necessary, and a few minutes later, it'd be done. Sorta like how the apache server works now, y'know.
It is interesting that people fall for the complex GUI approach to configuring things. You'd think people could figure out for themselves that there's a much easier way, at least after they've seen a few editable config files. I wonder why people are so thick-skulled about such things?
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
But I'd be pretty pissed if my ISP inconvenienced me, the consumer, because of their idealogical objections to Microsoft. Bottom line: they have to have realized that many of their customers would eventually run Vista, yet they made no effort to accomodate those users. The fact that the bug is Microsoft's is 100% irrelevant.
So are you seriously trying to tell me that you don't think that a server is required to implement features that the standard makes available to the client?
Do you really believe that such a well-established standard is that inconsistent?
"Vista SP1 will remove bugs and close security holes provided in the original Vista Release. Any new holes and bugs SP1 creates will be smaller. Or at least different. We think." ---Unofficial description from Microsoft concerning Vista SP1
A wise man once said, "Where is my other quotation mark?
WARNING! Parent link lead to a Steve Gibson article! WARNING!
If you were blocking sigs, you wouldn't have to read this.
You can do all that from the command line; it's just that on Windows, you don't have to.
If you were blocking sigs, you wouldn't have to read this.
So Vista isn't (formally) going counter to protocol, it's just going counter to a 15-year old custom. Nonetheless, Vista *can* cooperate, it just needs to be told not to raise the DHCP BROADCAST flag. And yes, that route goes via a registry modification.
In summary: a tropical storm in a teacup.
Yeah, and it's fine anyway. Microsoft will soon have this fixed via a Windows upda... oh, wait.
== Jez ==
Do you miss Firefox? Try Pale Moon.
You are of course aware that step-by-step instructions for opening the correct config file, scrolling to the right section, etc, would be about as long? (The note about comments inline in files is certainly valid, but support instructions on the web are for those who have already looked at any such info without luck...)
FWIW I implemented a dhcp server because in my opinion ISC dhcpd sucked, lacked features we needed, and was/is likely to have more security bugs ;).
My reading of RFC2131 (4.1) was if a client sets the broadcast flag, that means it wants the server to broadcast the reply back (if replying directly[1]). Seemed fairly clear AND reasonable to me. [1] The server should not broadcast if the reply is to a dhcp relay (giaddr is set) and thus not direct.
So if Vista is using the broadcast flag that way, then it's compliant, and a DHCP server that doesn't like that is one that's buggy and noncompliant (and same goes for the ISP).
In my experience the really non-RFC2131 compliant stuff appear to be some PDAs (some versions of windows mobile+PDA being offenders), I had to make my dhcpd less strict and even then some PDAs still wouldn't work (can't remember but those were probably doing stupid stuff in the vein of sending a dhcprequest for 0.0.0.0).
I also suspect some Macs running Parallels were doing some weird DHCP stuff. Can't remember exactly either but I think it was something like doing dhcp stuff using one hardware address then switching to using another every now and then (but requesting the same IP that was offered to the other hardware address).
Then some flash/shockwave stuff make dhcp requests too with hardware type 0 (not ethernet)- just to look for some flash server thingy or who knows what. Don't know why they have to use(abuse?) dhcp instead of using some other protocol.
It's amazing the stuff that's out there given DHCP is a fairly simple protocol.
Imagine the weirdness and bugs with things more complicated like DNS and IPSEC...
Uhh what the fuck? Windows XP worked perfectly fine with the ISP. Windows Vista doesn't. That's **Microsoft's** regression. Linux users were never mentioned, as it assumes they have followed the standard all along and don't release new distributions who break standards for no purpose other than to break them. This should give **Microsoft** a bad name for not being able to release a new version of their own OS without breaking virtually every component in the process.
"Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
And there appear to be only two reasonable explanations for why VIsta behaves this way:
1. Incompetence.
2. To intentionally fail on non-MS DHCP servers.
With anybody else I'd apply the "Never attribute to malice that which can be adequately explained by incompetence" axiom, but with Microsoft I'm not so sure.
No - the implemented features now suddenly, after almost 20 years of neglecting them, for no obvious reason used by Microsoft. These changes are NOT communicated with the rest of the world, and it is suggested that the fault lays with non-Microsoft products if it goes wrong. Only after people are scratching their heads and wonder what the hell is going on, they admit the changes they made. And THATS the problem!
... is this behavior of a company that want to make their software look better than anyone else by sneaky changing some parameters and hide that for a period of time?
They changed something. They did not tel it to anybody, and suggested the fault lies with anything but Microsoft products. Only after asking what is wrong, they admit it is a setting that nobody ever used before (still suggesting it is NOT the fault of Microsoft, but a fault of everybody else). Now tell me - is this behavior of a company that want to play nice with other OS-es? A company that said it wants interoperability with other OS-es? Or
The answer is obvious...
That's funny - Our Cisco DHCPd worked fine with Vista, but DHCP3 is choking.
I believe the spirit of RFCs is something like "Be liberal with what you can receive, and conservative with what you send out.", or something to that effect. Vista is adhering to the RFC. The worst you can throw at it is not adhering to the spirit of RFCs, but the problem only exists because the server software, if adhering to the RFC at all, doesn't adhere to the spirit of it, either.
Personally, I'd say that when standards compliant, the burden is on servers to accomodate client implementatins, not clients to accomodate server implementations.
What is this, fucking Digg? That is the worst summary I have read in my life....
http://watching-eyes.blogspot.com/
The DHCP broadcast flag was to be used for old servers. This was 15 years ago and it was being discouraged to use back then. There may be only 2 or 3 servers in the U.S. or even the world that activly require the DHCP broadcast flag now. All but the 2 or 3 servers worldwide don't support or break with the DHCP broadcast flag on. Why should any ISP try to support something that is outdated and was discouraged 15 years ago?
Obvious, yes, but not the one you'd arrive at.
You're making this out to be a Microsoft against the world situation. Our Cisco DHCP server worked fine with Vista, no problems at all. Only when we switched to an ISC DHCP server did we start having problems. In fact, the only DHCP servers I've seen so far having trouble playing nice with Vista have been ISC DHCP servers.
This is a failure on the part of ISC to make their DHCP server standards compliant, regardless of how esoteric you find the Vista client implementation.
As for Microsoft announcing the problems to be with non-Microsoft products - Wow. A corporation denying responsibility for issues it isn't responsible for, and using the failures of competitors to promote their own software. What's next?
That is what the Internet is for. You're projecting Windows' problems onto real computers. There is no reason why a router or hardware firewall should be necessary to add security -- they're both computers with instructions and flaws. Increasing the number of hardware pieces increases the number of failure points at the cost of also increasing latency and reducing actual bandwidth.
There are only three reason why a computer needs to be isolated from the Internet:
Help stamp out iliturcy.
From my understanding of the problem, Microsoft is no longer supporting the unicast response for DHCP like it did previously, even though that is the recommended way to do this. While Microsoft's implementation is valid (though not recommended), I can see why the ISP doesn't want to honor it. If a lot of Vista subscribers start doing this, there could potentially be a lot of broadcast packets. I.e. each time a Vista user connects, the DHCP server would send a broadcast response to everyone on the local subnet (which can be quite huge).
I remember scanning the broadcast network traffic years ago on my cable modem and it was tens to hundreds of DHCP requests packets per second. If most users start running Vista then this would double the broadcast traffic.
Broadcast should be avoided unless absolutely required.
This post is encrypted twice with ROT-13. Documenting or attempting to crack this encryption is illegal.
> The problem as reported is that the Vista DHCP client fails to obtain an address from Linux servers running (I'd presume) ISC dhcpd.
/. theorizing.
Look folks, this is pretty simple. If this was a 'linux' problem, even one of not accepting slightly malformed packets I'd say it was our problem to fix. But it doesn't have a damned thing to do with 'linux'. The problem appears to be one of Vista failing to interoperate with the ISC DHCP server and sorry, if that damned thing isn't the canonical reference implentation that EVERYBODY should be testing against somebody suggest a more widely deployed implementation? Not making sure that a new product works with both it and the tiny implemenation of DHCP in Linksys routers (the first runner up for most widely deployed implememtation) makes Microsoft at fault.
Of course it ISN'T quite so clear cut because I am not having problems with the ISC DHCP servers under my control successfully handing out addresses to Vista clients.
I really would like to read a clear explanation of the problem from a source in a position to know, not this endless
Democrat delenda est
Vista = Fistya
google "32 trillion offshore needs IRS attention"
Comment removed based on user account deletion
Don't use a flag that has no benefit other than being incompatible?
Come on - this flag is only set on VERY old equipment, and has absolutely NO need at all in Vista!! Now - why the hell this flag is set? To gain interoperability? Really?
A command-line registry editor? You don't mean using vim to write a C# program that edits the registry, then invoking csc on that file, then running the output? Where is it? And how does it handle the evil practice of using GUIDs as registry directories?
If, for instance, the registry were an actual directory, containing subdirectories for each nonterminal registry directory and files for each terminal registry directory, a text editor and tab completion in a sane shell could get you 95% of the way there. And pretty much nothing will get you the rest of the way, since registry keys have a default value, so you don't need to have the key present in order for the program to work.
These are the two ways in which GNOME's gconf is better than the Windows registry. The typical UNIX example of rc files only solves one of the issues -- there are unspoken defaults -- but comments can help with that. And of course the gconf xml files don't have comments, since they aren't intended to be edited manually.
Because isn't that what Hollywood asked Microsoft to do?
If you've seen one Broadcast Flag, you've seem 'em all.
"My country, right or wrong; if right, to be kept right; and if wrong, to be set right." --Senator Carl Schurz (1872)
Every one of the DHCP servers in the world, on every OS whether embedded or multi-purpose should be audited and downgraded (yes, this is a downgrade to a deprecated method) or replaced with obsolete equipment.
This should be done because Microsoft's Vista network programming team could not be troubled to code in something like "If DHCP request using deprecated method times out, retry with the standard method."
And no copying my idea. That's valuable Intellectual Property there.
Help stamp out iliturcy.
hmmm, I wonder how Microsoft could have put so little effort into the DHCP networking system of Windows Vista with all the time they spent developing it? Whatya think, they don't put much value in networking? They know it breaks connecting to open source DHCP servers and did it on purpose? Or are they such hacks at software development that over five years, billions of dollars, and 10's of thousands of developers to upgrade Windows XP to Windows Vista resulted in a flaw in their software to communicate with their number one threat in servers and growing threat in desktops?
IMO, make the ignorant bastards who purchased a computer with Windows Vista suffer the consequences of their action. They don't deserve to be connected to a network and should be told to call the company who sold them the computer for support. ie, make sure the money trail includes those who brought this on in the first place.
LoB
"Anyone who stands out in the middle of a road looks like roadkill to me." --Linus
Except that they would have to snail-mail the instructions to the affected customers since they may have no access to the Internet until this is fixed.
Help stamp out iliturcy.
Good point. It'd be soooo much easier for Aunt Millie to follow these instructions:
m eters\Interfaces\{GUID}
1. Find your DHCP config file and open it in your text editor
(Sudo if you have to)
2. Locate a line that looks like:
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Para
In this part of the file there may be multiple {GUID}s listed. Find the one that corresponds to the network adapter that is connected to the network.
3. Create a new line below the correct{GUID} and type DhcpConnDisableBcastFlagToggle = 1
4. Save your changes and close the text editor.
Technically the process is less steps, but is it *really* any easier for a non-technical person to accomplish? Besides, if you wanted you could make the same change in Windows using a text-editor.
It would be easier to say:
/etc/dhcp.conf
/etc/init.d/dhcp restart
sudo gedit
change DhcpConnDisableBcastFlagToggle 0
to DhcpConnDisableBcastFlagToggle 1
Save modified file and quit gedit
sudo
And you got all this unsubstantiated speculation from where, exactly?
Oh, that's right; you pulled it from out of your ass.
Just read your journal BTW, you seriously think Microsoft is organising Slashdot posters against you? Honestly, if Linus Torvalds, RMS, Bruce Perens, Theo de Raadt and other people with far more, and far more deserved significance in the F/OSS movement can go without all their postings on LKML/Slashdot/whatever being the victims of "M$ astroturfing", what the hell would Microsoft be doing going after some random nobody with nothing to his name except making everyone else who believes in a cause he supports look bad?
By summer it was all gone...now shesmovedon. --
Yes, it is that easy. And what would you change tomorrow should M$ decide that with service pack 1 for Vista the network order is little-endian?
Would anyone seriously suggest it was the ISP's fault if I ignored RFC-2131 when writing the dhcp client for a build of SkryneOS.
Twitter is quite right in saying that where Microsoft products depend on competitors upstream services there is no obligation on these competitors to deviate from published standards to facilitate Microsoft's product.
I think this is another instance of using the moderation drop-down because the reply button requires too much work.
"Linux is for noobs"-The new MS fud strategy
Having to use a certified appliance between an MS-OS and the real world would solve the luser problem at a stroke ;)
"Linux is for noobs"-The new MS fud strategy
It's a good technical decision.
Vista is not ready for prime time. Microsoft will surely fix this at some point, probably before Vista itself is ready to be connected nakedly to the net. Until then, it's a very good idea to discourage users from putting up a known unreliable system that'll just get rooted.
That, and the site's been down for years.
I just read Slashdot for the articles.
Thousands upon thousands of special cases dating back to dos and 16 bit processors. It's a large part of why, for the most part, windows just doesn't work.
The ISP is doing the right thing.
Before Vista, with XP, for reasons I don't understand I NEVER saw more than about 300kB/s or 2400kbps download speed on my high speed cable internet ... despite a speed measured at 1187kB/s or 9500kbps at speakeasy.net
I upgraded to Vista, same hardware, same connection, same speakeasy.net speed, only the OS changed. Now I usually get download speeds >1000kB/s. Perhaps there was some setting in XP that was wrong, who knows, but the speed improvement I've seen with Vista has been wonderful.
So I think the useless ISP in the story is intentionally blocking Vista users because they are scared they won't be able to provide the bandwidth that their customers will expect to have with Vista.
George Bush + Linux = "I will not let information get in the way of the fight against Windows"
Do you really want Swedes on the Internet? I mean, aren't they like the Canada of Europe?
Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly and applying the wrong fix.
The ISP has implemented the DHCP standard correctly. Microsoft has not.
The ISP is not "screwing" Microsoft, it's holding it to the standard as well they should.
Of course, more or less. But that's only relevant the first time you have to config something. The second time, you know how to open a file, search for a keyword, and edit the line. You don't have to relearn these things a zillion times, once for each special-purpose config thingy.
;-)
Actually, the simple-minded sort of time-and-motion observations (that I've been doing mostly out of boredom) say there's often still an advantage to the plain-text, command-line approach. This is in great part because of something that can get really annoying with most current GUI tools: Every time you open a new window, it's back at your home directory, and you have to repeatedly wend your way through the stack of directories to find the file. There's nothing like the "cd" command to remember a point in the directory tree, or a "set" command to give a short name to the path. And the GUI tools have this annoying way of suddenly closing a window when they think a task is done, losing the directory information and forcing you to "drill down" to it again as you make repeated tries to get thing configged right. The GUI folks might be able to invent something like the "cd" concept, but they don't seem to have done it. Even the vaunted Mac interface is bad here, forcing you to waste time repeatedly getting yet another tool to the same directory.
(Not that the CLI folks have solved all the world's interface problems, either. But they've been more honest at facing up to the problems, and kludging up partial solutions to a lot of them.
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
Yea - no Microsquash no workie workie - duh, what is new with that.
Good for the ISP not lining up to kiss some Microsquash ass and working a fix around their problem child (imo).
But more important - - - I live in America and surf for Swedish porn. When in Sweden whose porn do they seek ?
Once upon a time, a soon to be mommy and daddy loved each other very much (the lust was strong as well as the drinks)
3.) Security
4.) You are an idiot
thanks
Read the RFC genius: If 'giaddr' is 0x0 in the DHCPREQUEST message, the client is on the same subnet as the server. The server MUST broadcast the DHCPNAK message to the 0xffffffff broadcast address because the client may not have a correct network address or subnet mask, and the client may not be answering ARP requests. If 'giaddr' is set in the DHCPREQUEST message, the client is on a different subnet. The server MUST set the broadcast bit in the DHCPNAK, so that the relay agent will broadcast the DHCPNAK to the client, because the client may not have a correct network address or subnet mask, and the client may not be answering ARP requests.
Lundis Energi should have been testing Vista back in its early alpha release stages to ensure compatibility with their Linux based server system.
Yeah, you fucking genius. Instead of Microsoft testing their shitty future product with one of the most popular DHCP servers in the world, some random IT guys in Sweden should do that for them. Yeah.
I mean, after all - it's our responsibility to test if Microsoft product works, not Microsoft's.
Genius.
Lundis Energi should have been testing Vista back in its early alpha release stages to ensure compatibility with their Linux based server system.
Huh? With as much time and energy invested in Vista, wouldn't it be Microsoft's job to make sure Vista worked with existing servers? That is, if they really had any interest in working with existing servers. I have had similar problems trying to get Vista to work with a wirelss access point that has been running for 3 years with Win98, XP, and several Linux laptops.
Security reasons, my ass! This is just another example of Microsoft abusing their monopoly market position.
Vista - just say NO!
I find it hard to laugh really hard until Microsoft is gone. Mr. Pope, all the modern evils are in Microsoft, _not Google_ btw. Ironic that I happen to have a guest staying from that very town. Nobody wants Vista here (in Europe) and therefore a small chuckle. Clearly if they play, well, Microsoft... and every compliant OS works fine, tsk, tsk, I side with the people of Lund.
BillSF
MS is properly implementing a known, albeit outdated, DHCP option. The DHCP server is failing to respond rather than either implement/ignore the valid option. Incidentally, this issue is fixed in ISC http://www.isc.org/index.pl?/sw/dhcp/dhcp-v2.php, so it's also possible that ISP is ignoring an update that will fix the problem. Also, MS has stated which registry key to change to disable the offending broadcast flag option http://support.microsoft.com/kb/928233, so it's not correct to say that MS has not responded. Given that they are RFC compliant, they didn't need to do anything.
meh, -1 !!?? Please, mod parent up!
Please don't mod this funny, I am 100% serious.
Please don't mod this funny, as it doesn't count towards karma. There, fixed that for you.
Get your own free personal location tracker
Twitter is starting at -1 because moderators are finally catching on to his irrational behavior. His pathological hatred of Microsoft makes any attempt at rebuttal an exercise in futility. He's immune to the Reply button.
If you think developers are lazy, the average non-user is pathetic. I guess I'd be quite screwed-up if I used such an opiate for over ten years. I prefer to serve and I was shown NT (the __only__ Windows when we began the experiment) was unsuitable right in front of my face.
The answer: People are lazy! (not just developers) What I cannot tolerate is all the needless suffering this laziness has caused. I'm pissed. Most people who are shown Linux (and more likely *BSD from the younger ones) take to Unix real fast. If "Generation X" had full use of their computers, they'd be dangerous!
Sorry this seems so burnt-out, but two days of dealing with Windows "non-users" is beyond my limits. ANYTHING UNIX(tm) or "Unix-like" simply attracts smarter people. That's not simply because they used it in university, Unix users are simply smarter. (Duh)
That, and the site's been down for years.
As of January 14, 2004, the domain goatse.cx was taken offline, but many mirrors of the site are still available,
gleaned from http://en.wikipedia.org/wiki/Goatse.cx
and the trolls still try to direct you to a mirror.
The truth shall set you free!
It'd be soooo much easier for Aunt Millie to follow these instructions: ...
m eters\Interfaces\{GUID} ...
Heh. I've watched a fair number of Aunt Millie analogs trying to use both
Windows and Mac GUIs to do such things. I've watched as they get more and
more frustrated, and tell the world what they think of the idiots who imposed
such a godawful way of doing things on people like them. So no, I don't think
that the existing GUI config stuff is any better than the CLI stuff. It just
takes more keystrokes (most of the time), and changes more with every release.
2. Locate a line that looks like:
HKLM\SYSTEM\CurrentControlSet\Services\Tcpip\Para
3. Create a new line below the correct{GUID} and type DhcpConnDisableBcastFlagToggle = 1
One of the things I've seen with plain-text stuff is that the programmers know
that such garbage names will be seen by users, and those users will mock the
idiot programmers who imposed such names on the users. Especially things like
setting a "Disable" flag to 1 to turn something off; even the dumbest user
knows how perverse this is. Not that you don't see such things in plain-text
stuff; you certainly do. But there is subtle pressure (based on the desire
to not look too hostile to your users) to call it "BcastFlag" with 0=off
and 1=on.
Another observation is that, while geeks like us certainly can handle such
instructions, we often need several tries to get it right. I notice that
there's a spurious blank in the "Param eters" field, and I'd expect that it
will make the operation fail (or be ignored) if you type it that way. This
is one more thing that helps encourage programmers to be less user-hostile
and use a simpler heirarchy for such things. And not include blanks in names,
as is done so often in GUI-based packages.
Not that things are really all that much better in the CLI world. We still
have a long way to go before we can honestly claim that we're user friendly.
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
I believe twitter calls that "evangelization".
Web2.0: I love when people Flickr my cuil and digg my boingboing until my google is reddit and I start to yahoo
You're not passing the DHCP requests from the LAN. CentOS is your WAN side DHCP *client* and likely is providing your LAN side DHCP service (what the Vista boxes are talking to). Totally different scenario.
Quack, quack.
People who need hand holding to use email aren't going to be able to edit their registry.
Microsoft is relying on an optional feature whose use is discouraged. Servers are not required to implement it. Clients are required to be able to operate without it.
"M$" has not broken any standard. They are simply using the optional broadcast flag. It's antiquated, sure, and not *required* by the standard, but it IS PART OF THE STANDARD.
I certainly think MS should change their implementation so that it automatically omits the broadcast flag if the server doesn't respond, but your post is nonsense. MS isn't changing/breaking the standard by any stretch of the imagination.
SHOULD simply means that you MUST NOT require that particular feature to operate. Which is exactly what Vista is doing.
The headline "Vista Bug Costs Users In Swedish Town Their Internet" is clearly misleading; nobody here seems to disagree that this only affects Vista users.
And if this was important for the people affected (Lund is the largest university city in Sweden), this would be all over the blog-world in Sweden. But I have been unable to find any swedish bloggers writing about this. In fact all the blogs I have seen seem to end up at the same Inquirer article that we see in TFA, and the Inquirer goes into little detail and has no references. And if you understand swedish and look at the web pages of the city net of Lund, there is no mentioning of such problems, not even on their support pages.
Sweden is quite different from the US when it comes to internet access. In Sweden there is real competition, and if somebody do not like what the city net of Lund is offering, there is at least two (probably 3-5, but I only know two because I do not live there) other ISPs that people can choose from in Lund.
And the funny thing is that the city net of Lund isn't actually an ISP. They supply most of (but not all) the lines in the city for people who want 100mbps internet access, and they have a number of ISPs that supply internet access through these connections. Once you have your connection, you are directed to an internal web page where you are asked to select the ISP you want to use.
Most likely the problem with Vista is only for getting an initial IP address so you can go to this web page to select the ISP you want to use. And this is why nobody in Sweden really cares about this.
Of course, Vista is non-standard software; however, what is the *point* of this? Manufacturers still telling their customers it'd be all plug and play? It was always a good idea to run a router/firewall locally that allows for static IPs in order to avoid the DHCP mess in mixed / Windows environments. There is always some printer, or Windows machine, that can not deal with regular DHCP. There is no plug and play, but beyond that unfounded assumption, there is a lot left to explore.
n/t
SB
It's old. The more humans I meet, the more I like my cats. At least they are honest.
Only if it's connecting over infiniband or firewire. Over ethernet it SHOULD set that bit to zero.
Ah, the dangers in summarizing RFCs.
The Internet is the communications network that most of us on Slashdot use on a daily (or near daily) basis. An internet is a generic inter-network.
-Mr. Pedantic
Setting the BROADCAST bit is not following the recommended approach, but it's still standards compliant.
The Pirate Bay
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
Service: Provide internet
Process: Any means necessary.
Using *your* clients as hostages to save you some work is about as poor form as it gets. I can get onto my internet just fine with Vista through a linux server so I'm assuming it's possible.
Somebody over in sweden has a god complex and I hope it bankrupts them for their childish antics.
From http://www.metro.se/se/article/2007/08/28/14/2423- 48/index.xml
Lund’s Citynet Shuts Out Surfers With Vista
BROADBAND. In Lund there will be disappointment for those who bought a computer with Microsoft’s new operating system, Vista. More precisely, there they will not be able to connect to the Internet.
Lunds open “citynet” has a Linux server that is not compatible with Microsoft’s operating system. According to Lunds Energi, which runs citynet, the fault lies with a bug in Windows Vista, and there are no plans to replace the server.
“A swap could lead to problems of a different sort. Our engineers are looking at what we can do. But it is really up to Microsoft to take action on the matter,” says Åsa Holmander, product director at Lunds Energi.
At Microsoft, the whole thing has come as a surprise.
“I haven’t heard anything about this. Nobody has contacted us as far as I know. But if Lunds Energi gets in touch with us, then we can surely arrive at a solution,” says Michael Bohlin who is product marketing chief.
Lunds Energi refers customers to a solution at Microsoft’s support pages. But for Annie Johansson who will begin studying in Lund this autumn, these recommendations weren’t helpful.
“It made no difference and still the Internet worked fine at home in Ljungby. Now I get to borrow my little brother’s computer instead.”
ya i can see it now ... call in to tech support "ya go into reg edit and modify this key here... ok your good to go" 2 hours later "what did you do? you deleted a bunch of things from regedit and your wondering why the internet will not work and your computer wont load anything... im sorry call MS"
for a technician taking a user into the registry is like giving a terrorist a tour threw a reactor... you just dont want to do it if you want to have problems
(yes i know i suck at spelling fell free to correct my grammar and/or spellin i dont care, im still not going to change
You need the GUID of the NIC that it applies to. So you need a little more logic than a .reg file can offer. But nothing you couldn't handle in VC++ quite easily.
Newly written clients aren't supposed to use it.
Sufficiently advanced malice is indistinguishable from incompetence.
-- The act of censorship is always worse than whatever is being censored. Always.
While I'm not necessarily disagreeing with you, what would #4 have to do with the way Vista behaves? I didn't write it.
Not snarking, clarifying. It could well be that you're right, but you'd want to be bloody sure before arguing it too hard...
FWIW, I could describe a dozen cases of broken dhcp clients that work with the ISC dhcpd but fail with other implementations, but I won't. Instead, just google +"Xbox" +"dhcp", +"printer" +"dhcp", +"nslu2" +"dhcp", +"playstation" +"dhcp",
In fact, I'd be a little surprised if it does turn out to be a problem with the ISC dhcpd - it's streets ahead of every single other dhcpd available on Linux or the *BSDs.
What part of "a well regulated militia" do you not understand?
Warning Serious problems might occur if you modify the registry incorrectly by using Registry Editor or by using another method. These problems might require that you reinstall your operating system. Microsoft cannot guarantee that these problems can be solved. Modify the registry at your own risk.
But, the instructions are only 7 steps long. The disclaimer is there for idiots who can't follow a simple 7-bullet list of steps.
And man, it's not like reinstalling your operating system is some sort of incredibly difficult chore. I've had to do it numerous times. It's routine computer maintenance.
Microsoft's laziness aside, if "businesses, schools, and grandmothers" are using computers in their day-to-day lives, they ought to know how to operate and maintain the things ... kind of like how you ought to know how to check your car's fluids, change a tire, etc.
Of course, there are two entire industries built on people's ignorance of how to maintain both computers and cars, so, I know that I'm dreaming here. People love convenience too much.
But more to the point, the how-to isn't hard, and people ought to be more comfortable learning how to maintain their machines (whatever they may be).
Then again, Vista does not agree with the "conservative with what you send" principle, considering they are using a flag they do not need to use to operate correctly and in fact is deprecated and should only be used in legacy clients.
Both vista and dhcpd would adhere to the standard if they can provide an explanation of their respective behaviours. However there appears to be no possible justification for MS' actions, while the dhcpd behaviour could be attributed to dropped support for now obsolete legacy clients (which must have been almost unneeded until vista appeared). Instead of bashing on dhcpd why don't you provide the necessary explanation that would make vista compliant?
Also, the standard declares that the correct solution is to modify clients in order to follow what is considered the correct behaviour (not to require broadcast answers). So, in this particular case, the burden's on the client, not the server.
See the DISCUSSION subitem in the quote I sent in my earlier post to see why dhcpd should not accept vista's requests.
Basically the whole thing is a small fuckup by dhcpd and a fucking huge one by vista (or plain malice, as I suspect).
GPG 0x1B479C78
For command line registry access, use reg.exe (type "reg
The command line tool is pretty well written, real easy to put in batch files, and can even handle changing individual values (something that you can't do with a
The registry is evil because it is a binary blob, not because of how difficult it is to work with.
But I seriously think the network admins should get off their Linux high horse and work for the greater good of the community to solve this problem... I would say they, more than Microsoft is to blame here. They found one specific case where Microsoft software failed, and are, based on what I'm reading, failing to help anyone fix the problem (including microsoft). Microsoft on the other hand has to try and to make sure Vista works properly on networks with every other platform that is still relivent, all while trying to implement some new networking protocols which may help end user networking experience one day. A much bigger job then emailing up microsoft with a note saying "our servers are configured xyz and it seems Vista doesn't work, here are some relivent details...", then perhaps, security etc allowing making some adjustments to let Vista work in the here and now so your users can browse the net... To me this more sounds like some jerk admin using this as an oppertunity to try and make people hate microsoft and switch to the software they feel everyone should be using regardless if that is actually doing them any good service or not, more sort of a religous crusade that shouldn't be a part of the world of techology.
Perhaps they aren't conservative in what they send, but if the ISC DHCPd was any better at being liberal in what they accept, this problem wouldn't exist, so they're both to blame in that respect.
Microsoft don't need to justify their decision to set the bit to what it's set to. One bit that is within standards is certainly not a "huge" fuckup, and since it appears to be supported fine by other DHCP servers, one can only assume that it's a lacking implementation on the part of ISC.
Our other DHCP servers work fine with Vista. Only the ISC DHCPd doesn't play well with it. How would the other servers know how to support this bit, if not from the published standard?
exactly the point.
It is very interesting how the ISP is blamed here. He could, he should, yada yada. M$ delivered a broken configurations and the ISP has to fix this. Generally not wrong. However, as Linux user I wonder a bit, why he should do this? I have enough examples where ISP refuse to support Linux. If it works, fine. If not, problem of the user. I never heard similar ISP blaming comments about this in any forum. More likely: "Oh yeah, Linux is the outsider, only 0.00001% of all user use it, heh heh heh."
And how often do I hear about the superiority of Windoze. When some WiFi card does not work under Linux: Linux is not fit for the general desktop. If some WiFi card does not work under Vista: The stupid manufacturer was not able to deliver proper drivers on time.
I begin to think the only reason that Windoze works at all is because everybody bends over for M$ and paves their path.
Sorry, but even if those voices, which say the ISP could have acted on behalf of their customers, are right, and they are, I still deem them hypocrites.
Mod parent down for being wrong in letter and wrong in spirit.
The purpose of the RFC broadcast bit, set in Vista and part of the DHCP RFC, is to request a broadcast response. One reason for this bit, which it is the responsibility of a DHCP server writer to support, is (as mentioned in previous posts) to support TCP/IP implementations which simply can't cope with a non-broadcast packet before obtaining an IP address.
Another (which I haven't seen discussed yet in the Comments) is for network management: one machine can listen in for all these requests and log/detect conflicts/unexpected packets, where such machine need not be the same physical box as the DHCP server. By configuring a Vista box to only accept broadcast responses, you make it a bit harder for some ne'er-do-well to unicast a false DHCP offer that isn't seen by the logging box.
Put simply, Microsoft are attempting to improve security, and they are doing it while adhering to standards. It's intellectually dishonest and thoroughly harmful to genuine Linux advocacy to label this as Microsoft's failing; if this article was written to parody the irrational zealot who leaps before he looks, it is first rate.
In other words, Microsoft has found some security bug in MSWVista^H^H^H^H^HLook-Mah-No-Hands when running DHCP the corret way.
Maybe they think they are trying to wrap a man-in-the-middle vulnerability in some sort of protective covering?
The standard says that servers SHOULD implement the BROADCAST flag, just like it says that OSes that don't need it (like Vista) SHOULD leave it unset. "SHOULD" means that, unless you have a good reason, you should do it that way.
In fact, since the BROADCAST flag is a workaround for implementations that were considered old in 1993 and use of it is discouraged (see RFC1542 section 3.1.1), DHCP servers that don't implement it have good reason and are more standards-compliant than Vista. (I would be entirely unsurprised if there network setups out there where the BROADCAST flag cannot be implemented due to broadcast packets from the DHCP server to the client being dropped.)
I'd call that real competition.
The thing is, from doing some quick Googling and looking at the code, it would appear that ISC dhcpd does in fact support the BROADCAST flag, at least for some packets (presumably the ones that need it). It's odd that you're having trouble with it.
Comment removed based on user account deletion
Comment removed based on user account deletion
doesn't vista require online authentication? it should be up to microsoft to then take care of this since their product just doesn't work in the town.
Setting the broadcast bit should not break your DHCP server.
That's also true. I'm not defending the DHCP servers in question, though I would like a bit more details about what the actual bug there involves, and why they didn't respond. Alas, googling brings up more smoke than light.
My point is that Microsoft is not blameless here. If you know anything more about their decision I'd really like to know what led them to start setting this flag now, when it's been discouraged since 1993* in booth DHCP and BOOTP, and since their TCP stack doesn't actually seem to need it. This is unfortunately symptomatic of Microsoft's general approach to standards.
* And its interesting to note that it's only required for Infiniband because of the size of Infiniband addresses... NOT because of the reason the flag was originally created.
I beg to differ.
Start>run>mmc
Import your scripts, and have whatever you want at your fingertips.
Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
Take a look at all the spyware/malware that is all-to-easy to install, the IE bugs, etc, and a wise user will have long ago concluded that Windows probably shouldn't be used on networks anyway. Windows is for legacy apps and games. Just run 'em in a VM, on top of whatever modern OS system you use for really getting things done.
Wow... imagine if the situation were reversed, their networking software was Windows based, and it was denying access to people who use teh Lunix.
You just know all the FOSSies would be tearing their hair out and gnashing their teeth and screaming for blood... but when this town starts denying access to people using Vista... suddenly all the FOSSies are fine with that.
Remember, teh Lunix is "all about choice"... so long as you don't dare to choose Microsoft.
Well, whatever.
If you think MS does not have to justify their use of a deprecated behaviour, fine. I guess this loss of connectivity is correct result then. They're both standard compliant, isn't that great? Good luck embracing and extending, but remember that vista isn't that popular yet.
I'm tired of feeding a troll.
PS: when dealing with dhcp, one does not assume anything, one reads rfc-951, rfc-2131 and its revisions and updates.
GPG 0x1B479C78
Yeah, you're right. I know what it does, I just should have written to 10% instead of by 10%...
Don't thank God, thank a doctor!
Well, since you asked, Microsoft's ME II, better known as Vista, is causing unhappy faces everywhere I go. It isn't just that people don't want to use it, or that it's insecure and buggy or that the very word vista has "failure" attached to it. It isn't that Vista isn't even compatible with Microsoft's own SQL Server.
Most of the people that I know only care that it's not possible to deploy Vista with industry standard tools. A rollback is likely, and there are substantial unresolved issues preventing deployment.
Although I'm aware you don't appreciate twitter's attention to these matters, I do. I do appreciate twitter's attention to these things quite a lot.
Thanks, twitter.
Help stamp out iliturcy.
Having to edit the registry like a complete nerd is why Windows will never be ready for the desktop.
The decision of the Internet authorities of Lund must be praised because they have now made the commitment that all users of Windows (any flavor should do) : The big must be fixed by the provider of the Operating System, nobody else. I am 65 years old and I still remember when we wree hearing im Werners Lee talking about the CERN and HTML to provide a free platform trough the Internet, free for all, and Bill Gates had just sent his 8 satellites (by the way they must be still circling the earth, I assume) and Microsoft postulated that these kind of "private" network was the ONLY alternative to the emerging Internet. He wanted toi own what we now know as Internet, but he has never committed his own company to respect and fully support the rules of te Internet Engineering Task Force has issued. Microsoft never admits openly a bug, well knowing of the weakness of their own OS coding. Microsoft su`lies just the "temporary" bug that hopefuully will preven a certain kind of mal-ware to disrupt your computer OS, something you paisd for not only by the license, but now buying completely new hardware to run this VIsta-monster! Many people today, mainly young people can really beleive that Bill Gates invented the Internet. Well it did no happened that way! Why should Bill Gates ensure the users (paid) pf his weak and ressource-swallowing OS Vista that will ever comply to 100% all the Internet agreements all around the world rthat have made possible that everybody (except Internet Explorer Users) see the same pàges we design fro all the other Operating Systems? What benefit is there for Bill Gates to ensure that if you are using Outlook instead of Thunderbird or Forefox to read your e-mail, that your computer donot acquire instantly hundreds of spywares and every conceivable kind of bugs? Only if you use Linux (Ubuntu, for example) or Mac OS X you can feel safe! I really as a Dane and Chilean I do applaud the brave and correct decission of the city of Lund (By the way their inhabitants do own a sizable amount of Macintoshes). Windows is a bother. I do use it just to verify that my PHP-MySQL-CS2 code is working or not in the latest flavor of incorrectness : Internet Explorer Hellmuth Stuven Lira