James_A

Thanks for this list. I did a similar list some time ago when we actually needed some of these Hotfixes. However, at that time, I missed 843503 (which seems to completely replace the earlier 829759). What do you mean by "These files aren't easily available at Microsoft"? Once the numbers are known, all of them are easily available on request by email with just a few mouse-clicks. You've already done the hard part, which is finding the KB numbers. .

KB943729 file names are a little non-standard at the best of times. For Português (Brasil) it's actually Windows-pt-BR-KB943729.exe, found here: http://www.microsoft.com/downloads/details...61-247ce3f6c4f8 .

According to Asus, this board does support Phenom processors, but not the 9950. Spec here: http://www.asus.com/products.aspx?modelmen...p;l3=0&l4=0 CPU support list here: http://support.asus.com/cpusupport/cpusupp...uct=1&os=17 .

Actually, cluberti, you've completely missed the point of my question. Perhaps I was a little too brief. So, let me explain. In Yzöwl's post of screenshots (see post #8) the behaviour of "my usual Gecko based browser" and IE7 are shown as somewhat different. Based upon that, this post then took a somewhat OT turn into an argument over the different rendering of different browsers. Now I have come along, with the implied asseveration that most of this discussion is actually based upon a false premise. Because MY "usual Gecko based browser", namely Firefox, renders the CODEBOX essentially the same as Yzöwl's IE7 (give or take a few pixels) and not like the first screenshot at all. We thus revert to the original question by Yzöwl. We know that the CODEBOX rendering can be different on different versions of IP.Board and also that it will be radically different in the upcoming version 3 (as confirmed by a discussion between myself and Tarun on another forum). So, has anything changed on this board, or not? That is, after all, the original question and it has not yet been answered. .

The SANS Internet Storm Center (SANS ISC) have just posted some interesting new information about Conficker/Downadup, in the handler's diary: Some tricks from Conficker's bag. First, there is some additional technical detail of an already-known behaviour of Conficker, namely the ability to patch the MS08-067 vulnerability in memory. Unless you have applied the REAL patch, this brings with it the risk of re-infection when disinfecting. Secondly, Conficker uses an undocumented function to deletea all System Restore points. That is why you cannot remove Conficker by returning to an earlier Restore Point. On the plus side, this does save you from searching and deleting Restore Points when disinfecting!!! There are some 24 undocumented functions of srclient.dll, including the ResetSR() function used by Conficker, in addition to the documented SRSetRestorePointA() / SRSetRestorePointW() and SRRemoveRestorePoint(). For the curious, a full list can be found on the Sysinternals forum: http://forum.sysinternals.com/forum_posts.asp?TID=15352 in the second post. .

So why does my copy of Firefox draw the Codeboxes the same width as Yzöwl's copy of IE7? (Give or take 20 or so pixels) BTW Spam's browsershots disappeared overnight -- yesterday's interesting page is today's dead link. .

The trouble with looking for general trends in the HDD -v- SSD debate is that there is a very wide range of performance available. In the UK tests I referred to earlier, on 15 different SSDs there was a 3:1 performance ratio between the fastest and the slowest SSD when reading. There was a 7:1 performance ratio between the fastest and the slowest SSD when writing. A plain (cheap) Toshiba 2.5" laptop HDD was quicker at writing than over one-quarter of the SSDs. SSD parts that we don't even know how to measure properly (Memory controllers, Wear-levelling, Erasing NAND in blocks instead of pages, before writing) all made a difference and introduced unexpected latencies. If you want to throw (or, more provocatively, waste) large amounts of money on your storage devices then a pair of Seagate Cheetah 15K.6 HDDs run as a RAID 0 array driven by an Adaptec 5445 SAS RAID controller blew everything out of the water, even the Intel X-25 Extreme SSD. Almost 300MB/sec. That's three times faster than the average SSD read rate. It's also four times faster than the average SSD write rate. By "average" in this context, is meant the SSD drive with middle-of-the-pack (Median) performance. How much do you actually need this performance? How much are you willing to pay? Where does reliability and noise, never mind price, feature in your decision? .

This confirms the other information in your post. It's a 550MHz / 256kB L2 cache / 100MHz FSB (ie PC100) / 1.65 volts Vcore. It takes about 15 watts max and most systems using them have a 90W or 145W PSU .

I'll second going with just one 128MB module for the moment, just to test. That's plenty fo a 500MHz P III running Win 98SE. The only downside is that it might slow Firefox down a bit if you start opening more than a couple of tabs. And, I would swap the Ethernet card for something else, just to test. I did not explain my Packet Collision remark, but I've seen a laptop grind to a halt due to a mis-configured Ethernet card. Swapping in and out HDDs and 40-wire or 80-wire cables is unlikely to make performance gains of the order you need. Those PSU voltages are way too low, so low that I doubt their accuracy. The answer to "Vcore is 1.68 - what should it be?" is anything from 1.60 volts up to 2.05 volts depending on exactly which processor model and stepping you have. 1.60 volt is usually for FC-PGA and the Slot-1's are more likely to be 2.00 volts, if it actually IS a 500MHz, or 1.65 volts if it actually is 550MHz. What does the string stamped on the top say? .

Aaaaah! My mistake, I forgot about the crippled console in XP Home. The User Accounts app in Control Panel only does half the job (as you already know) so a quite easy and relatively safe task in Windows 2000 or Windows XP Pro becomes very difficult. I don't know whether the Win 2000 version (lusrmgr.msc) would work or not but, like you, I wouldn't even try without an image backup. To try the "shuffle the profile" idea surely needs a total of THREE accounts: (1) the old account; (2) the newly created account; and (3) an account that is used to switch the profile in the ProfileList registry key from the old account to the new account whilst neither (1) or (2) is logged-on. Or at least, that is what Microsoft advises when copying a profile. Instead of (3), however, another copy of the OS can be used to load the SOFTWARE registry file into Regedit and it can be edited (and unloaded) from there. Once again, I wouldn't try this without an image backup. .

Another possibility is packet collision with the Ethernet adapter. Check the settings for Half/Full-duplex and for 10/100Mbps. Davicom mention this on their website, but I find their English somewhat difficult to understand! .

From the information you have given it seems that the advice of transferring the profile to a new user is almost exactly the OPPOSITE of what you are trying to do. However, I am confused by what you have done already. In your first post, second paragraph you wrote "so I renamed his user account" but in the next paragraph you wrote "The user profile entry (reached through My Computer properties) is named Marvin" which suggests that the user account still has the same name. If you open up the Computer Management console (which can be reached in several ways, includingreached via My Computer -> Manage... ) and look in the Users sub-section of Local Users and Groups, what do you have in the First and Second columns? From the information given in your post, it seems to me that you still have "Marvin" in the first column and the new name in the second column. Is that the case? (If not, then I have misunderstood what you have done). If I am right, then renaming "Marvin" is done by right-clicking on it in the first column and selecting Rename .

If you've read the http://www.hfslip.org/ page as given by TommyP and still don't know:- Yes No .

Whilst agreeing with the posts by CharlotteTheHarlot and by jaclaz, the goal-posts on this subject just moved with the release of exFAT for Windows XP. See the KB article for details and the download link. That would mean re-formatting the drive first, of course... .

Are you sure about that? KB954600 still installs strmdll.dll even when WMP9 is already installed. .

Note I was speaking about "the list", not this specific update. QFE's are indeed not public nor available on WU (unless we're talking about IE updates, in which case these are on WU but you won't be on QFE unless you do it manually once... I digress).I was speaking about the list too... And I am aware of forcing the QFE branch too... But none of this "clarification" of each other's posts is helping the OP, so I shall stop right there. The fact remains, as the public sees it, that there are updates released for Windows which do not make it to WU and they don't all fall into a single category when it comes to level of testing and support. How does one find these? There is no summary page or list provided by Microsoft. It all depends what one means by "optional" as well. There are websites dedicated to listing all the Hotfixes that fall into the "available on request" category -- but as for those that are freely available by just clicking the "Download" button (like the example already quoted) there is, paradoxically, less available information, so I have suggested one (slow) way: search the Download Center manually -- maybe with a search like http://www.microsoft.com/downloads/Results...rtCriteria=date. Also, if you can speak, or rather read, German, then http://patch-info.de (run by Ottmar Freudenberger) is very quick to report all manner of updates. .

Some are, some aren't. Although there are plenty of QFE's in the Microsoft KB, usually recognised by the words Hotfix Available and the fact that you have to email or otherwise contact Microsoft PSS to obtain them, there are also fully supported Public updates that, for one reason or other, are availabe from the Download Center, but are not offered by WU. One way to discover these (to answer the OP's second point) is to use the Download Center's own search facilities, which can be ordered to show downloads by reverse date (latest first) rather than by Popularity. This particular one already IS Public: http://www.microsoft.com/downloads/details...;displaylang=en Actually, this particular update has a chequered history, because the original vulnerability was only on Vista, not on XP. Fixing the AutoRun behaviour was incidental. Microsoft have privately admitted that releasing the non-Security update for XP with the same KB950582 as the Security Update for Vista was a mistake. Of course, since it was released in early 2008, everyone is now panicking about AutoRun because of the spread of the Conficker/Downadup worm and KB950582 on XP has gained a new Security status. Not exactly, there's no need to do anything because the update makes the change for you. See the HonorAutorunSetting information given HERE for the full explanation. .

More information in http://www.msfn.org/board/index.php?showtopic=128880 .

Peter Gutmann has added a second epilogue to his original paper criticising the article by Craig Wright. If you read both of the papers, it doesn't help matters to find that the acronym MFM now has TWO meanings: Modified FM (Frequency Modulation) and now Magnetic Force Microscope. Wright seems to confuse the two with a heading giving the second meaning, closely followed by a diagram using the first meaning. Gutmann also uses both meanings, but in the context in his paper it is clear which means which. That doesn't mean that Gutmann disagrees with Wright. On the contrary, Gutmann says: A comment at SecurityFocus has pointed out that an MFM does not have sufficient resolution to read anything off a modern HDD: the HDD bit density is twice as high as the MFM resolution. This supports Gutmann's last comment. I agree with jaclaz: the myth is thoroughly debunked and can now be regarded as 100% fact-free. .

It's not clear from your description, but do you have two DVD-ROM drives on one cable? If so are they both jumpered to CS (Cable Select) or one to Master, one to Slave? If both are set to CS, try setting one to Master, one to Slave. This once worked for me on a quite different board, but which uses the same AMD chipset. I know CS is supposed to work with an 80-wire cable, but in my case it did not. Rather than find out why not, I just took the route that worked. .

Yes, that's one of three tools I know of that you can use to disinfect with (the others being F-Secure's tool and Microsoft's own MRT version 2.6 - January 2009). Symantec have a write-up of their W32.Downadup Removal Tool HERE, but they do say that you should disconnect each computer from the network to use it:- The above instructions do seem to be written with home users more in mind, as I can't see large networks doing that to every computer one-by-one!! However, since the worm very aggressively protects itself and floods the network with half-open connections, it may have to be taken down or partially taken-down anyway. F-Secure's tool is available HERE. Of course, the infected computer will not be able to connect to either Symantec or F-Secure, since Conficker (a.k.a. Downadup) blocks both. F-Secure suggest using their raw IP address, which is ftp://193.110.109.53/anti-virus/tools/beta/. I have already mentioned Microsoft's removal instuctions in my previous post. They now have a new blog post with more information and useful links HERE. US-CERT have also issued an alert HERE Their images illustrate that, in a network, a missing MS08-067 patch is not your only problem. For example:- .

I too was unable to access msfn.org due to nameserver failures, since about one hour ago: ns1.msfn.org returned (SERVFAIL) ns2.msfn.org returned (SERVFAIL) Independently verified by OpenDNS .

It IS possible to spread Conficker (a.k.a Downadup a.k.a Kido) via any external drive/USB device, provided AutoRun or AutoPlay is enabled. I'm not sure how much you already know about Conficker, so apologies if you already know what follows... The worm uses three separate modes of propagation: Via Autorun.inf+a randomly-named file in the RECYCLER\... folder Via network shares (Inside a network it goes hunting for Servers to (re-)infect) Via the MS08-067/kb958644 vulnerability Most people seem to be making the mistake that once you have patched with kb958644 your troubles are over. Not so. At the very least you need to disable AutoRun and AutoPlay, then disinfect the entire network including all devices that attach to a USB port. Remember also that AutoRun is broken in Windows (including XP, 2003 and Vista), unless you have patched with kb950582, so a Group Policy to disable this may not work. See http://support.microsoft.com/kb/953252 for more information. Network propagation of Conficker includes increasing the max number of half-open tcp connections to 268435456; hunting for servers and users with NetServerEnum and NetUserEnum, password cracking, taking over the DNS API and blocking Microsoft and most well-known AV and Security websites...... The AV vendor F-Secure (who are, I think, bigger in Europe than in the USA) have plenty of useful information on this, including: When is AUTORUN.INF really an AUTORUN.INF? and F-Secure Malware Information Pages: Worm:W32/Downadup.AL If you cannot reach www.f-secure.com, then you are already infected with Conficker. .

It seems my infection information is waaay out-of-date. Currently some 9 million computers are infected. That does not include anything (e.g. memory stick or external drive) connected via a USB port. .

Not personally, but it is estimated that only one in three Windows networks are fully patched against it and some 3 million computers are actually infected. (That last figure varies, depending on who you believe.) Microsoft have even published a KB article with detailed removal instructions. See KB962007: Virus alert about the Win32/Conficker.B worm. KB962007 is well worth a read, to check if you've missed something... .