Nomen

To recap: I have a few Soyo socket 478 motherboards with Intel 82801DB IDE controller. I have installed long ago the IAA which replaces ESDI_506.pdr with IntelVSD.VXD and IntelATA.MPD for the IDE drivers. I attached a 250 gb SATA drive using the small IDE-SATA adapter (the adapter plugged directly into the second IDE port on the motherboard) and from DOS (format + fdisk) I have formatted the drive into 4 volumes (65/65/65/44 gb). During POST startup the drive is detected and displayed properly by the bios in the secondary master IDE position. In Windows I have copied several hundred MB from my existing C drive to the last volume (44 gb) on the sata drive and these files seem to be ok. I am sure this would not have worked if I was using the original ESDI_506.pdr and maybe there are other alternatives to the 48 bit LBA problem when it comes to IDE drives, I've never attached to a win-98 system any IDE drives larger than 80 gb. For me it is somewhat rare to have any IDE drives larger than 80 gb but I do have some 320 gb IDE that I was using to build some XP systems 10 years ago (I still have 1 or 2 of those drives still sealed in mylar bag). But for some IDE controllers the IAA files seem to work ok (this is the first time I've ever tested them under these conditions of having >137 gb IDE drive).

I just tried Opera 12.02 and got different results the first time - it can't connect to bgp.he.net because of security protocol. I went in and enabled all protocals in opera and got the same error. I then renamed my hosts file (so no hosts file) and restarted Opera and it worked, and I could get info on IP's without getting the javascript message. I then restored my hosts file and restarted opera and it is still working so I don't know what's going on but I'm happy it's working again.

I bought a few of these small IDE - SATA boards on Ebay recently. They are bi-directional adapters and can be plugged directly into either a motherboard IDE connector (to provide a SATA port to connect to a SATA drive) or can be plugged into an IDE drive (to allow an IDE drive to be connected to a motherboard SATA connector). I know there are lots of IDE/SATA/USB adapters around - this board has no USB connectivity. I'm wondering about connecting a large SATA drive to an IDE port on Win-98 systems and the 137 GB problem. A system I'm thinking of trying this on has an Intel 82801DB Ultra ATA Storage controller and is using IntelVSD.VXD and IntelATA.MPD for the IDE drivers. I do not appear to be using ESDI_506.PDR. I recall that Intel made a set of drivers (Intel Application Accelerator?) way back for certain controllers and (I think) it had the side effect of not having the 137 gb problem (28 bit LBA vs 48 bit LBA). Anyways, I'm going to try a spare 250 gb SATA drive with one of these boards and see if I can read/write properly to the entire drive. I might split the drive into multiple 80 gb volumes and then make sure I can access each volume correctly. Other than that, I have some motherboards that don't have IDE ports where I run win-xp and 7 and have a bunch of old IDE drives to clean up and off-load files and this seems like a good way to connect these drives.

I was sort of expecting more activity in this thread, but a lot of the old timers must be gone. I was expecting someone to say something about what (if anything) is going on with the RLoew files (in general) or the sata drivers in particular - but nothing. Or is that happening in the special projects sub-forum? I've been experiencing problems (freeze-ups when accessing SATA drives) with my current setup and I'm exploring alternate solutions for connecting SATA drives to my primary Win-98 system. I have a bunch of SATA driver files scattered across various drives and might have to pull them all together and start experimenting again. I picked up a couple of 4-port SIL3514 boards (2-port SIL3112 boards are hard to find, even 2-port SIL3512 are hard to find) and will mess around with them at some point. I also have some small converter boards that I'll post about in my next post.

I used to stay on top of this but that was a while ago. I'd like to pick up a few PCI sata controller cards and was wondering which of these now have working win-98 drivers and which ones don't. I seem to recall that the only working controllers were the 2-port 3112 cards, not the 4-port cards and not the 351x cards, but maybe I'm wrong on that. And is there now a public-domain RLoew sata driver that works with these Sil controllers?

I obtained the above SST files and ran the updroots thing - got no messages after each command (so I assumed no errors). Does this take effect immediately or is a reboot required? Is there any sort of on-line test or check to see if these certs are working correctly?

Earlier in this thread someone posted a link to rzbrowser-tls12-20180504.7z. I had already downloaded and installed that package in Sept 2018 so I don't know if there's anything newer. I still use FF2 on my win-98 system and haven't really experienced any reduction in usefulness over the past year for the sites I visit. One thing that has been a downer for me is that I somewhat often use https: //bgp.he.net/ with Opera 12.02 and at some point a week or two ago that stopped working. I can bring that page up in FF2 but when I enter an IP to check I get a "this site requires javascript so please turn it on" message (and I'm pretty sure I have it turned on). If anyone can perform an IP lookup on bgp.he.net using any browser running on win-98 please tell me.

Looking at web searches it's clear that I'm not the only one that has or would want to continue to use ghost 2003 to clone their windoze 7 drives but from one of my above posts it's clear that starting with vista and certainly with 7 that there is something about the ntfs boot or disk structure that ghost 2003 (and even newer versions) don't seem to know about or replicate correctly. Ghost is not a sector-by-sector copier so it has to know about the logical structure of the file system it's dealing with when it clones drives. It would be nice to find some instructions on what *exactly* to do, with what tools (bootrec.exe, bcdboot, bcdedit, bootsect, startrep, or other, etc) once a ghost 2003 clone has been made. Maybe it's all here: https://www.veritas.com/content/support/en_US/article.100001014 or here: https://www.digitalcitizen.life/command-prompt-fix-issues-your-boot-records ? On a slight tangent, I've looked at the radified website to find something definitive about the max hard-drive size that ghost 2003 can handle. It seems that it can't clone a 2T drive. I find it odd that I can't find such a basic parameter like that in the ghost faq on that site. I wanted to create an account there to post a question about that, but I can't seem to find a way to create an account.

Well I put the clone drive (as the only drive) in a PC with a CD rom drive and booted a win-7 CD and selected the repair and it did something very quickly and then put the drive back into the target pc and it booted up just fine. I would like to know how to "fix" a drive like this in this state by slaving it to a working win-7 PC and performing what-ever system-level task or operation to the boot records or what-ever but I have not seen any such instructions on how to fix a drive like that under those conditions. Happy new year, by the way. It must be new year somewhere in the world by now.

Still reading your post, but just to comment: > Allow me to doubt that you had at any moment two partitions active, as this is - >besides not allowed - almost impossible to obtain with *any* "Normal" tool Ah, yes. I was wrong. Disk Management shows in real time that when using command line diskpart to mark partition 1 or 2 active that only 1 of them becomes active - the other one (if active) becomes non-active. So when the "system reserved" partition is Active, I get the first error in this thread upon boot (0XC000000e can't find required device). When the second (and only other) partition is marked active, I get the "bootmgr" is missing message. When this drive is slaved to active Win-7 system, system reserved becomes E drive and the other partition becomes F. E drive has bootmgr, bootsect.bak, $recyclebin (folder), Boot (folder), System volume information (folder). So I decide to copy bootmgr and Boot (folder) from E to F. I can copy the Boot folder, but when I try to copy the bootmgr file (which I'm told is 382 kb) I get the message "You'll need to provide administrator permissions to copy this folder - Local disk" ? So all of a sudden bootmgr becomes a folder? In any case, bootmgr and boot folder is present on "system reserved" partition, and that partition is marked active, and windows resides on the other partition and everything is there but something else must be wrong because the target system doesn't boot with this drive. Could the slaving of this drive to another PC have imposed drive lettering to these paritions that is screwing stuff up and must be removed? I still would like to know if bootrec.exe can be used to fix this, without having to somehow boot a win-7 install disk and mess around with recovery option.

I attached the clone drive via usb adapter to a running win-7 system and used drive management to set the second partition to active. The drive has 2 partitions, the first one being very small (100 mb) named "System Reserved". I put the clone back in the target PC and tried booting it again, this time I got the message "bootmgr was missing. Press cntrl alt delete" I did the USB thing again with the drive and now set the first partition to active as well. So they were both set as active. Put it back into the target and booted - still got the bootmgr message. Maybe only the first partition must be set active - not the second? How does win-7 deal with 2 active partitions on the boot drive? Putzing around the net I found this: "Bootmgr was introduced in Windows Vista which Microsoft released to the public on January 30, 2007. In previous versions of Windows, that's before Vista; a program referred to as NTLDR was the boot manager. This means that Windows XP users won't get the bootmgr is missing error. Bootmgr is essential for the boot sequence to begin; without it, the operating system will not load. In other words, if ‘Bootmgr is missing’ then your computer won't boot. In this post, we discuss how to fix ‘Windows Bootmgr is missing’ errors in Windows." https://www.ghacks.net/2017/05/02/fixing-bootmgr-is-missing-error-in-windows/ What I'm looking for now is a way to fix this boot mgr thing without having a win-7 CD handy but instead do it while the drive is attached to a running win-7 system via usb. Is this possible? I guess what I'm looking for is the equivalent to the dos "sys" command. It doesn't look like the "diskpart" command shell can do this? Maybe this will work? Windows USB/DVD Download Tool https://www.microsoft.com/en-us/download/details.aspx?id=56485 Seems that BootRec.exe would work the best, but I can't find the damn file anywhere! Ok, I've expanded winre.wim from the cloned drive and have a copy of that entire thing on my primary drive on the PC that the cloned drive is slaved to via USB. Looking at the commands for bootrec /fixmbr I don't know how to tell bootrec *which drive* to operate on... ?

I've cloned XP dozens of times back in the day, with Ghost 2003, and the clone always booted. I've rarely cloned a win-7 drive, but it seems that every time I do, the damn thing won't boot and I have to putz with a setup CD or drive tool of some sort to "fix it", but I never figure or am never told what the hell the problem was. An example error is: 0XC000000e The boot selection failed because a required device is inaccessible. Why isin't Ghost cloning this thing so that it's bootable? Is this a known thing for ghost 2003 or is there some ghost setting that I don't have right? What bit or what-ever on the drive is not flipped in the right direction to cause this?

This is what I've always used: [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\WindowsEmbedded\ProductVersion] "FeaturePackVersion"="SP3" [HKEY_LOCAL_MACHINE\SYSTEM\WPA\WEPOS] "Installed"=dword:00000000 [HKEY_LOCAL_MACHINE\SYSTEM\WPA\WES] "Installed"=dword:00000000 [HKEY_LOCAL_MACHINE\SYSTEM\WPA\PosReady] "Installed"=dword:00000001 I last used that a few days before the supposed end of the update support for XP (July?) while creating several new master XP installations for several motherboards and it worked fine. But in any case - can I get an answer to the question -> is WU still updating XP-POS systems?

A couple of days ago I was testing Klonezilla and used to to clone an XP master drive. The drive had SP3 but hadn't been powered up since october 2012. It was up-to-date at that time. I started the clone and was surprised to see, within a few minutes, the gold shield in the task bar telling me there were updates available. Seemed to be a few dozen updates, the last on the list being the XP end of life notification. I selected all except for the EOL notification and they downloaded and installed, and then no other updates were offered. I then ran my POS reg file (the one that creates 4 registry keys) and restarted several times but still no other updates were offered. I was thinking that maybe I'll re-clone the drive and then run the reg file before I let it try accessing the WU server. Or is what I'm seeing known behavior - that MS is no longer offering POS updates through WU?

Also mentioned here: Only 44 hits in all the interwebs for the quoted phrase "Target state is Absent. Client id: Deepclean." I see that I have a DeepClean.log in c:\windows\logs\CBS dated nov 16. It only contains entries dated Nov 16. These are the first few lines of that file: =========== 2019-11-16 20:06:21, Info DISM Service Pack Cleanup UI: PID=8432 Superseded Service Packs 0 - CScavengeCleanup::GetSpaceUsed 2019-11-16 20:06:35, Info CBS DC: Ensuring the online components hive is loaded to load maps... 2019-11-16 20:06:38, Info CBS DC: Clearing cache... 2019-11-16 20:06:38, Info CBS DC: Finding superseded packages... 2019-11-16 20:06:38, Info CBS Skipping: Microsoft-Windows-CodecPack-Basic-Package~31bf3856ad364e35~x86~~6.1.7601.17514 due to applicability ============ Would like to know what initiated or triggered that activity, based on what settings, and what the results are for future system stability / functionality. This install of windows was created using 7lite (or what-ever it's called) back in oct 2016 and had all known/good kb's rolled in and has not performed a windows-update check since then and has automatic updates disabled.

I'm flipping through my Win-7 event viewer and I see (as of Nov 16) a bunch of entries in the setup logs as follows: Initiating changes for package KB3087039. Current state is Installed. Target state is Absent. Client id: Deepclean. A reboot is necessary before package KB2532531 can be changed to the Absent state. There are dozens if not a hundred or so of each of those, with all sorts of KB numbers. What's going on here?

I'm putzing with a new Synology NAS and was wondering about NFS for XP. When it comes to SMB, XP is only capable of SMB-1 and no service pack seems to exist to give it SMB-2 capability. When it comes to NFS, there seems to be something called "Windows Services for UNIX" for which the MS links no long work and archive.org does not seem to have them. So I'm wondering if they would be on any MSDN or technet CD's. If this package exists on MSDN then I will likely have it, but I'd need to know which CD to look for. Any ideas? Anyone ever get NFS working on XP?

I'm getting this protocol error again: Firefox can't connect securely to (...).storage.googleapis.com because the site uses a security protocol which isin't enabled. This is with Roytam's 5/4/2018 firefox with DLL's copied from Retrozilla 2.2 (2/23/2019). As mentioned just above, I had seemingly fixed this error by copying all files from root of Retrozilla folder to the Firefox folder, copying over all DLL's with the same name. This allowed the content from the above URL to be displayed. Just this morning I'm seeing this same error again. I have no googleapis cookies, I've cleared the firefox cache, and my system time and date clock are correct. I can copy the link to the offending googleapis file (its a jpeg) to Opera 12.02 and it is rendered just fine. It is also rendered just fine on Netscape Navigator 9.0.0.6. With netscape 7.2, I first get a "website certified by unknown authority" which according to Details is Google Internet Authority G3. I can accept the certificate, which I do, and then I get this error: Netscape 7.2 and (...).storage.googleapis.com cannot communicate securely because they have no common encryption algorythms. Retrozilla 2.2 can display the image with no protocol issues. I had both Retrozilla 2.2 and Roytam firefox (with retrozilla DLL's) side-by-side while viewing https://www.ssllabs.com/ssltest/viewMyClient.html and they seem identical. Opera 12.02 has a much shorter list of Cipher Suites, but all the ones it is is showing are also showing on Firefox. But Opera is supporting "OCSP stapling" and Firefox is not (neither is retrozilla). Any ideas how to get to the bottom of this? Are security protocols and ciphers negotiated based on browser user-agent? Edit: And today, looking at the same website that has all these storage.googleapis jpg files on it's site, everything is working fine again. What is going on with this?

For quite a while I've been running Firefox Community edition (Roytam's version) compiled on 5/4/2018. See page 17 of this thread. Just today I began to get security protocol errors on some components of a website I regularly view. The components are small clip-art or jpg elements that are hosted on storage.googleapis.com. I can post the URL of an example of one of these if asked. I tried one of those URL's on Opera 12.02 and it had no problem. So I don't know what security protocol or cipher googleapis is using for this, but what-ever it is, opera 12.02 has it. Anyways, I've just downloaded and installed this newer version of Retrozilla, and it *does* work on these URL's, so there must be some new protocols or ciphers added to it. My question now is - how do I get this new install of retrozilla to look like my current Roytam version (which is installed in programfiles / mozilla firefox and was not touched or modified by the installation of this newer version) ? I would like my bookmarks, search window, menu bar with quick links to be imported / copied to the new install. Is there an easy way to do this? Like copy some files over to it? edit: I copied all the files in the root of /program files/retrozilla/ into my existing /program files/mozilla firefox/. Just copied the files in the root, not the sub-folders. These files are mostly dll's. So I now have retrozilla.exe and firefox.exe in that folder. I ran the firefox.exe (5/4/18, 7128 kb) and it works fine. It doesn't give me the security protocol error that I was getting yesterday. I note that retrozilla.exe (2/23/19) is only 156 kb. It doesn't run when I launch it from /program files/mozilla firefox/. It looks like the only protocol difference is that Retrozilla 2/23/19 has support for 2 additional algorithms: Signature algorithms SHA512/RSA SHA512/ECDSA Those functions must be contained in a dll that firefox.exe is (now) using.

I don't think that's why I was seeing .Net updates. I performed a new XP install and activation today (july 20) and got 130 updates from M$. I then did the POS thing and only got 30 more updates (on other systems I've gotten 60). I don't think any of them were .net. On the systems where I did get .net updates after doing POS, those systems had installed .net (version 2?) as part of installing the video driver (some extra stuff other than the drivers got installed, and they came with some version of .net). On the system today, after the 30 POS2009 updates, no more were offered. I then installed a bunch of .net versions (2, 3.5, 4) and after that I got a whack more kb updates, and they were almost all .net (a few weren't). Not to get side tracked too much, but what other stuff should I install (that I might need or want some day) on a new XP install with POS that would trigger more KB updates to be offered (assuming the update servers are still up) ? Are there any MS Office updates that are only offered to a POS system?

Yesterday I installed XP on an AMD-based ASROCK motherboard (again, a board with no floppy controller). This one had a single SATA controller, and all ports could only be set to either IDE or AHCI. So I installed XP with the ports set to IDE. After all other drivers were installed, I added a SATA pci card. XP found the card and I gave it the AHCI drivers for it. I then shut it down, moved the XP drive to the card, set the on-board SATA ports to AHCI and restarted. XP booted up just fine, and now it detected the on-board SATA controller. I looked at the device ID string for controller, and searched all inf files for the string in all the driver files I had downloaded and expanded from Asrock. I found only 1 file buried deep in one directory tree that had the matching device ID. It worked, and I installed the driver. Shut down the computer, removed the PCI card, reconnected the drive back to the on-board controller, XP started just fine. Only after all this, with the drive controller in AHCI mode and all other drivers installed, do I perform XP activation (I used telephone method - not smart phone option). It would have taken me hours of trial and error to find the right driver for a grub-based F6 floppy install. Do all updates (about 130, avoid WGA) then run POS reg file, get a bunch more. Why does MS not offer any .NET security patches/updates for regular XP-SP3 like they do for POS2009?

I could scream. Scream, I tell ya. I finally realized that I'm putzing with Sata/AHCI drivers for a controller that my hard drive is not even connected to! This motherboard has a Marvell 2-port controller that I have set to AHCI in the bios but my hard drive is connected to an "on-chip" controller that has 6 ports, the first 4 of which can be set independantly of the last 2. So I finally clued in that I will connect my hard drive to port 5. Port 5 and 6 set to IDE in the bios. I set ports 1-4 as AHCI (SATA). I boot XP, and bingo - it finds new hardware. It loads some new / different driver files and *now* it has the drivers for the "on-chip" controller. I shut down and connect the hard drive back to port-1. Port 1-4 is still set in the bios as AHCI/SATA. I boot back up, and XP has no problems. The Marvell controller I was putzing with the drivers for the floppy F6 XP install turn out to be eSATA ports on the back-plate of the motherboard. So - it seems you can get XP running in AHCI mode when it was originally installed on a drive in IDE mode if you can get the SATA drivers installed. I was lucky my board has a multi-channel controller where some ports can be in IDE mode and others in AHCI/SATA mode. If that wasn't the case, then another way to do this would be to temporarily install a PCI sata card, have XP see the card as a new device and install the drivers, then reboot with XP drive connected to the PCI card. But before rebooting, set the on-mother board SATA controller to AHCI/SATA mode. That way, XP will boot and "see" the motherboard sata controller and then you install the drivers, power down, pull the PCI controller out and re-connect the drive to the motherboard sata controller.

Ok, I've just read the manual. I think I'm gonna feel like a dummy. If you want, the manual is here: http://download.gigabyte.us/FileList/Manual/mb_manual_ga-990fxa-ud3_v.4.0_e.pdf Sata/AHCI instructions start on page 70. They say to copy the files to a USB floppy drive (?). I wouldn't have thought that would work. But the main thing is that I'm supposed to select two drivers during the F6 session. Perhaps that's why it's asking for the floppy a second time. I'm going to see if I have a USB floppy drive somewhere, but I'll try this again and select both drivers.

I made some progress but in the end it didn't work. One thing I did was use the following sequence of grub commands: find --set-root /floppy.img map /floppy.img (fd0) map --hook chainloader (fd0)+1 rootnoverify (fd0) map --floppies=1 boot I tried your instructions - and they did work. I was able to get a directory listing of A drive. But then I re-booted and the floppy A drive was gone, so I thought there was a problem and I needed to do something different with grub. So that's when I stumbled on the above code. I was under the impression that the floppy image would persist (ie survive an actual re-boot) but it seems it doesn't. During XP installation if you press F6 and don't have your A drive setup working and you abandon the install, you now have a dual-boot drive (DOS, and Windows setup). At that point I choose DOS, do the grub thing, get the dual-boot choice and now choose Windows setup. Very soon at that point windows lets you press F6, and bingo, it sees a floppy disk with SATA hardware choices. I had to mess with creating the right files and oemsetup.txt and stuff to get that right. A few times I make a choice and get a strange error, or a certain file can't be found. Ok, I get past that and it continues and I get the EULA screen where I have to press F8 to continue, and then it wants to restart and says the DOS mode install is over. But soon after restarting it says "put the floppy in the drive A and press enter" and I keep pressing enter and nothing happens. I can only quit at that point I think. I was reading somewhere else where you are asked twice for the floppy and you can set up a ramdisk in DOS and it will work the first time but it won't work the second time (because dos is not invoked hence no ram drive). I thought grub would make MBR disk changes so that the virtual floppy image would always be present during a boot, but it doesn't seem to be - or there are more commands in grub to make that happen than what I've been able to find. I would also like to know how to use grub to check if the floppy image is "contiguous". What I've found about that is hard to follow / execute.

I've never used grub4dos, so yes if there are instructions how to do this then please show me where. Thanks.