Dave-H

I'll give that a try, but I don't hold out much hope. I do think the problem is almost certainly the Silicon Image driver for the eSATA card. The Windows 10 driver supports GPT disks of that size connected to the card, the XP driver doesn't. It's still odd to me that the disk is being seen in Device Manager though. As there seems to be no driver for XP later than the one I'm using, I guess I'm stuck. I will have a search for a Vista driver for the card, if there's a chance that can work on XP it might be an answer.

Thanks @jaclaz. I actually already use Hotswap! and have been doing so for many years! I use it to unmount the drive in the enclosure before I de-power the enclosure. With a SATA connection instead of an eSATA connection, it does still work to unmount the drive, which is fine. The problem is with starting to use it, I used to just power it up and the system would mount it. Now that doesn't happen of course, I have to reboot for the drive to be recognised.

Thanks @roytam1! Previous thread locked.

Continues here -

LOL @jaclaz yes, happy days! The external enclosure I'm now using thankfully does not have the same problems as the one I was using. Normal disks work fine on the eSATA connection and the USB connection, with no compatibility issues. I do now have a cable to connect the enclosure to a port on the motherboard, bypassing the interface card. It works fine, but of course I'm losing all the hot-plugging advantages of an eSATA connection. I don't want to have the enclosure powered up all the time, and I used to be able to just power it up and the connection would be recognised and the disk mounted. Now I have to reboot, which is annoying. I guess you can't have everything in this world though!

Thanks very much @Andalu that's really useful! Am I right in assuming that the ASMedia cards you refer to don't exist as PCI or PCI-X cards? I have two PCI-E slots on the motherboard, but they both have graphics cards in them! This is because of my multi-boot configuration. One card is for Windows 98 and has no Windows 10 drivers, the other card is for Windows 10 and has no Windows 98 drivers! I can use both of them in Windows XP. There is another small slot on the board which Supermicro describe as a 'PCI-U' slot. This is intended for some sort of internal expansion card. It's also a PCI-E x8 slot, and I'm using it for one of my Silicon Image cards. Unfortunately the eSATA socket on the card is not physically accessible (I'm only using the SATA port on the card edge to connect my Blu-ray drive). I might be able to fit an ASMedia card into that slot, but connecting my eSATA disk enclosure would be very difficult without a lot of physical surgery! Am I also right in assuming that an ASMedia card would have no Windows 98 drivers? That may not be a real problem, as the Silicon Image card in the PCI-E slot at the moment doesn't have Windows 98 drivers either. If I'm only using it for the eSATA enclosure and the Blu-ray drive, that's not an issue as I wouldn't expect to use either of them on Windows 98 anyway. Thanks again for the help! Cheers, Dave.

Well all I can say is that I've been using Seagate Barracuda drives for quite some years now, and I've always found them to be very reliable. I've certainly never had one fail. I only ever used them as data drives, my system drives are all SSDs now of course. Thanks for the heads up though!

Yes indeed. We usually do it when new builds are issued, so if you could create a new thread for your next builds @roytam1 that would be good, and I will then lock this thread. Thanks, Dave.

I've now done another test, connecting the 3TB disk directly to the SATA port on the Silicon Image 3512 card, using a different cable of course. Exactly the same result, appears in Device Manager, but not in Disk Management, so it's not a problem just with the eSATA connection. My conclusion is that it's not a physical limitation of the 3512 card or its cabling, as it all works fine in Windows 10, but it must be something to do with the 32 bit driver that the card is using in Windows XP. Something is getting lost, which is preventing Windows XP from seeing the disk properly. Whether this is fixable, I don't know, but I'm not hopeful as it looks as if the driver I'm using is the last one.

Yes indeed, but it works fine with the eSATA/Interface card route too in Windows 10. That surely means that the hardware is physically capable of supporting the disk fully, the only difference being the driver that the card is using. It is 64bit on Windows 10, and 32bit on Windows XP, but I doubt that is the source of the problem. The Windows 10 driver is just a later version, and maybe supports things that the XP driver does not. That's why I'm trying to confirm that the driver I'm using on XP is indeed the very latest one available.

Thanks. I'm pretty sure that I've already looked there, but I will look again.

Of course, Windows 10 would almost certainly be better at intercepting and correcting errors than XP, but that's not the issue. The warnings I mentioned in the event log on Windows 10 were a bit of a red herring which I probably shouldn't have even mentioned here, and they've gone away now anyway. Something as fundamental as Disk Management not seeing the drive at all in one operating system and seeing it fine in another, surely can't be just down to a dodgy cable?!

Thanks, but if the hardware is suspect, why would it work perfectly on Windows 10? I also have a Silicon Image 3132 card shoehorned into the system. I can't connect to the eSATA port on that card because of physical problems (the machine's, not mine!) I tried connecting the drive to the SATA port on that, with the same result. I was hopeful because the BIOS on that card did report the correct disk size, which the other one doesn't, but in XP the result was the same. So, unless I can find another driver for the 3512 card, I don't think I'm going to get anywhere. All the later ones I've found so far are either RAID drivers, or they're 64 bit only. I doubt partitioning the disk into smaller partitions will help if XP Disk Management can't see the drive at all. This is what I'm seeing in Device Manager if I query the volumes on the drive. I guess it's a no-go. I think I can work out a hardware configuration solution, although it's not ideal. I will have to get a suitable eSATA-SATA cable to connect the external enclosure directly to the motherboard, and get another to connect the drive that was on the motherboard SATA port to the eSATA port on the card. That should be OK, although not very physically elegant!

Thanks everyone again. I've now disabled the Core Storage Volumes Driver service, and that error has gone away, with no apparent ill effects. I really don't think that there's anything physically wrong with the hardware. It all works fine in Windows 10 exactly as it should. I'm not too worried about the 'retry' warning I was getting in Windows 10, I think it's gone away now anyway, although I will keep an eye out for it. I don't think it's got anything to do with the disk not being seen at all by Disk Management in XP. The eSATA interface card is a Silicon Image 3512 SATALink controller. As far as I can see it is using the latest driver 1.3.71.1, from 2008. I will keep looking to see if there's a later one.

OK, back again! I now know what the problem is, but I don't know how to get around it. I didn't get anywhere with diagnosing the problem with the Core Storage Volumes Driver. Process Monitor and File Monitor didn't reveal anything relevant, at least not to me. However, it turns out that that driver isn't needed anyway, so I guess I will just disable it in the registry and forget about it. It may be needed when reading HFS formatted disks, but I think I'm very unlikely to ever need to do that. Using smaller partitions made no difference either. The problem turned out to be the drive enclosure connection. I tried connecting the disk directly to the motherboard, and it all works. I can see and use the whole 3TB partition, so the Paragon driver is working fine. This is strange, because connecting the disk using the eSATA connection in the caddy works absolutely fine in Windows 10, and indeed the alternative USB connection works fine too. Exactly the same hardware configuration doesn't work in XP though, Disk Manager can't see the disk at all. I obviously don't want to use the disk physically mounted in the machine, the whole point is that it's an easily removable archive and system backup, which I can grab if there's a fire! Can anyone think why it would work in Windows 10 and not XP? I will see if I can find a later driver for the interface card. It's not as if XP isn't seeing the disk at all, it is. It appears in Device Manager fine, just not in Disk Management!

Thanks @D.Draker and @jaclaz Sorry, yes of course, wrong terminology there, I'm very good at that if you remember! The enclosure is a high quality one and not that old, so I would be very surprised if it was that, also the eSATA cable was replaced recently with a new good one too. The same hardware worked fine with the previous 2TB disk. I'm hoping the error message I was seeing has now gone away, I updated the driver on the eSATA interface card, and I hope that's fixed it. This was only on Windows 10 of course, so actually off-topic here. I only mentioned it in case it gave any clues. The main problem is getting XP to see the drive at all in Disk Manager! It could be because the Core Storage Volumes Driver isn't running, but as I said, I have no idea what that driver actually does. I will investigate with FileMon and Process Monitor as suggested. If I don't get anywhere with that, I will try formatting with smaller partitions and see if the disk is then seen. Thanks again for the responses. Cheers, Dave.

@jaclaz @Cixert Any input here? The only thing apparently wrong with the HFS installation is that I'm getting an error message on every boot in the System Event Log. It says that "The Core Storage Volumes Driver service failed to start due to the following error: The system cannot find the file specified." I don't know what that service is supposed to do, I can't find much information about it. The strange thing is, what file can't it find? Its registry entry points to system32\drivers\csvol.sys and csvol.sys is present and correct in that folder, where you would expect it to be. I have checked that it's the correct 32 bit version, and it is. This is very puzzling, any help would be appreciated! Thanks, Dave.

Sorry to bump, but I have a problem with a 3TB disk I've just bought to replace by old 2TB archive disk, which is showing signs of dying. The drive is in a StarTech external enclosure, but connected by eSATA, not by USB. The eSATA connection is via a Silicon Image add-in card. I have formatted the drive as GPT using Windows 10, and it works fine there, although I'm getting occasional errors in the event log - "The IO operation at logical block address 0x5bc5e0 for Disk 4 (PDO name: \Device\Scsi\SI31121Port5Path0Target0Lun0) was retried." I'm not sure if that indicates a problem or not. The main issue is with Windows XP of course. I have installed Paragon HFS for Win v.10.5.0.95, which seems to be working OK. However, although the disk appears correctly in Device Manager as a disk drive, and shows no errors unless you try to populate the volumes on it, it does not appear at all in Disk Management, so I can't assign a drive letter to it. Anyone any idea why that might be? Thanks, Dave.

Working again now!

Yes, it seems to have failed again. I'm on it!

@D.Draker Should be fixed now!

@D.Draker Sorry for the delay in responding. Yes, there is a problem, I can't make a new topic either. I have reported it to xper. I guess it's something which went wrong after the last server outage. Cheers, Dave.

@D.Draker Is the problem still there?

Just as an aside, I have for quite a while now been using the latest script version of 'Fluff Busting Purity' on 360Chrome, with OrangeMonkey, which still works perfectly of course. As long as scripting extensions still work, I hope that's going to be an option for quite a while yet with some extensions.

I tried with an incognito window with uBlock enabled, and it still crashed. I also tried connected through ProxHTTPSProxy, and it still crashed. Sorry I should have mentioned before, as someone did earlier, that it works fine in Firefox 52.9, rather unexpectedly! Obviously it's not a site that many people would use, but I always think these things are worth reporting, as what fails on one webpage could easily fail on others.