installing linux to d: vs partition on c:
 

So my c: drive is an ancient 10GB and contains winXP along with a bunch of other misc. apps.. I think it's got something like 1GB of free space, and that's after a defrag. My d: drive is a nearly-fresh 100GB only just beginning to fill with media files. I have searched for threads on installing Linux to a separate physical drive, and while I'm pretty sure a lot of people do it, I've only found a couple bits of info. I just want to know if there is anything different about it vs installing to a partition on c:.

Mainly, where would LILO physically install, c: or d:? And once I am at the bootloader screen, how does it affect boot-up with Linux being on a drive other than c:? Does it behave in exactly the same way it would if Linux was on c:?

Once I have these questions answered, I will hopefully be creating a partition on d:... then I will probably be back to ask more questions:)

By the way, I will be installing PCLinuxOS (from my current live CD) if it makes any difference.

Lilo (Grub?) will install on C: (well, no in /dev/hda1, in /dev/hda instead) and it doesn't matter where the Linux kernel is, it is automatically configured in the install process, don't worry about that.

BTW, if D: (/dev/hda2 or /dev/hda5 or /dev/hdb, not sure about what you meant) is NTFS, you'll have to format it and you'll lost your data, so don't save your data there.

D: is my 2nd hard drive. I will have to format it? What about people who install Linux on the same physical drive as Windows and all their other data?

Some distros used the free space that's left in d: or whatever. Others don't and you have to partition yourself (and maybe lose your data). But I don't know about PcLInuxOS. Anyway, you'd better backup your data before doing anything.

C and D are Windows terms and must not be used in Linux forums.....;)

Seriously....
**Some** bootloader MUST be on the mbr of the first drive or on a floppy.
You can put any number of OSes on a single drive--limited only by space.

One approach to consider:
1. Convert 10GB drive to a hood ornament or paperweight. "Ancient" = going to fail one of these days.
2. Buy a new drive--40GB or so (Cheap)--and set it up with multiple partitions to install Windows, and maybe 2 or 3 flavors of Linux.
2A. If $ are tight, use part of the 100GB for OSes, and the rest for data.
3. Format the 100GB drive as FAT32 so that Linux and Windows both see it easily.

You did not mention where your backup is. Naturally, you back up all important data before doing any of this.

Yes, my apologies for using heathen conventions in these sacred halls:)

Okay, I like the idea of just having all OSes on a single (partitioned) drive.. and believe me, I fear for the life of my old 10gig (it's been in active service for 7+ years). I guess I was hoping for a quick fix. I understand that if I really wanted to, I could go ahead with the 2-separate-hard-drive plan, but maybe you have me convinced to move everything to a new hard drive.

So on that note, FAT32 is the format of choice, yes? I wish I understood this better. Can someone tell me how to find my hard disk properties in PCLinuxOS? I can navigate to it in a similar fashion as I do in Windows, but it doesn't tell me diddly - just it's location and that it's mounted. I may have to reboot into Windows and have a look. For all I can remember, both my drives may already be formatted FAT32.

Besides, C: and D: refer to partitions and not to drives.

Two distro's that I've installed. Mandrake and SuSE have had no problems resizing the Windows partition to leave room for Linux. I did this on my laptop which originally had a single NTFS partition with Windows XP. They still recommend backing up the contents first, just to be safe. Defragging the drive first is also a good idea.

Fat32 would be the choice for a common data drive that both OS's can write to. However, you don't want to use it to install linux on. Use something like Ext3 or ReiserFS.

A fat32 drive will not hold files larger than 2 GB, so if your media files are larger, you don't want to use fat32.

ho boy.. okay, I'm sorry to ask Windows questions here.. but here goes.

I've rebooted into winXP and had a look at my drive properties. Some things have popped up that I had forgotten about. First the good news - there's actually more like 3GB of free space an the old 10gig. Okay, good. But why on earth do I show TWO separate drives in explorer for the same drive? It shows C:, formatted NTFS, with 6.45GB of data, and E:, formatted FAT32, with 6.2GB of data. These are both the same drive! So how is it formatted, NTFS or FAT32?

3GB of free space ON the old 10gig, sorry.

Because Windows sometimes does weird things......

Can you run fdisk from your Linux?

3GB free space on the existing 10GB antique is going to be tight---return to previous suggestions about new use for this.....;)

Ah, well, you're right of course... hrmph. Until I migrate everything to the other hard drive or a new one altogether, I'll just keep running the liveCD. Fiddlesticks.

Both NTFS and FAT32. It is possible to divide a single IDE drive into 64 seperate partitions. When windows boots it will automatically assign c: to the partition with the boot flag set active. d: will be assigned to the first partiton of the next physical drive found and so on for all physical drives. Next it will assign drive letters to the rest of the first physical drive partitions and then go to the next physical drive. You only have one so it is e:. BTW with XP you can override the automatic settings.

So you have two partitions on the first physical drive. c: is NTFS and e: is FAT32. It looks like your 10GB drive might be bigger then you remember.

oops.... having minor problems here.

Sorry, I just got back to my desk after a day away.

Actually, no. This drive is most definitely a 10gig. And the letters c: and e: are most definitely assigned to the exact same partition. I guarantee it. But I don't understand it.