PDA

View Full Version : Adding 2nd hard drive



cameronj
07-10-2002, 01:40 PM
Dunno if this qualifies as an expert question, but so far after 4 days on that other forum no one has had a clue, so who knows. Anyway, here's the copy and paste, its long and involved. Any help is GREATLY appreciated.

--

Followed Hinsdale's instructions for my Sony SVR2000. 30 hour unit, adding a second drive (80 gigs). Throughout the install process the drive was reporting 33 gigs in the bios. I ignored it (since I didnt see anywhere in the instructions showing how to fix it) and just went along with the install.

Sure enough, on install in the Tivo all I had was 63 gigs! Aieee! Can anyone give me help with this? I did have to qunlock, but I presume that had nothing to do with it. I'm going to go back into the bios and check things out, but I dont really expect to find anything. I have a maxtor 80 gig drive. Thanks

Cameron


Yesterday 12:56 AM

cameronj

Following-up: I put the drives back into the computer, and in the bios the old Tivo drive is back to reporting 9 megs! Is that normal? I figured qunlock did its thing, and that was that. Does the tivo set the drive back to locked status?


cameronj

Get a load of what I just found:

"The Maxtor Diamond Max drives 4K080H4, 4K060H3, 4K040H2 (aka D540X-4K) are identical to the drives 4D080H4, 4D060H3, 4D040H2 (aka D540X-4D), except that the jumper settings differ. A Maxtor FAQ specifies the Master/Slave/CableSelect settings for them, but the capacity clip jumper for the "4K" drives seems to be undocumented. Nils Ohlmeier reports that he experimentally finds that it is the J42 jumper ("reserved for factory use") closest to the power connector. (The "4D" drives use the J46 jumper, like all other Maxtor drives.)

However, it may be that this undocumented jumper acts like the IBM jumper: the machine boots correctly, but the disk has been clipped to 33 GB and setmax -d 0 does not help to get full capacity back. And the IBM solution works: do not use any disk-clipping jumpers, but first put the disk in a machine with non-broken BIOS, soft-clip it with setmax -m 66055248 /dev/hdX, then put it back in the first machine, and after booting run setmax -d 0 /dev/hdX to get full capacity again. "

The drive I have IS one of the ones in the list above, and it DOES 'clip' the drive at 33 gigs.

Can anyone else help me now?

For what its worth, the jumpers look like this:

. . . - .
------|--
. . . . .

The vertical line represents the current jumper config, which enables slave operation and the horizonal dashes are the only way I could get it to see my spacing.

cameronj
07-10-2002, 01:41 PM
cameronj

Holy moly!

What a pain in the ass! Turns out the Maxtor "Maxblast" software releases their drives from the Linux x86 drive size limitation of 33 gigs. Guess that's why you buy retail drives, not OEM.

Check this out though - the disk that I had been using for a boot floppy, since I didnt have a single other one... Maxtor Maxblast II (reformatted of course).

Oh the irony is just thick...

Good thing I waited before posting this, because the software did not work.

I need some linux help now.

Apparently, according to the site that I took the big quote from before, "On recent Maxtor drives the call setmax -d 0 /dev/hdX will give you max capacity again. However, on slightly older drives a firmware bug does not allow you to use -d 0, and setmax -d 255 /dev/hdX returns you to almost full capacity. For Maxtor D540X-4K, see below. "

Now, I've downloaded "setmax.c" no problem. HOW oh how to I get from that file to running it? Do I just boot into linux (I have a CD that can do that, I think) and run the command line? It seems to me that the C file should be compiled or something, but I have no clue about linux... HELP!!!

cameronj

Grr... wonderful... I decided to pack it up for the night (I KNEW i should not expect all the stories of 30 minute upgrades to apply to me) and plugged in my old drives as they were before... Guess what? Primary master and slave fail. I went into the bios to autodetect, and it detected the primary drive perfectly, the slave is 80 gigs and was found to be 80 megs. I crossed my fingers and hoped that was good enough, but on reboot I got the same error - both drives fail.

Ug!

Finny

Cameronj,

Finny here. I actually did get mine to work & was amazed how easily it worked. I originally was using MFStools ver 1 which I had from a previous TiVo upgrade that I did.
First off, I found out that the Tech support guy at West Digital didn't know what the hell he was talking about. He told me that a "special" jumper setting to use which didn't work at all. Once I set the jumpers back to the way they were listed on the drive & I made a boot CD of MFS Tools ver 2, then it all fell into place.

Two things: first off, you may have to use qulock more than once. Sometime once you reboot, the drive will set itself back. If the drive is not reporting the correct size: DO NOT PROCEED. I have an old shitty BIOS, & once I had the jumpers correct, the qunlock did it's magic & the MFS TOOLS did the rest. It seems that the Linux boot disk will recognize the full size regarless of the BIOS limitations. At least that's how it was on my computer which normally can't recognize anything larger than 32gigs. Using the MFS Tools v. 2, everything worked.

Give it a try. Good Luck.

-Finny

cameronj

Weird. In my readings I definitely came across a documented limitation of older linux distributions of 32 gigs, just like mine is displaying. I presume that the distro included with MFSTools is not prone to that problem, but still, that does not solve my problem. I guess when I go in and retry, I'll avoid the bios, but everything I do indicates a 32 gig drive to me. I'm really pissed that I bought this specific hard drive, and bought 3 of them at that! $82... saved $10 on each but it cost me 6 hours last night! ARG!

cameronj

Cmon folks, someone must have seen this error before. I tried just putting the drives back in and running the setup again, but i get all sorts of errors telling me that the total sectors mismatch on the big drive.... Next step i'm going to boot into windows with the new drive to see if it can see 80 gigs there.

stormsweeper

How old is your computer?

If need be, I'll make a dd dump of the Maxblast floppy and email it to you.

The linux distro on the CD is a 2.4.x system, and doesn't have the limitation you're talking about.

cameronj
07-10-2002, 01:41 PM
cameronj

My system has a mobo from maybe 18 months ago, it is a BX system. Nice system tho, it really cannot be a bios limitation because in Win2K i have a 40 gig drive and an 80 gig drive that work just fine. In fact the system recognizes the 40 gig drive that I have kept connected the whole time, it just makes no sense that it can't see the new 80 gig drive.

I hvae a maxblast 2 CD. what would you suggest that I do with it? I played with it for an hour last night and didnt see anything that helped my cause.

Thanks

cameronj

OK well I popped in teh maxblast cd (rather than the floppy i had used last night) and ran powermax. I am doing a low level format right now. Why the hell not. Any ideas, anyone?

cameronj

OK, low level format did no good (probably unsurprisingly). I booted into Win2K with my 40 Gig PC HD, my 80 gig PC HD, and my 80 gig new tivo drive connected. Using Partition magic I formatted the Tivo drive to Linux Ext2 or something, figuring that was the most likely correct choice of all I was given. Now I'm going to try to do it all again, unless someone else has a good idea.... please? someone?

cameronj

Still nothing... i feel like i'm talking to myself here... believe me, I am doing plenty of that here in my apartment, I hardly need to do more of it online...

Anyway here is the exact error message I get:

/# mfsadd -x /dev/hdc /dev/hdb
/dev/hdc2 success
/dev/hdc3 illegal seek
mfs_volume_load_header: total sectors (57,489,485) mismatch with volume header (123,543,572).
mfs_volume_load_header: Loading anyway
mfs_load_zone_map: Primary zone map corrupt, loading backup
mfs_load_zone_map: Secondary zone map corrupt, giving up
mfs_load_zone_map:Zone map checksum error
Unable to open MFS drives.

Someone PLEASE help (that's me, not the computer talking).

Where is everyone! All the other threads I see, tons of people jump in and the problem gets fixed immediately. Is this really such a unique problem? Why am I cursed????

stormsweeper
Get your hands on a different PC, preferably a newer one than the one you are using. The problem is almost certainly related to that PC.

cameronj

Hm...


Well, I have enough parts here to build myself an Athlon PC, I think, but it seems very weird that the two PCs I've tried are not good enough. I haven't read about anyone having this problem even with one PC - this is not an old PC, it is a P3-800...

Oh well, i'll give it a shot...

Finny

Yeah, I don't see how it could be the PC. I was able to get mine to work on an old Dell PII 450.

Have you tried MFSTools ver 2?

Also, be sure that you are not running any disk overlay programs that allow you to use the full capacity of the the drive.

Compleley disable you BIOS from looking at the drives. Then run MSFTOOLS ver 2 from a CD Boot.

That's what worked for me.

-Finny

cameronj

I've been using MFSTools 2 the whole time. I don't exactly know how to disable the bios from looking at the drives, but I certainly have not done any auto-detect functions (sometimes). There's just nothing, really, that it could be except some weird thing with the hard drive. I *have* other drives that are bigger than 32 gigs, reporting as their true size! That in itself should prove that the problem is not with the computer or the bios. Even in Windows I can get the full size of the new Tivo drive, just not using MFSTools.

cameronj
07-10-2002, 09:24 PM
WHEW!

cameronj
07-10-2002, 09:29 PM
That last post was not a mistake. I just had to express how relieved I was to finally get it.

Thanks for the posts guys, but those were not the problem. I had definitely tried all the jumper combos, probably a half dozen times each. And I had tried other drives and they gave me the same errors. This is on TWO computers.

The total saga, I'd estimate, ran me about 12 working hours over 3 days.

The solution?

Build another computer.

Then everything worked like a charm. Ain't that a bitch? Everyone says it works perfectly. The very small percentage of people who have problems, have problems that everyone has seen before, many times usually. Mine occurred on an old PC (P200) with random internals (SIS chipset, SIMMS etc etc) and a new PC that was totally industry standard (P3-800, BX chipset, Soyo board with latest bios, 256 in DIMMS etc etc etc). BOTH of those systems had the same exact issue. Could only see 32 gigs on that hard drive, while they could both see more than that on other drives. What a frustrating little bitch!

So what did I do? Well, had to solve it eventually, so it should be illustrative of just how many other things I had tried that I scraped together the parts from my room tonight and built a THIRD PC, this one a 650 Duron with a system built on the KT133A, hardly the image of a good IDE system (which I figured was where any problem had to be, somewhere between the IDE chip and the hard drive, inclusive).

And yep, since I had already done everything but mfsadd, that's all I did on the new PC. So 10 seconds later it was done and instead of the last time where it said "SUCCESS! Old size = 30 GB new size =62 gigs for an addition of 32 gigs!" This time it said "SUCCESS! Old size = 30 GB new size = 129 gigs for an addition of 99 gigs!" (dont ask me how I got 99 gigs out of an 80 gig hard drive, I ain't askin questions!).

So, I'll say it again... WHEW.

It feels good, because i EARNED IT ;)

I'm posting this in hopes that someone else who someday (a decade from now probably) has the same problem can be helped. Adios!