Re: Problem - Solved - Re: Starting with Linux

From: BobS (spam_at_spam.com)
Date: 11/18/05 

Date: Fri, 18 Nov 2005 16:55:50 GMT

Valentin,

No doubt that you thought it was hardware related and your overall
suggestion was spot on. I just like to make sure that all possibilities are
looked at before telling someone their hardware is dead meat when it may not
be. According to some on-line sources, memory for that system should not be
that expensive and will certainly be a lot less expensive to upgrade than
buying a new computer.

As I'm writing this, I dusted off an old P3 Celeron 300MHz system that was
stashed away and only has 128Mb memory. I threw in a 16Gb drive, fdisk'd it
and running a memory test on it now. Looks good so far, so I'm going to be
loading up v10 (retail package) in a couple of moments and I should be
seeing the same symptoms Rod is seeing.

If I do, then I'll boost the mem to 512 and reload. Stay tuned....

Bob S.

"Valentin Guillen" <usenet1@myrealbox.com> wrote in message
news:pan.2005.11.18.15.56.11.273566@myrealbox.com...
> On Fri, 18 Nov 2005 15:08:33 +0000, BobS wrote:
>
>> Rod,
>>
>> You just solved the mystery and you do have a hardware problem - not
>> enough memory for ver 10. Minimum requirements are 256Mb and the
>> Recommended amount is 512Mb. So here's what I believe is happening:
>
> I've been following this thread for days, and I was certain that the
> answer would manifest itself. It has!
>
> This is the reason that I originally sugested that Rod stick with version
> 7.3. I had a laptop which had 7.3 on it. It ran just fine with 96mb of
> RAM. I made sure that SuSE had a 100mb swap partion when I installed the
> Linux. After upgrading the laptop to SuSE 9.0, I had bootup times of 4~5
> minutes. I *expected* this. After bootup, the computer is eminently
> USABLE, but slow. I expected this. I did not and will revert to 7.3, but
> I understand WHY I have a five minute bootup time. I like the features of
> 9.0, so I will not revert to the earlier version. The earlier version had
> everything I needed at the time.
>
> My laptop is a Pentium 133mz, with the 96mb RAM. It is simply not worth
> investing in more specialized RAM for the dinosaur. With the RAM
> limitation, it is imperative that there be a sizable swap partition
> created during installation, because it will be swapping like crazy during
> usage. That's precisely what it should be doing.
>
> To reiterate a point in my earlier post, use the 7.3 to learn on, and when
> ready, upgrade both the hardware and software to a much newer version.
>
> Regards,
>
> vg
>
>
>> 1. SUSE does not see enough memory to support the operating system
> much
>> less DMA (direct memory access) for the hard drives - so it turns DMA
>> off. With DMA off, you will notice some system degradation and with only
>> 128Mb of RAM, it's swapping like crazy out to the hard drive which is
>> running slow because DMA is off because there's not enough memory to
>> support it. See the big -slow- circle there?
>>
>> 2. I'll bet that part of the memory is also being shared by the video (4
>> to 8Mb probably) further reducing the amount of system memory available.
>>
>> 3. I haven't checked but I'll bet ver 7.3 has lower minimum memory
>> requirements of only needing 32 Mb as WinME does and that's why it runs
>> faster.
>>
>> http://support.microsoft.com/default.aspx?scid=kb;en-us;272156#kb1
>>
>> 4. I think your only choices now are to either upgrade the memory, stay
>> with ver 7.3 (or a version that doesn't require as much memory) or a new
>> computer.
>>
>> 5. That computer probably uses PC100/133 type memory and it's getting
>> hard to find locally but you can try on-line. Just be sure to look up
>> the exact memory type you need. If it's the older PC100 memory, it
>> comes in several flavors (chip configuration) and can be crazy in price
>> just because its getting hard to find. Hopefully you can use the PC133
>> memory which is relatively cheap (www.pricewatch.com ) and click on
>> Memory System, and look for PC133 type. You'll need to know what the
>> max amount of memory that your computer can support in each slot. It's
>> cheap enough on line so if you can load it up to the max amount at a low
>> cost - then do it and don't worry about that 127Mb stick you have in
>> there now. If your system supports two, 512Mb sticks, then go for it
>> while its still available.
>>
>> 6. eMachines
>> http://www.emachines.com/support/product_support.html?cat=desktop&subcat=Other
>> . Use this to enter your model number and look up the specs if you can't
>> find the manuals.
>>
>> Be glad to offer any help I can, such as how to decide what memory is
>> needed (need model # of computer) and how to change out the memory
>> (piece o' cake..) but you gotta take the case off that computer first.
>>
>> Choice is yours,
>>
>> Bob S.
>>
>>
>> "rod s" <rod@surfanytime.co.uk> wrote in message
>> news:4dcbba47bbrod@surfanytime.co.uk...
>>> Hello again,
>>>
>>> I will try to answer the questions in as much detail as I can:
>>>
>>> In article, BobS wrote:
>>>> Rod,
>>>
>>>> Need to get unconfused here. You have Windows on one hard drive
>>>> (master) and Linux on a second hard drive - is that correct?
>>> Yes,
>>>
>>> (When I started experimenting with Linux (couple of years ago) I added
>>> this second drive (40GB). The Windows was resident on the original hard
>>> drive and as far as I understand it the loading process of Linux puts
>>> it all on the second (slave) drive. I have not had to tell it any of
>>> this in these recent trials, so I must have told the v7.3 what to do
>>> originally and Linux must detect this somehow when (re-)installing any
>>> version.
>>>
>>>> 1. What version of Windows are you running (Win98SE, WinXP, WinXP
>>>> w/SP2...)
>>> It is Windows ME (millennium edition) Me 4.90.3000
>>>
>>>> 2. Name brand and model of computer (if any).
>>> It is emchines - GeuineIntel - Intel(r)Celeron(tm)processor 127.0MB RAM
>>>
>>> Is that enough... I could perhaps find the books which came with the
>>> computer.
>>>
>>>> 3. Does Windows run correctly (fast) when you boot to it (yes / no)
>>> I have never had any reason to suspect it is running slow. Some progs
>>> take 30s to load eg Firefox but I assume this is due to bloated
>>> program.
>>>
>>>> 4. When you had ver 7.3 running, was the DMA set On or Off the hard
>>>> drives?
>>> I don't know.. I suppose if it comes to re-installing v7.3 for test
>>> purposes, I could check this and other points. Systematically.
>>>
>>>> 5. When ver 7.3 was running was the DMA for the CD drive On or Off?
>>> Same answer as for q4.
>>>
>>>> 6. When ver 10 is running, you said YaST shows DMA is "NOT" on for
>>>> "both" the hard drives
>>> After installation, I looked in YaST-hardware-IDE DMA MOde. There are
>>> three lines.
>>> current DMA......device........Required Mode off hda
>>> no change off hdb no change ultraDMA/33 CD
>>> no change
>>>
>>>
>>>> 7. What was the other "place" in Linux that indicated DMA was "ON"
>>>> when running ver 10?
>>> elsewhere on the same screen you can change the DMA. You are offered in
>>> a submenu:
>>> DMA Off
>>> DMA On (default) <-----it says!!
>>> DMA/16
>>> UltraDMA/16
>>> UltraDMA/33
>>>
>>> I have never altered the CD setting
>>> I have tried DMA On
>>> I have tried UltraDMA/16
>>> I don't think I tried any of the others.
>>>
>>>> 8. Was that for the hard drives or CD?
>>> see q7
>>>> Even though the hard drives support DMA, and it sounds like your
>>>> hardware is new enough (BIOS) that it should also support DMA, it
>>>> appears you are having a conflict. YaST should be turning DMA to ON
>>>> for the hard drives and OFF for the CD/s. Double check that.
>>> see qq6-7
>>>
>>>> If YaST is turning DMA to OFF for the hard drives then maybe ver 10 is
>>>> not identifying the motherboard correctly or the drives.
>>> OK
>>>
>>>> The one thing that is troublesome is that you said you turned DMA to
>>>> ON and that buggered the boot record - which is a typical symptom
>>>> under Windows when the hardware really doesn't support DMA even though
>>>> the software shows selections to turn it on.
>>> I suppose that's it... the boot process hung (if memory is right)
>>>> I've had to reload Windows on more than one occasion trying to see if
>>>> the hardware really supports DMA.
>>>
>>>> It would be nice if you could get at the hardware to unplug the
>>>> windows drive so we can do some elimination here but you indicated
>>>> that's not possible. Anyone else around that could do that for you?
>>> The problem is space - it would be a major undertaking to get at the
>>> computer - and (I hope this doesn't sound as if I am ungrateful for all
>>> your patient help) I'm not sure it is worth the effort.
>>>
>>>> At any rate, try answering the questions above and then when you
>>>> reload ver 10, do it by the numbers, one step at a time, assume
>>>> nothing, write down any errors it gets or anything you don't
>>>> understand so we can try to sort it out.
>>> Well, when I loaded it yesterday, it didn't seem to report any errors -
>>> the whole process went smoothly. But I agree it is only sensible to do
>>> this systematically so that one does not rely on memory.
>>>
>>>> If I get the chance in the am, I'll be loading ver 10 up on one of my
>>>> test boxes here (old hardware also) and I'll let you know how it runs
>>>> - or I'll be asking you for advice....;-)
>>> LOL
>>>> Don't give up on it yet. We just need to get it back to ground zero
>>>> and go from the known to the unknown - one step at a time. So keep a
>>>> pencil and paper handy to jot down anything that looks flaky. It just
>>>> may be the clue needed.
>>>
>>> The word 'just' is rather encouraging. I am not planning to do anything
>>> till I hear from you - with a plan of campaign and then anything I do
>>> will be taken down in writing and may be used in evidence.
>>>
>>> Though all this installing and re-installing is very time-consuming,
>>> there is one bright side. The computer I use for messages such as this
>>> is a different one, so that I am not reliant upon the 'test' machine
>>> for immediate communication. This machine is not a Windows one - it is
>>> a RISC-PC if that means anything to you. But if it helps in all of this
>>> I do have another older Windows machine which is running Windows 98 but
>>> only has the one hard disk.
>>>
>>> --
>>> /\ _ \
>>> |==============| / / / \__\ |=====================|
>>> | Rodger Davis | /_/ / / / \ |rod@surfanytime.co.uk|
>>> |==============| / \_/\_/ /___) |=====================|
>>>
>>>
>

Relevant Pages

  • Re: Problems with IE after clean Install of Win XP & SP2
    ... I think hardware check may be my next step. ... The system is a Dell Dimension 8600, 232 GB hard drive, 1GB ram. ... Memory performace seems fine - Commit figures (Total=315 MG and Peak ... I* have only loaded Avast antivirus and MS Office 2003. ...
    (microsoft.public.windowsxp.general)
  • Re: 32-bit programs on Windows x64
    ... KERNEL address space which can be in the paged pool (which is not the same as the pageable ... conceivably care about) and the amount which can be in the non-paged pool (the limit ... pageable memory is all that matters in terms of user performance, ... if there is 8.0 GB of RAM on the system? ...
    (microsoft.public.vc.mfc)
  • Re: Virtual Memory in use at 100%
    ... I just had a Hardware Malfunction again. ... I had to restart the server so I ... hardware problem but a software memory leak and they told me that the best ... > RAM compatibility, etc, can be the torture of anyone doing hardware/OS support. ...
    (microsoft.public.windows.server.sbs)
  • Re: running Linux with no swap space (but lots of RAM)
    ... Some amount of caching is necessary to achieve such I/O scheduling. ... The amount of memory allocated to ... Someone who thinks writing to /dev/null would result in lots of data being ... TB of RAM and fits in an ATX case using no more than 550 watts of power. ...
    (comp.os.linux.development.system)
  • Re: High memory
    ... memory and then copy it up above 1MB...but if you want to put ... outside the CPU, memory is seen in a completely different light...this ... 1MB with "real mode memory" labelled on it or anything...the actual memory ... the system bus to actual hardware devices...hence, ...
    (alt.lang.asm)