OT: Linux Recommendations For Old Pentium PC

Home » CentOS » OT: Linux Recommendations For Old Pentium PC
CentOS 24 Comments

I’ve got a very small footprint rack server with a 4TB drive in that I wish to be a Bacula storeage device. However, it’s got an old board / processor in it.

Can anyone recommend a Dist that would work on it?

24 thoughts on - OT: Linux Recommendations For Old Pentium PC

  • I would use FreeBSD (and I do use FreeBSD for bacula, now bareos backup server and storage hosts), it has really small “footprint”, and it is quite widespread.

    Incidentally, I was using bacula for very long time, but recently I
    switched to bacula’s fork: bareos. You may want to consider the differences before you finalized everything in stone.

    Valeri

  • Hi Valeri,

    Thanks for this. I haven’t looked at FreeBSD since the 1990’s or there abouts, but I’ll give it a look.

    I’m also looking at lubuntu, but was hoping that there was a lCentOS. We tend to like what we’re used to.

    I’d be interested in your views on the differences between bacula and Bareos. I do have one Bareos storeage device but that’s just in Bacula compat mode.

    Gary

  • I’m fresh out of FreeBSD world. Depending on the port, it can be easy and predictable, or an absolute confusion-fest.

    Cheers, Bee

  • Gary Stainburn wrote:

    CentOS will work, but you might start with minimal (but make sure it includes networking).

    Please note that I installed CentOS 6, just a few months ago, on an HP
    Netbook from ’09, and it runs perfectly well.

    mark “see? I didn’t say anything about systemd….”

  • CentOS 6 requires a PAE supporting CPU. Subject referenced Pentium CPU.

    Pentiums do not support PAE, and so would not run CentOS 6 without fun and games and an alternative kernel.

    I previously had a Dell X1 with a Pentium M CPU, which also didn’t advertise PAE support, so couldn’t run the stock CentOS 6 kernel, which made installation a little more interesting.

    If you’re really stubborn, there are options for mashing it on anyway, but I’m not sure I’d bother. In my case I think I just ran anaconda within C5 to install C6 onto another LV, put a non-PAE kernel on, then booted into the C6
    install.

    jh

  • It is counter productive, and this list is wrong place to tell some alternative system is better than one or another Linux, hence this is the rant, ignore it, everyone who can:


    Linux kernel is IMHO overburdened by quite a lot of stuff that doesn’t belong there. Hence higher chance of bugs (and almost all bugs in kernel have security implications). Adding to that not too rare glibc security patches, all in all in my observation on average you have to reboot Linux box once every 45 days. That became a statistics after switch from
    2.4 to 2.6 kernel as I recollect, and one of my friends started to use word “Lindoze” when he was looking where to migrate his servers to those days…

    All in all for your hardware if I were to pick the system that is widely used and has small footprint and small demands to hardware specs, I
    would use FreeBSD.

    I hope, this helps.

    Valeri

  • I did try CentOS 6 32-bit because I believe that was what was on it last time. Unfortunately this time it refused to see the install image on the DVD

    I did also wonder about repositories and how long they’ll be available for it.

    Gary

  • FreeBSD ports should not be confused with FreeBSD system. Each of ports is maintained by different maintainer(s), some of them get obsolete, sometimes quickly, and not every software that is ported deserves in sane sysadmin’s opinion to be offered to the users.

    And the same can be said about RPM collections (which are many, and one huge one would be Fedora’s one) or deb packages collection of Debian
    (and its clones).

    All in all, if one gets confused sometimes, one can get confused using any open source system.

    On the other hand, before starting to offer some software to users, every sysadmin analyzes it carefully and tries to predict if it will stay alive for long time. As it is huge pain to migrate users to some alternative once the software of your choice becomes dead… And that is how sysadmins earn their salaries IMHO.

    Just my $0.02.

    Valeri

    —
    ++++++++++++++++++++++++++++++++++++++++
    Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247
    ++++++++++++++++++++++++++++++++++++++++

  • I would not use system that has EOL (End Of Life) in a really close future. That would be waste of my time. Just mentioning.

    Valeri

    —
    ++++++++++++++++++++++++++++++++++++++++
    Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247
    ++++++++++++++++++++++++++++++++++++++++

  • I’ve been using it for years. I know the difference. You run FreeBSD and you install ports. The two come hand-in-hand.

    There’s no confusion. The maintainers, the admins, are far and few between on FreeBSD. The very reason I’m here is due to to just that. That, cannot be said of the Linux world. Your last paragraph is on point, and some people earn their “keep” regardless of how many errors they make. Historically, that’s the same for IBM and Microsoft, and everybody that employed those technologies because “IBM is too big to fail”. Well documented in business cases for decades now, something that a lot of tech people simply don’t understand.

    Cheers, Bee

  • We use mysql as database backend for bacula, and it becomes heavy loaded, over time especially wenn restoring respectively generating filelists. So, not sure if such old CPU provides enough compute power …

  • You’re giving two very mixed signals here.

    “Old Pentium,” as someone else said, can mean anything back to 1993, but “4 TB drive” suggests something far newer than that.

    I ask because that affects the expected energy draw of the server. If it’s old, it could be 200 W or so. If you’re using “old” rather loosely, then it could be down in the double digits.

    Here’s why it matters:

    https://www.rapidtables.com/calc/electric/energy-cost-calculator.html

    At 12 pence per kWh — typical for power in some places in your country, based on your TLD — it’s going to cost you about 1 pound per watt consumed if it runs all day every day. If it draws 35 W, that’s £35/yr. If it draws 200 W, that’s £200/yr.

    If the cost is high enough, then it’s probably cheaper to buy a new energy-efficient server, which then lets you buy something that will run any Linux distro you want.

  • This is important for the machine hosting director. If database is hosted on different machine even that shouldn’t be awfully loaded in my opinion. As far as the box hosting storage daemon is concerned, that dosn’t need much of resources (like CPU or RAM – unless one uses NFS
    which I wouldn’t), the only things to pay attention for that box would be network connection capacity and/or filesystem speed, whichever becomes a bottleneck.

    I hope, this helps.

    Valeri

  • Warren Young wrote:

    Good point. If it recognizes a 4TB drive, then it has to have a controller card from around ’10 or newer. I don’t know that an “old Pentium” can address that.

    Don’t they also call 686’s Pentiums?

    mark

  • My recommendation, take it for what its worth:

    32-bit distros to me are a short lived proposition IMO.

    Example: I’m running CentOS 6, 32-bit version. I recently ran into an issue where a package
    (clamav) started using a 64-bit library for decompression of files. End result, end of scanning for email viruses as this lib won’t run on 32-bit AND, the lib hasn’t been updated for 32-bits in
    6 years.

    Forced to move to CentOS 7 to get 64-bit libs. CentOS 6 is still a viable supported OS until end of 2019 or 2020.

    So, ditch the box and get something that runs 64-bit..Your time will be better spent!!!

    Jay

  • This isn’t the director, it’s just a remote storeage device. All it has to do is talk network and HDD. Why is why I’ve retained it as it’s a 1U half depth rack server and fits into my already overcrowded cabinet.

  • Hi Warren,

    I had considered power consumption but only with regard that it is a small footprint system, both physically and in terms of processing power etc. I had not considered the lack of energy efficiency. The server has a MSI
    MS-9628 board with a Pentium M processor, and the one modern 4TB HDD.

    There is one tiny fan in the PSU and another tiny processor fan on the CPU.
    From this I (possibly wrongly) assumed power consumption would be low.

    It used to have 2 x 1.5TB drivers with software RAID until that died. I am still in the process of installing lubuntu so I don’t know how effective it will be. I had considered putting CentOS6 32-bit back on, but has been said elsewhere that’s very near EOL. Having said that, some of my (soon to be replaced) SAMBA boxes are still runnning F9.

  • You could well be right, but I’m a fan of taking measurements over guessing. :)

    If you were in the US, I’d recommend either of these from personal testing:

    https://amzn.to/2NMWXJq
    https://amzn.to/2oyz5Oz

    I wrote the review voted most helpful for the latter item. You might want to read it.

    A clamp meter + line splitter is more fiddly to use than a Kill-a-Watt, but you can use the clamp meter for many more things, so it’s a better overall value unless you simply will not be doing those other things.

    Neither of those will work for you due to US vs UK AC line connector differences, but these two items appear to be roughly equivalent:

    https://www.amazon.co.uk/dp/B01DSQ30FO/
    https://www.amazon.co.uk/dp/B00H99EECU/
    plus:
    https://www.amazon.co.uk/dp/B075D512KM/

    I’ve found the same clamp meter that I reviewed above, differing only in OEM labeling.

    I’m not too happy with my guess for the best line splitter, for two reasons:

    First, I’m unfamiliar with your UK power plugs, so I don’t know if that’s actually a UK plug.

    Second, two of the reviews of that line splitter point out a design flaw that you might care about. I’d have passed it by if I could find something better on the Amazon.co.uk site, but that is the best I found, alas.

    You need a line splitter to do this test with a clamp meter, else you get a zero reading since the electromagnetic field from the neutral line cancels that of the hot line: you have to measure one or the other, separately. It doesn’t matter which one you measure: the current through both lines is the same, differing only in direction, which is what we mean when we call something an electrical “circuit”.

    Alternate plan: build a “broken circuit:”

    https://tangentsoft.com/elec/broken-circuit.html

    You can either make one for an inline current meter, as shown, or take the basic idea to DIY your own line splitter.

  • Gary Stainburn wrote:

    This means that you’ll need a 32-bit OS, and that PAE might not be recognized by the OS. Debian 9 32-bit would work; CentOS 6 32-bit might not.

  • It previously had CentOS 6 on it, but I wanted to avoid that as it now has a limited shelf life. However, when I tried lubuntu the Bacula versions were not compatible. Now Putting CentOS 6 32-bit back on and crossing my fingers.

    I’m really impessed that the m/board recognised the 4TB drive