Re: NTFS @#$%^&*(

Discussion in 'NZ Computing' started by EMB, Aug 21, 2008.

  1. EMB

    EMB Guest

    Carnations wrote:
    >
    > Perhaps more correctly, I have never experienced this sort of problem on
    > any *nix computer.


    Those of us who have been involved with *nix for a while[1] have however
    experienced all sorts of 'interesting' problems at various times (today
    included).

    [1] Well over 2 decades in my case.
     
    EMB, Aug 21, 2008
    #1
    1. Advertising

  2. EMB

    EMB Guest

    Carnations wrote:
    >
    > Yes - but that would be through using machines under heavy load running
    > 24/7/365 and then seeing what those boxen throw up wouldn't it?


    Most of the machines I actually have to care about run 24/7/365.
    Whether they run 'doze or *nix doesn't make much difference to their
    reliability - it's only when someone fucks about with them and changes
    things that they fall over. I bounced a 'doze box today that had an
    uptime of about 6 months - the bounce didn't resolve the problem which
    turned out to have been caused by a brain fart on the part of one of my
    colleagues. Sorting it out didn't require another reboot - 'doze boxes
    that are looked after properly are not much different to equivalent *nix
    machines, it just takes a bit of care in selecting what to install (or
    not install) on them.
    >
    > While I can't say I've been using *nix as long as you, I have been using
    > *nix in various incarnations and roles for about 7 years now, and have
    > been Microsoft free on all my BAU computers for about 5 years now.


    I only use 'doze on the desktop due to inertia - it's too much like hard
    work to have to make a paradigm shift when I go home. I'd rather just
    use what I generally have to use at work.
     
    EMB, Aug 21, 2008
    #2
    1. Advertising

  3. EMB

    Enkidu Guest

    EMB wrote:
    > Carnations wrote:
    >>
    >> Perhaps more correctly, I have never experienced this sort of problem
    >> on any *nix computer.

    >
    > Those of us who have been involved with *nix for a while[1] have however
    > experienced all sorts of 'interesting' problems at various times (today
    > included).
    >
    > [1] Well over 2 decades in my case.
    >

    ....and will have come across interesting solutions. eg the bash
    construct <(bash commands). This allows things like:

    diff <(ls <dir-one>) <(ls <dir-two)

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 22, 2008
    #3
  4. EMB

    Enkidu Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 00:28:01 +1200, EMB wrote:
    >
    >> I only use 'doze on the desktop due to inertia - it's too much like
    >> hard work to have to make a paradigm shift when I go home. I'd
    >> rather just use what I generally have to use at work.

    >
    > I'm the opposite to that - I use all *nix at home, and have to put up
    > with a Dozy box at work.
    >
    > I reboot the doze box every night, because if I don't it goes
    > doolally.
    >

    What did you do to it? It's my experience that a Windows box is stable
    if only people don't *fiddle* with it. I can't remember the last time I
    rebooted my laptop and then it would have been because I was on call.
    >
    > But my *nix boxen at home... well I might restart my
    > desktop box once every month or so if I felt like it. My file server
    > had an uptime of over 400 days prior to a recent thunderstorm, as did
    > my firewall box.
    >
    > For me, I have to put up with the Doze box doing stupid things at
    > work, such as using the inane Micro$oft "Credential Manager" - which
    > I got disabled because it was just causing too much grief.
    >

    That's an HP thing, isn't it?
    >
    > I am also aware of Doze application servers that have regular
    > scheduled reboots because it has been found that without doing that
    > those boxes end up crashing.
    >

    I'd say that was because of whatever application the app server was
    serving. I've had Windows boxes up for more than two years without a reboot.

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 22, 2008
    #4
  5. EMB

    Enkidu Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 21:38:30 +1200, Enkidu wrote:
    >
    >> ...and will have come across interesting solutions. eg the bash
    >> construct <(bash commands). This allows things like:
    >>
    >> diff <(ls <dir-one>) <(ls <dir-two)
    >>
    >> Cheers,

    >
    > Interesting! being able to do a diff between the results of two different
    > Bash command strings.
    >

    Yep, that's what I said - "interesting".

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 22, 2008
    #5
  6. EMB

    Enkidu Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    >
    >> What did you do to it? It's my experience that a Windows box is stable
    >> if only people don't *fiddle* with it.

    >
    > I've never fiddled with it. In my experience Windows boxes are
    > fundamentally unstable. It does, however, also have some Computer
    > Associates "Dashboard" garbage installed on it.
    >

    I don't know why people think that a system is 'fundamentally unstable'
    when any amount of non-OS stuff is installed on it. By the same token my
    car is unstable if I put a tonne of stuff on the roof rack!

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 22, 2008
    #6
  7. EMB

    Enkidu Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    >
    >> I'd say that was because of whatever application the app server was
    >> serving. I've had Windows boxes up for more than two years without a
    >> reboot.

    >
    > What? No monthly "critical" patching with the requisite reboot?
    >

    NT needed reboots for patches. 2000 less so, and XP even less.

    NT SP6a would run forever without a reboot.

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 22, 2008
    #7
  8. EMB

    Jasen Betts Guest

    On 2008-08-22, Enkidu <> wrote:
    > EMB wrote:
    >> Carnations wrote:
    >>>
    >>> Perhaps more correctly, I have never experienced this sort of problem
    >>> on any *nix computer.

    >>
    >> Those of us who have been involved with *nix for a while[1] have however
    >> experienced all sorts of 'interesting' problems at various times (today
    >> included).
    >>
    >> [1] Well over 2 decades in my case.
    > >

    > ...and will have come across interesting solutions. eg the bash
    > construct <(bash commands). This allows things like:
    >
    > diff <(ls <dir-one>) <(ls <dir-two)


    ooh! pipes as filenames,

    $ echo <( echo )
    /dev/fd/63

    ah... so that's how it's done.

    /dev/fd/ is a kernel 2 thing isn't it?

    Bye.
    Jasen
     
    Jasen Betts, Aug 22, 2008
    #8
  9. EMB

    EMB Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    >
    >> I'd say that was because of whatever application the app server was
    >> serving. I've had Windows boxes up for more than two years without a
    >> reboot.

    >
    > What? No monthly "critical" patching with the requisite reboot?


    Why patch something that isn't at risk? Servers that are on the
    internal network (as app and DB servers generally are) generally don't
    need patching just for the sake of being "up to date" as they are at
    zero risk of being exploited. External facing servers are a slightly
    different proposition, but good firewalling minimises that risk too.
     
    EMB, Aug 23, 2008
    #9
  10. In article <>,
    says...
    >
    > > What did you do to it? It's my experience that a Windows box is stable
    > > if only people don't *fiddle* with it.

    >
    > I've never fiddled with it. In my experience Windows boxes are
    > fundamentally unstable. It does, however, also have some Computer
    > Associates "Dashboard" garbage installed on it.
    >


    Nah. My XP Pro will run for weeks and months without a reset.

    I just make sure I get no stupid invasive crap autostarting, like
    realplayer etc etc, stuff that keeps downloading more stuff and
    spreading through the computer like a fungus. All 'autoupdating'
    etc etc gets disabled. Keep it slim and trim and there are no
    problems.

    Works for me.

    -Peter
     
    Peter Huebner, Aug 23, 2008
    #10
  11. EMB

    Enkidu Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 23:01:25 +1200, Enkidu wrote:
    >
    >> I don't know why people think that a system is 'fundamentally unstable'
    >> when any amount of non-OS stuff is installed on it. By the same token my
    >> car is unstable if I put a tonne of stuff on the roof rack!

    >
    > The only junk I use on it on a regular basis is MS Orifice.
    >

    I've not found the XP/Office combination the least bit unstable.

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 23, 2008
    #11
  12. EMB

    Enkidu Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 23:04:11 +1200, Enkidu wrote:
    >
    >> NT SP6a would run forever without a reboot.

    >
    > And how would you patch the kernel?
    >

    If you're going to start talking about Linux, here's a quick guide:

    http://www.linuxdocs.org/HOWTOs/Kernel-HOWTO-6.html

    But you cannot (yet) patch a kernel without a reboot.

    If you are talking about NT SP6a the machine I referred was never
    patched after SP6a, the last one for the NT4 OS.

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 23, 2008
    #12
  13. EMB

    Enkidu Guest

    EMB wrote:
    > Carnations wrote:
    >> On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    >>
    >>> I'd say that was because of whatever application the app server was
    >>> serving. I've had Windows boxes up for more than two years without a
    >>> reboot.

    >>
    >> What? No monthly "critical" patching with the requisite reboot?

    >
    > Why patch something that isn't at risk? Servers that are on the
    > internal network (as app and DB servers generally are) generally don't
    > need patching just for the sake of being "up to date" as they are at
    > zero risk of being exploited. External facing servers are a slightly
    > different proposition, but good firewalling minimises that risk too.
    >

    Not quite zero risk. In fact, quite a big risk, from disaffected
    employees and so on.

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 23, 2008
    #13
  14. EMB

    Enkidu Guest

    Carnations wrote:
    > On Sat, 23 Aug 2008 11:12:57 +1200, EMB wrote:
    >
    >> Carnations wrote:
    >>> On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    >>>
    >>>> I'd say that was because of whatever application the app server was
    >>>> serving. I've had Windows boxes up for more than two years without a
    >>>> reboot.
    >>> What? No monthly "critical" patching with the requisite reboot?

    >> Why patch something that isn't at risk? Servers that are on the
    >> internal network (as app and DB servers generally are) generally don't
    >> need patching just for the sake of being "up to date" as they are at
    >> zero risk of being exploited. External facing servers are a slightly
    >> different proposition, but good firewalling minimises that risk too.

    >
    > Would you suggest the same for desktop computers that are on the internal
    > network?
    >

    Assuming that they are allowed to browse the Internet, they are at risk.
    Also laptops which may leave the 'home' network could bring in viruses
    and so on. You have update them regularly, so why not all machines? It
    can easily be done automatically.

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 23, 2008
    #14
  15. EMB

    Enkidu Guest

    Carnations wrote:
    > On Fri, 22 Aug 2008 22:58:49 +1200, Enkidu wrote:
    >
    >>>> ...and will have come across interesting solutions. eg the bash
    >>>> construct <(bash commands). This allows things like:
    >>>>
    >>>> diff <(ls <dir-one>) <(ls <dir-two)
    >>>>
    >>>> Cheers,
    >>> Interesting! being able to do a diff between the results of two
    >>> different Bash command strings.
    >>>

    >> Yep, that's what I said - "interesting".

    >
    > How useful have you found it to be?
    >

    The diff directories thing is very useful, especially for sorting out
    copies and the like that fail for one reason or another.

    Cheers,

    Cliff

    --

    "I LOVE IT!!" - my biggest fan and follower, on a newsgroup, somewhere.
     
    Enkidu, Aug 23, 2008
    #15
  16. EMB

    EMB Guest

    Carnations wrote:
    >
    > Would you suggest the same for desktop computers that are on the internal
    > network?


    If firewalled and locked down suitably[1] it's less of an issue to leave
    them unpatched than it is to roll out an improperly tested patch and
    introduce stability issues across the entire network.

    [1] Floppyless, CDless, USB storage disabled desktop machines on a
    network with real-time virus scanning of both mail and web traffic are
    fairly hard to compromise. Add a certain amount of content filtering at
    the firewall and IMO not much is going to be lost by not patching the
    machines.
     
    EMB, Aug 23, 2008
    #16
  17. EMB

    EMB Guest

    Enkidu wrote:
    > EMB wrote:
    >> Carnations wrote:
    >>> On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    >>>
    >>>> I'd say that was because of whatever application the app server was
    >>>> serving. I've had Windows boxes up for more than two years without a
    >>>> reboot.
    >>>
    >>> What? No monthly "critical" patching with the requisite reboot?

    >>
    >> Why patch something that isn't at risk? Servers that are on the
    >> internal network (as app and DB servers generally are) generally don't
    >> need patching just for the sake of being "up to date" as they are at
    >> zero risk of being exploited. External facing servers are a slightly
    >> different proposition, but good firewalling minimises that risk too.
    > >

    > Not quite zero risk. In fact, quite a big risk, from disaffected
    > employees and so on.


    A risk which is generally not mitigated by any amount of patching.
     
    EMB, Aug 23, 2008
    #17
  18. EMB

    EMB Guest

    Enkidu wrote:
    > Carnations wrote:
    >> On Sat, 23 Aug 2008 11:12:57 +1200, EMB wrote:
    >>
    >>> Carnations wrote:
    >>>> On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    >>>>
    >>>>> I'd say that was because of whatever application the app server was
    >>>>> serving. I've had Windows boxes up for more than two years without a
    >>>>> reboot.
    >>>> What? No monthly "critical" patching with the requisite reboot?
    >>> Why patch something that isn't at risk? Servers that are on the
    >>> internal network (as app and DB servers generally are) generally don't
    >>> need patching just for the sake of being "up to date" as they are at
    >>> zero risk of being exploited. External facing servers are a slightly
    >>> different proposition, but good firewalling minimises that risk too.

    >>
    >> Would you suggest the same for desktop computers that are on the
    >> internal network?
    >>

    > Assuming that they are allowed to browse the Internet, they are at risk.
    > Also laptops which may leave the 'home' network could bring in viruses
    > and so on. You have update them regularly, so why not all machines? It
    > can easily be done automatically.


    To a certain extent. If I release XP SP3 via our internal WSUS server
    our WAN will grind to a halt for the next month or so - it's not quite
    as straightforward as it seems.
     
    EMB, Aug 23, 2008
    #18
  19. In article <>,
    says...
    > On Sat, 23 Aug 2008 12:28:03 +1200, Peter Huebner wrote:
    >
    > > I just make sure I get no stupid invasive crap autostarting,

    >
    > So you don't have MSIE installed?
    >


    Er ...???

    The fact apart that I do not as a rule use MSIE on the internet (in
    fact, most of my OS & what little MS software I use has no permissions
    through the firewall), I have yet to see MSIE, by itself, without luser
    input, go out on the 'net, fetch bazillons of gadgets, toolbars, media
    players, ad content, yet more invasive 'software' and other unwanted
    crap and install it without user permission, all things being equal.

    So I think you missed te barn with that shotgun, sorry.

    -P.


    ---

    Irony is not like coppery, brassy or tinny.
     
    Peter Huebner, Aug 23, 2008
    #19
  20. EMB

    Dave Doe Guest

    In article <48af7e24$>, says...
    > Enkidu wrote:
    > > Carnations wrote:
    > >> On Sat, 23 Aug 2008 11:12:57 +1200, EMB wrote:
    > >>
    > >>> Carnations wrote:
    > >>>> On Fri, 22 Aug 2008 22:01:30 +1200, Enkidu wrote:
    > >>>>
    > >>>>> I'd say that was because of whatever application the app server was
    > >>>>> serving. I've had Windows boxes up for more than two years without a
    > >>>>> reboot.
    > >>>> What? No monthly "critical" patching with the requisite reboot?
    > >>> Why patch something that isn't at risk? Servers that are on the
    > >>> internal network (as app and DB servers generally are) generally don't
    > >>> need patching just for the sake of being "up to date" as they are at
    > >>> zero risk of being exploited. External facing servers are a slightly
    > >>> different proposition, but good firewalling minimises that risk too.
    > >>
    > >> Would you suggest the same for desktop computers that are on the
    > >> internal network?
    > >>

    > > Assuming that they are allowed to browse the Internet, they are at risk.
    > > Also laptops which may leave the 'home' network could bring in viruses
    > > and so on. You have update them regularly, so why not all machines? It
    > > can easily be done automatically.

    >
    > To a certain extent. If I release XP SP3 via our internal WSUS server
    > our WAN will grind to a halt for the next month or so - it's not quite
    > as straightforward as it seems.


    Why? Do you not get WSUS to store the updates (the default).

    --
    Duncan
     
    Dave Doe, Aug 23, 2008
    #20
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Bay

    NTFS permission question

    Bay, Oct 5, 2003, in forum: MCSE
    Replies:
    2
    Views:
    490
    Richard Hensley
    Oct 8, 2003
  2. Joshua

    Question about NTFS

    Joshua, Oct 8, 2003, in forum: MCSE
    Replies:
    12
    Views:
    670
    Marlin Munrow
    Oct 15, 2003
  3. Bay
    Replies:
    4
    Views:
    9,129
    Kjell
    Oct 23, 2003
  4. Rick

    NTFS permisions

    Rick, Oct 22, 2003, in forum: MCSE
    Replies:
    8
    Views:
    2,209
  5. Tech
    Replies:
    3
    Views:
    719
    Plato
    Apr 6, 2004
Loading...

Share This Page