Re: FS corruption... some help maybe??

Herbert Wengatz 42850 (herbert.wengatz.extern@ae3.hypo.de)
Wed, 21 Jul 1999 13:32:46 MET DST


+> On Wed, Jul 21, 1999 at 12:17:43PM +0200, Herbert Wengatz 42850 wrote:
+> > +> > my system is only up max of 16 hours a day and then a manual
+> > +> > shutdown (till I get a UPS then I'll set up an automagic reboot
+> > +> > on sundays at midnight or something)
+> > +>
+> > +> You clearly have been using Microsoft Windows (NT/9x) to much. that
+> > +> Microsoft OS does need regular re-boots, but there is no reason to re-boot
+> > +> a Linux system just to re-boot it. I can understand turning it off when
+> > +> you aren't there without a UPS, but when you get a UPS, just leave it up
+> > +> 24/7. There is "almost" nothing that a re-boot can fix that cant be fixed
+> > +> by using other tools like ifconfig, etc... It is not uncommon to see
+> > +> uptimes well in excess of 70 days. I have heard of Linux 1.2.x systems
+> > +> staying up for longer then a year ( and by this time, this might include
+> > +> 2.0.x systems also)
+> >
+> > I can only support that! - I have an old 486 with 133 MHz at home, working as
+> > a gateway for me at home. I had it running for the first time with a Linux
+> > kernel version 2.0.30 for about 96 days, when it crashed. - The reason was -
+> > oh, no! - Not Linux! It was a broken CPU-fan! - I used this occasion, put in
+> > a more silent fan for the power supply, as well as another harddisk and
+> > upgraded to 2.0.36. - Since then, I never rebooted the system again. It
+> > currently has an uptime of about 165 days and I expect it to run far longer.
+> > My only fear is a powerfailure or that that bloody CPU-fan will break again.
+> >
+> > Can you tell me, what use rebooting is?
+>
+> Maybe because of nightly running applications, not releasing
+> shared memory and other IPC stuf, producing amazingly zombies,
+> preventing useful tasks to run (such as backup & Co).
+> Yes, scripts could be written to attempt to cleanly put
+> things up, but usually the best way is to reboot regularly.

You are *defintely* WRONG ! - Believe me.

Best is, to debug those programs which are not releasing the "shared memory
and other IPC stuff, producing *amazingly* zombies". - Or to switch to applications
which don't behave this way.

Any program not freeing it's shared memory when exiting (not important for which
reason, whatsoever) is IMHO *broken*.- Same goes for zombie-producers.

If you are suffering from such stuff - go and fix it!!!

This will dave a lot of your time and nerves.

Rebooting is *definitely* NO SOLUTION at all. - You may call it a bad habit from
Micro$oft-infested sysadmins, or in the best case, it's a minor and dumb workaround,
but is IN NO WAY a SOLUTION, since it SOLVES _nothing_.

+>
+> And, with Linux + autofs, taking into account dynamic modifications
+> of the NIS automount tables is possible without rebooting but
+> is quite painful with 100 or so machines. Sun's automount is
+> much better in reconfiguration...

OK. So there you have it. - Linux' automounter has to become more stable and
as you say not as good configurable as Suns automount. - This is a point where
Linux can improve. - Go and tell or better help the people who work on that!

I'm working here with 800+ Userworkstations (Sun's), not counting the servers...

+> However, whenever we can keep the uptime rising, we do it :-)

Great to hear this! :-))))) Keep up this intention!

A really smooth runnning network really satisfy you, when you can reach this.
Try to make everything "automagically" working. - When you have cronjobs which
send mails, set up a filter which starts the matching counter measures.
Write lots of cronjobs and atjobs. - This will keep a lot of problems away from
you. And this will save you a lot of time which you can spend for more interesting
and important stuff.

+> >
+> > At work (where I use Sun workstations) I have three machines (USER-Workstations!
+> > Where there are sitting users at the thing, logging in and out and crashing
+> > their applications all day long ;) ) - and these three machines have an
+> > uptime of far more than 280 days!!!!! (one is over 290 !)
+> >
+> > Can anybody please show me *any* NT-User-WS with such uptime?

Singin' with Elton John: "I'm still standing..." ;-)))))

Best regards,

Herbert

The first time Microsoft makes something that doesn't suck, is when
they start making vacuum cleaners.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/