"...today, I login and get "bus error" for everything on /usr and /sbin..."
"...my main site with 25000+ users crashed and stayed crashed for six hours
until I rebooted."
"...this is definitely a kernel issue. At least, it's not a hardware
issue..."
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=21447&forum=37
"Linux is perfect for servers"
"Linux is perfect for Web servers"
"Linux is perfect for server applications"
> "A couple of weeks ago my /home suddenly went read-only, affecting
> thousands of users."
>
> "...today, I login and get "bus error" for everything on /usr and
> /sbin..."
>
> "...my main site with 25000+ users crashed and stayed crashed for six
> hours until I rebooted."
>
> "...this is definitely a kernel issue. At least, it's not a hardware
> issue..."
>
>
> http://www.centos.org/modules/newbb/viewtopic.php?topic_id=21447&forum=37
<snip>
Hmm. I read all 14 postings and the jury seems to be out on this on. The
consensus seems to be that it's probably a hardware issue. Maybe yes, maybe
no. I don't know and neither do you.
Kernel issue? It seems funny that the OP didn't mention what kernel he was
running. Was it 2.6.18? I couldn't find any widespread evidence of that
kind of bug on that kernel. Strange, eh. ;)
Attila
A kernel issue, only affecting CentOS? Yes, strange indeed. ;-)
--
Duct tape is like the force; it has a light side and a dark side, and
it holds the universe together.
Additional:
Some are suggesting it is a *drive* failure, which would make more sense.
IMO, DFS's post is just trolling *again*.
--
Windows Error: 001 - Windows loaded. System in danger!
Do you mean troll as in goblins? or troll as in fishermen? I think the
fisherman image is more appropriate although the intention is as you say.
He drags his net through various gnu/linux forums were people tend to post
things about their problems (surprise, surprise) and comes up with things
like this. Today's catch is not very good. ;)
Attila
>Some are suggesting it is a *drive* failure, which would make more sense.
>
>IMO, DFS's post is just trolling *again*.
Correction: "fsckwitted trolling"
> "Linux is perfect for servers"
> "Linux is perfect for Web servers"
> "Linux is perfect for server applications"
Micoshaft names RHAT and Canonical as their enemies!
----------------------------------------------------
BEAWAHAHAHAHAHAHAHAHAHAAAA!!!!
They got nothin better to do?
I'm thinking about that RHAT advert...
�First�they�laugh�at�us
��Then�they�fight�us
���Then�we�win
So we are now at stage 2 where micoshaft is picking a fight
by naming RHAT and Canonical as their enemies.
A few days earlier they named Intel and HP as enemies too.
And Google? - yet another enemy that has destroyed
a large chunk of micoshaft already.
Looks like micoshaft is catching its death from all around.
And only a few days ago micoshaft's Balmer claimed micoshaft
has been fighting Linux for YEARS!!!!
Thats despite all the trolls.
Verbosity makes things unclear.
Either Balmer is lying, or the windummy trolls are lying
or both are liars as Linux continues its meteoric rise
and micoshaft revenue heads south.
Windummies loons all thrown into the trollard soup after
years of lying about Micoshaft claiming it doesn't care about
Linux or that Linux is too insignificant when it was
the exact opposite.
As micoshaft sales have lost 30% of market.
how does micoshaft define competing with Linux?
It can't be anything serious.
All we see here are a bunch of windummy trolls ranting
anti-Linux slogans.
Is that what micoshaft calls competing with Linux?
Is that how they lost 30% market share to Linux?
I am sure all reporters, journalists, bloggers,
Linux engineers, and big bosses of big companies
would all like to know now what Balmer meant by
competing with Linux. Please define.
Looks like a hardware problem, albeit a rather nasty one. I've seen bad
mainboards and supply units cause trouble like this (usually caused by
crappy capacitors), with the disk controller going haywire every now and
then, or video cards crapping out under a slightly increased load. And
after a reboot, everything seems fine -- until the next glitch or brown-out
in the supply voltage causes the next file system checksum error.
In my experience, replacing either the mainboard or the supply should fix
this.
The chances of both HD's failing within a week or so are not zero,
especially when they're identical types, bought at the same time -- but
still quite slim indeed.
And oh, I've seen dual-boot systems where Linux exhibited this kind of error
(disks being set to read-only), while XP seemed to chug along just fine.
One reason for this is that Linux immediately responds to file system
errors by changing affected partitions to r/o, while XP by default silently
tries to recover from the error. I prefer Linux' behaviour, because I want
to know when a machine is starting to break -- but I think that many users
prefer XP's way: keep going until stuff really goes badly wrong.
Richard Rasker
--
http://www.linetec.nl
So did I, before posting.
> and the jury seems to be out on this on.
> The consensus seems to be that it's probably a hardware issue.
Yeah, those impartial Linux juries are quite unbiased.
> Maybe yes, maybe no. I don't know and neither do you.
The victim (and all the evidence) points to Linux crapware.
> Kernel issue? It seems funny that the OP didn't mention what kernel
> he was running.
Clearly he's a MS shill.
> Was it 2.6.18? I couldn't find any widespread
> evidence of that kind of bug on that kernel. Strange, eh. ;)
He's an "asstroturfer" making it all up.
Sorry DFS but why would anyone think the guy is a MS shill? Certainly I
didn't and you're quoting my text. I have done quite a bit of help of
various forums and newsgroups and any kernel-related thread won't get very
far (or even get started) without indicating what kernel we're talking
about. Problems with kernels tend to circulate widely because the same
kernels are used by most (all?) distros and if there's are problem for one,
it's in everyone's interest to find out about it and deal with it. Do you
think you can follow that logic. Anyway, your opions seem to be faith-based
so there's no point trying to have a rational discussion. Putting your own
sick thoughts into a rational discourse for, example, "He's
an "asstroturfer" making it all up." is not the way to discuss issues. Have
a nice day DFS. ;)
Attila
> Attila wrote:
>> DFS wrote:
>>
>>> "A couple of weeks ago my /home suddenly went read-only, affecting
>>> thousands of users."
>>>
>>> "...today, I login and get "bus error" for everything on /usr and
>>> /sbin..."
>>>
>>> "...my main site with 25000+ users crashed and stayed crashed for six
>>> hours until I rebooted."
>>>
>>> "...this is definitely a kernel issue. At least, it's not a hardware
>>> issue..."
>>>
>>>
>>>
http://www.centos.org/modules/newbb/viewtopic.php?topic_id=21447&forum=37
>> <snip>
>> Hmm. I read all 14 postings
>
> So did I, before posting.
>
>
>
>> and the jury seems to be out on this on.
>> The consensus seems to be that it's probably a hardware issue.
>
> Yeah, those impartial Linux juries are quite unbiased.
Perhaps that guy should have asked his question in a Windows forum? ;-)
>> Maybe yes, maybe no. I don't know and neither do you.
>
> The victim (and all the evidence) points to Linux crapware.
The only reason why he suspects Linux is at fault is because the machine has
developed problems with two separate disks within a week or so.
Still, all symptoms point in the direction of hardware failure. Linux' ext3
file system and its drivers are the most rock solid available, and there is
no evidence whatsoever that Linux has bugs in this respect.
In my experience, the vast majority of these intermittent failures are
caused by problems in a machine's mainboard or supply -- most often bad
capacitors (leading to glitches and brown-outs in supply voltages). But
I've seen other causes as well, such as contact problems, e.g. BGA chips
with bad solder joints -- producing weird errors, unless you press down
'just there' on the mainboard. And sometimes mainboards simply die, without
any known cause. And oh, I have a specimen still lying around here, which
reports a missing -12V upon boot, while measurements clearly show that the
minus 12 volt supply voltage is present everywhere. This board quite often
works fine, but sometimes produces weird errors -- that have nothing to do
with Linux whatsoever. So even though it appears to work most of the time,
it's quite useless.
>> Kernel issue? It seems funny that the OP didn't mention what kernel
>> he was running.
>
> Clearly he's a MS shill.
>
>> Was it 2.6.18? I couldn't find any widespread
>> evidence of that kind of bug on that kernel. Strange, eh. ;)
>
> He's an "asstroturfer" making it all up.
I think he's still somewhat inexperienced -- each hardware sysadmin worth
his salt should know about smartmontools and other hardware diagnostic
tools, and he clearly wasn't aware of the existence of any of these. Also,
he doesn't take into account the possibility that the mainboard or supply
can fail with identical symptoms. Again, the only reason why he suspects
Linux is because two HD's developed problems independently within a
relatively short time frame. Of course I can't prove that Linux is /not/
the cause of his problems, but in this case, hardware problems are far more
likely.
The odds of that happening are the kind of odds casinos love.
> Still, all symptoms point in the direction of hardware failure.
> Linux' ext3 file system and its drivers are the most rock solid
> available, and there is no evidence whatsoever that Linux has bugs in
> this respect.
Except that two filesystems exhibited the same slopware issues within two
weeks. And a reboot "fixes" the issue, at least temporarily.
He says "I will be switching to another O/S on this exact same machine with
the exact same hard drive, let's see what happens."
I bet his CentOS-caused readonly /home issue goes away (only to be replaced
by the next
set of glitchy Linux problems of course).
You have trouble recognizing sarcasm?
> Certainly I didn't and you're quoting my text. I have done quite a
> bit of help of various forums and newsgroups and any kernel-related
> thread won't get very far (or even get started) without indicating
> what kernel we're talking about. Problems with kernels tend to
> circulate widely because the same kernels are used by most (all?)
> distros and if there's are problem for one, it's in everyone's
> interest to find out about it and deal with it. Do you think you can
> follow that logic.
I can't follow the logic of "advocates": I hate Microsoft, therefore I must
use their products to make a living my entire life.
> Anyway, your opions seem to be faith-based so
> there's no point trying to have a rational discussion. Putting your
> own sick thoughts into a rational discourse for, example, "He's
> an "asstroturfer" making it all up." is not the way to discuss
> issues.
You're a little strange, fella.
> Have a nice day DFS. ;)
G'day!
Is there any other type of troll?
> or troll as in fishermen?
Does not compute. Are fishermen famed for trolling on newsgroups or
something?
> I think the fisherman image is more appropriate although the intention is
> as you say. He drags his net through various gnu/linux forums were people
> tend to post things about their problems (surprise, surprise) and comes up
> with things like this. Today's catch is not very good. ;) Attila
That's TRAWL. Fishermen man trawlers with nets that trawl the seas...
Not Troll.
--
| spi...@freenet.co.uk | "I'm alive!!! I can touch! I can taste! |
| Andrew Halliwell BSc | I can SMELL!!! KRYTEN!!! Unpack Rachel and |
| in | get out the puncture repair kit!" |
| Computer Science | Arnold Judas Rimmer- Red Dwarf |
> Attila <jdka...@gmail.com> wrote:
>> Do you mean troll as in goblins?
>
> Is there any other type of troll?
>
>> or troll as in fishermen?
>
> Does not compute. Are fishermen famed for trolling on newsgroups or
> something?
>
>> I think the fisherman image is more appropriate although the intention
>> is as you say. He drags his net through various gnu/linux forums were
>> people tend to post things about their problems (surprise, surprise)
>> and comes up with things like this. Today's catch is not very good. ;)
>> Attila
>
> That's TRAWL. Fishermen man trawlers with nets that trawl the seas...
> Not Troll.
Trolling is used for salmon, mackarel and other pelagic fish by anglers.
--
Any idiot can run XP. And usually does.
Are you a trawling troll? L:-)
--
Many enraged psychiatrists are inciting a weary butcher. The butcher is
weary and tired because he has cut meat and steak and lamb for hours and
weeks. He does not desire to chant about anything with raving psychiatrists,
but he sings about his gingivectomist, he dreams about a single cosmologist,
he thinks about his dog. The dog is named Herbert.
-- Racter, "The Policeman's Beard is Half-Constructed"
> Attila <jdka...@gmail.com> wrote:
>> Do you mean troll as in goblins?
>
> Is there any other type of troll?
>
>> or troll as in fishermen?
>
> Does not compute. Are fishermen famed for trolling on newsgroups or
> something?
>
>> I think the fisherman image is more appropriate although the intention is
>> as you say. He drags his net through various gnu/linux forums were people
>> tend to post things about their problems (surprise, surprise) and comes
>> up with things like this. Today's catch is not very good. ;) Attila
>
> That's TRAWL. Fishermen man trawlers with nets that trawl the seas...
> Not Troll.
Ahhhh. Mais non, mais non, mon ami.
This is from the Cambridge Advanced Learner's Dictionary:
"troll (FISH)
verb [I or T]
to try to catch fish by pulling a baited line through the water behind a
boat:
Boats trolling for mackerel.
They were trolling the colder waters of the Channel."
That's what I used to do as a lad. ;)
Attila
You're quite correct, Andrew & Peter. I should have talked about a line
rather than a net or else used the word "trawl" but then I couldn't have
made my joke. ;) I stand corrected.
Attila
G'day mate. ;) I don't use MS products but let's not let facts get in the
way of your fantasies. You gotta live your dream cause it's soooo much
better than reality for you. Enjoy the ride. ;)
Attila
Round these parts we just wait for the Salmon to jump out of the
stream while swimming up stream, and shoot them.
Target practice and fishing all in one!!
(only kidding)
> Attila <jdka...@gmail.com> wrote:
>> Do you mean troll as in goblins?
>
> Is there any other type of troll?
Not that I'm aware of..
>> or troll as in fishermen?
>
> Does not compute. Are fishermen famed for trolling on newsgroups or
> something?
Only Flatfish! =-)
>> I think the fisherman image is more appropriate although the intention is
>> as you say. He drags his net through various gnu/linux forums were people
>> tend to post things about their problems (surprise, surprise) and comes up
>> with things like this. Today's catch is not very good. ;) Attila
>
> That's TRAWL. Fishermen man trawlers with nets that trawl the seas...
> Not Troll.
How 'bout:
He trawls google for problems, & trolls COLA with them.
--
Simon says stand! Simon says sit! Format drive C:! Ha! Gotcha!
So that would make him "a trawling" troll". Then can be creative, "The
trawling troll truly trolled while trolling in a trolley." Can we set that
to music? ;)
Attila
> On Thu, 06 Aug 2009 16:53:12 +0200, Attila wrote:
>
But how do you keep your powder dry? Or is it the salmon swimming upstream?
But if they're swimming upstream then you are downstream (of the salmon)
and so they're moving away from you. Wouldn't it be better to use the
FOSS/Gnu/Linux solution (I'm trying to be relevant here, Mosh, bear with
me) and to move upstream of the salmon and then shoot them as they swim
towards you?
(me too -- only kidding)
;)
Attila
> William Poaster wrote:
"Computer says 'NO'"
--
I'm a internaut and I'm OK. I surf all night and I sleep all day.
> Moshe Goldfarb wrote:
>
>> On Thu, 06 Aug 2009 16:53:12 +0200, Attila wrote:
>>
You're giving me a headache :)
<snip>
>>>
>>> Round these parts we just wait for the Salmon to jump out of the
>>> stream while swimming up stream, and shoot them.
>>> Target practice and fishing all in one!!
>>>
>>> (only kidding)
>> But how do you keep your powder dry? Or is it the salmon swimming
>> upstream? But if they're swimming upstream then you are downstream (of
>> the salmon) and so they're moving away from you. Wouldn't it be better to
>> use the FOSS/Gnu/Linux solution (I'm trying to be relevant here, Mosh,
>> bear with me) and to move upstream of the salmon and then shoot them as
>> they swim towards you?
>> (me too -- only kidding)
>> ;)
>> Attila
>
> You're giving me a headache :)
Ah shucks, Mosh, you're just saying that to make me feel good, you old
flatterer. ;)
Attila
>"...today, I login and get "bus error" for everything on /usr and /sbin..."
>"...my main site with 25000+ users crashed and stayed crashed for six hours
>until I rebooted."
>"...this is definitely a kernel issue. At least, it's not a hardware
>issue..."
>http://www.centos.org/modules/newbb/viewtopic.php?topic_id=21447&forum=37
What the fuck do you care? You don't run linux.
Why the hell are you so obsessed with an operating system you don't use?
Why not ask Roy Schestowitz, [Homer] and others the same
question?
They sure seem obsessed with Microsoft and Windows.
As I said, it's quite unlikely that the disks themselves went bad, one after
the other. But file system errors don't just result from bad disks -- the
other failure mechanisms I mentioned can do the same.
>> Still, all symptoms point in the direction of hardware failure.
>> Linux' ext3 file system and its drivers are the most rock solid
>> available, and there is no evidence whatsoever that Linux has bugs in
>> this respect.
>
> Except that two filesystems exhibited the same slopware issues within two
> weeks. And a reboot "fixes" the issue, at least temporarily.
Sure, a hard reset and a subsequent fsck can correct quite a few errors.
Temporarily, as you say.
>> I think he's still somewhat inexperienced -- each hardware sysadmin
>> worth his salt should know about smartmontools and other hardware
>> diagnostic tools, and he clearly wasn't aware of the existence of any
>> of these. Also, he doesn't take into account the possibility that the
>> mainboard or supply can fail with identical symptoms. Again, the only
>> reason why he suspects Linux is because two HD's developed problems
>> independently within a relatively short time frame. Of course I can't
>> prove that Linux is /not/ the cause of his problems, but in this
>> case, hardware problems are far more likely.
> He says "I will be switching to another O/S on this exact same machine
> with the exact same hard drive, let's see what happens."
>
> I bet his CentOS-caused readonly /home issue goes away (only to be
> replaced by the next set of glitchy Linux problems of course).
I bet that Windows (if installed) will appear to perform with less glitches
at first, until it crashes with severe data loss (as silent recovery only
goes so far), eventually showing that it was a bad mainboard, supply or
even disk controller was the cause after all. Been there, seen it. These
symptoms you attribute to "Linux crapware" have in my (quite extensive)
experience always been caused by hardware problems.
I use Windows, and I like to argue about Linux and Windows.
And make you dung-eating cola liars look like the raging idiots you are.
Still waiting on proof of a "million .tmp files"... how long do we have to
wait, liar?
Yep.
And that one was such an easy one to prove.
A simple screen shot, or cli dir dump etc.
AzGonads likes to tell stories and when he gets asked for proof
he falls apart at the seams.
You want a million .tmp files? How soon do you need them?
--
You never have to change anything you got up in the middle of the night
to write.
-- Saul Bellow
> On 2009-08-07, the following emerged from the brain of DFS:
>> Still waiting on proof of a "million .tmp files"... how long do we have to
>> wait, liar?
>
> You want a million .tmp files? How soon do you need them?
Do you deliver to Ma?
Go fuck yourself. I'm not going to jump hoops for a lying piece of shit
like you.
Hahahahah!
Another LinuxTard has been Rope-A-Doped........
Notice how they get vulgar when they are caught lying?
Tomorrow evening is fine. Put them all on your desktop and show us how they
look. Please.
Another frothing cola man-child that can't back up his lies and stupidity.
You wacks are emotional basket cases.
Yes they certainly are.
Notice how silent HPT is today....
He got his ass kicked the last 2 days.
I'm using awesome, so I don't really have a desktop, but I'll figure
something out :-)
--
Have a place for everything and keep the thing somewhere else; this is not
advice, it is merely custom.
-- Mark Twain
Okay, the one million .tmp files are in place:
% pwd
/home/tommy/one_mio
% ls -1 | wc -l
1000003
% du -h
29M
I'll provide you with a screenshot later today. What have I won now?
--
If one cannot enjoy reading a book over and over again, there is no use
in reading it at all.
-- Oscar Wilde
Ma? I'm afraid I don't get this...
I can ftp them to you if you like. It's only 29 megs ;-)
DooFu$ hasn't got his nickname for nothing. He's just another dumb M$
fanboi troll.
--
Why do psychics have to ask you for your name?
> Moshe Goldfarb wrote:
>> On 07 Aug 2009 01:40:11 GMT, TomB wrote:
>>
>>> On 2009-08-07, the following emerged from the brain of DFS:
>>>> Still waiting on proof of a "million .tmp files"... how long do we have to
>>>> wait, liar?
>>> You want a million .tmp files? How soon do you need them?
>>
>> Do you deliver to Ma?
>
> Ma? I'm afraid I don't get this...
Massachusetts .
> I can ftp them to you if you like. It's only 29 megs ;-)
--
Two most common elements in the universe: Hydrogen & Stupidity.
Ah, right.
>Okay, the one million .tmp files are in place:
>% pwd
>/home/tommy/one_mio
>% ls -1 | wc -l
>1000003
>% du -h
>29M
>I'll provide you with a screenshot later today. What have I won now?
The difference is that linux can delete those million temp files in
an hour or so on the NTFS volume or in just a few minutes on a
decent file system, whereas vista needs the better part of a week.
I don't care what vista application left the temp files. The problem is
that vista curled up and died when left in such a situation. It devoted
most of it's resources doing god knows what; probably indexing all of them
for desktop search.
You're following up your bullshit lie about Vista creating a "million temp
files" with another bullshit lie about needing the "better part of a week"
to delete these files that never existed in the first place.
You're a total loser.
He's trying to lie his way out of a lie.
And not doing a very good job of it.
Massachusetts.
Ah, that was what it was all about...
Results on Linux:
tommy@mordor:~$ date +%s && rm one_mio/*.tmp && date +%s
1249657953
bash: /bin/rm: Argument list too long
Ooops...
This worked though:
tommy@mordor:~$ date +%s && rm -rf one_mio/ && date +%s
1249658577
1249658642
65 seconds. Not bad. I can't say what Vista'd do as I don't have it
installed here.
--
Q: Know what the difference between your latest project
and putting wings on an elephant is?
A: Who knows? The elephant *might* fly, heh, heh...
>Results on Linux:
>Ooops...
>This worked though:
Vista needs at least 5 days. I timed it and it couldn't do 100 files/second,
and most of the time barely managed 25 files/second.
Fortunately, I had a knoppix live CD. I used that to clear out the
temp files. It's hilarious needing linux to maintain windows systems.
Here's some basic remedial math for the idiot. And I'll even use his (wrong)
claim that Vista "barely managed" 25 files/second.
1,000,000 files
25 files/second = 40,000 seconds
40,000 seconds = 666.667 minutes
666.667 minutes = 11.11 hours
11.11 hours = 0.46 days
But somehow this ignorant MORON concludes that deleting 1 million files at
25 files/second works out to "at least 5 days."
Yeah, I did the math too. That was a bit silly of AZ...
Anyway, I still had my Vista machine running at work, so I am creating
the one million files right now. Unfortunately I don't know a lot
about batch scripting, so I had to run a /bin/sh script in cygwin to
generate the files. Man, that's slow... I can't say if this is because
of Vista or because of the shell running in cygwin though...
Does anyone know a command to display unix time from a Windows command
prompt? And how does one strings commands together in the Windows
shell?
--
You're almost as happy as you think you are.
Hmm, lets change the subject line...
--
Break into jail and claim police brutality.
> Anyway, I still had my Vista machine running at work, so I am creating
> the one million files right now. Unfortunately I don't know a lot
> about batch scripting, so I had to run a /bin/sh script in cygwin to
> generate the files. Man, that's slow... I can't say if this is because
> of Vista or because of the shell running in cygwin though...
Both.
Making a 100 Mb bzip2 tar-ball on Cygwin with XP is excruciating.
--
O, it is excellent
To have a giant's strength; but it is tyrannous
To use it like a giant.
-- Shakespeare, "Measure for Measure", II, 2
You would be better off using the freely downloadable PowerShell. Much
more powerful than standard batch files.
Same thing with using mutt. Too slow to be useful :-(
--
The difference between the right word and the almost right word is the
difference between lightning and the lightning bug.
-- Mark Twain
You're right. Vista can't even manage 25 files per second. It must
have been more like 4 per second.
>1,000,000 files
Newsflash: millions is plural. The system I was cleaning up had
MILLIONS of tempfiles.
Perhaps it would have only been three days. Probably vista wasn't even
managing 5 files per second. I really don't care. It was months ago
and only assholes like you are still obsessing over it.
> On 2009-08-07, the following emerged from the brain of Chris Ahlstrom:
>> After takin' a swig o' grog, TomB belched out
>> this bit o' wisdom:
>>
>>> Anyway, I still had my Vista machine running at work, so I am creating
>>> the one million files right now. Unfortunately I don't know a lot
>>> about batch scripting, so I had to run a /bin/sh script in cygwin to
>>> generate the files. Man, that's slow... I can't say if this is because
>>> of Vista or because of the shell running in cygwin though...
>>
>> Both.
>>
>> Making a 100 Mb bzip2 tar-ball on Cygwin with XP is excruciating.
>
> Same thing with using mutt. Too slow to be useful :-(
ssh works pretty well, though.
Quite often, though, I use cygwin's shell for copying files, because
XP's Explorer waits for network shares to time out when you simply click on
a *local* resource. Crazy.
--
You are deeply attached to your friends and acquaintances.
Newsgroups: comp.os.linux.advocacy
From: AZ Nomad <aznoma...@PremoveOBthisOX.COM>
Subject: oh, god! I hate vista!
Message-ID: <slrng9ur5c.q...@ip70-176-155-130.ph.ph.cox.net>
User-Agent: slrn/0.9.8.1pl1 (Linux)
Date: Sun, 10 Aug 2008 17:33:16 -0500
[quote]
Today I was trying to help a neighbor whose 5 month vista system is getting
slower and slower.
I installed spybot and the new version, 1.6, wanted to catalog a pile of
temp
files. Apparently, there are 2.5 million temp files. No fucking idea what
process is leaving them behind. Probably the "yahoo search" in HP advisor
that came factory installed.
In linux, I'd be able to remove them all in a few minutes. Vista is
going to
need about 30 hours to get rid of them. Wether I do it from the command
line,
let spybot do it, or do it in safe mode, it can only remove about
20-30/second
I also tried to turn off file indexing for the hard drive. Vista wanted to
set attributes on every single fucking file in the entire system. I gave up
on that idea; I didn't want to wait the week it would require.
Unlike a vista laptop I recently struggled with that "only" had 512M of ram,
this system has 3 gig of ram, 6 ghz of processor and a 70MB/s sata drive.
And vista being a bloated pig run like shit.
Right now I'm back at home, copying a hundred or so of movies to a usb
drive,
doing 3 simultanious transcodes. I'd never know it. Even though the
CPUs are
pegged at 100%, the file transfer is pluging away find at 40M/second and the
response time of the rest of the system is just like when the system is
idle.
In vista, if I so much as opened the properties dialog for "my computer",
the window deleting the temp files would slow to 2-3/second. Vista is
*such*
a piece of crap.
[/quote]
"6 ghz of processor" you say? "2.5 million temp files"?
Why must you always resort to lying about that which can so easily be
disproved?
>>>>>>Okay, the ***one million*** .tmp files are in place:
>>>>> The difference is that linux can delete those million (**singular**)
>>>>> temp files in
Idiot. Go take some remedial math classes. Perhaps you can find a smart 8
year-old to help you learn arithmetic.
> Perhaps it would have only been three days. Probably vista wasn't even
> managing 5 files per second. I really don't care. It was months ago
> and only assholes like you are still obsessing over it.
Blah-blah-blah. More lies from the lying idiot who can't even do simple
math. Simple math proves that you're simply spewing ignorant bullshit.
Ssh, file management and burning CDs are my main uses of cygwin on my
Vista box at work.
I find file management under Vista just not accaptable. It's so
bloated and messy it makes my head hurt. I'd go nuts if I didn't have
cygwin (and Total Commander).
--
Talkers are no good doers.
-- William Shakespeare, "Henry VI"
It's called "LIEing for LIEnux" and it's a tactic used by
overzealous Linux "advocates" to make Linux look good.
> Newsflash: millions is plural. The system I was cleaning up had
> MILLIONS of tempfiles.
uh huh... sure it did.
> Perhaps it would have only been three days. Probably vista wasn't
> even managing 5 files per second.
Earlier you said 25 files per second.
> I really don't care. It was months ago and only
> assholes like you are still obsessing over it.
You're still a liar. You know it and I know it and everyone else knows it.
What do you call it when the overzealous Windows "advocates" lie to make
Windows look good?
--
Rick
>> It's called "LIEing for LIEnux" and it's a tactic used by overzealous
>> Linux "advocates" to make Linux look good.
>
> What do you call it when the overzealous Windows "advocates" lie to make
> Windows look good?
It is equally bad either way.
--
[INSERT .SIG HERE]
I took a look at it. Quite verbose syntax on first sight.
The 'time /t' command works just fine, and it seems you can string
commands together with '&&' just like in a proper shell. Good. Now I
just have to wait for the one million files to be generated. After 12+
hrs I'm almost at 300000. The same script on my Debian box took less
than an hour by the way.
#!/bin/sh
cnt=0
while [ "$cnt" != "1000001" ]; do
touch $cnt.tmp
cnt=`expr $cnt + "1"`
done
If someone can provide me with a Windows batch script alternative, that
would be greatly appreciated. I suppose that would run a lot faster.
--
You have the power to influence all with whom you come in contact.
> The 'time /t' command works just fine, and it seems you can string
> commands together with '&&' just like in a proper shell. Good. Now I
> just have to wait for the one million files to be generated. After 12+
> hrs I'm almost at 300000. The same script on my Debian box took less
> than an hour by the way.
Not surprised.
> If someone can provide me with a Windows batch script alternative, that
> would be greatly appreciated. I suppose that would run a lot faster.
Maybe a little.
--
The camel died quite suddenly on the second day, and Selena fretted
sullenly and, buffing her already impeccable nails -- not for the first
time since the journey begain -- pondered snidely if this would dissolve
into a vignette of minor inconveniences like all the other holidays spent
with Basil.
-- Winning sentence, 1983 Bulwer-Lytton bad fiction contest.
> On 2009-08-07, the following emerged from the brain of Chris Ahlstrom:
>> After takin' a swig o' grog, TomB belched out
>> this bit o' wisdom:
>>
>>> Anyway, I still had my Vista machine running at work, so I am creating
>>> the one million files right now. Unfortunately I don't know a lot
>>> about batch scripting, so I had to run a /bin/sh script in cygwin to
>>> generate the files. Man, that's slow... I can't say if this is because
>>> of Vista or because of the shell running in cygwin though...
>>
>> Both.
>>
>> Making a 100 Mb bzip2 tar-ball on Cygwin with XP is excruciating.
>
> Same thing with using mutt. Too slow to be useful :-(
It's Liarmutt. And he's sucking up again with zero proof and is also
discredited in this very same thread. Trust *nothing* he says.
--
http://www.faqs.org/faqs/windows-emulation/wine-faq/
"Nope, we know what an emulator does, and wine doesn't." - AH
** http://groups.google.com/group/comp.os.linux.advocacy/msg/dec7cb073d761af4
Thanks for the script. It took about 25 minutes to run it on Vista.
Deleting the files took about 50 minutes.
On my Debian box creating the files took about 45 minutes. Deleting
them took about 1'20".
Here are the screenshots:
http://www.drumscum.be/mio_win.png
http://www.drumscum.be/mio_lin.png
> If someone knows what the DOS command equivalent for the Unix command
> 'time' is, I'd be happy to know what it is. :)
Uh? 'time' *is* a DOS command.
--
Q: How many psychiatrists does it take to change a light bulb?
A: Only one, but it takes a long time, and the light bulb has
to really want to change.
Is deleting a directory the same as deleting the contents and then the
directory with fat32/ntfs?
It might be interesting to let the files sit on the machine for a few
days so that desktop search gets them indexed and try the mass delete
again.
I still don't have a good idea why that vista machine I repaired was
so incredibly slow at deleting the mass of tempfiles. My current theory
is that vista was updating the desktop search index upon file deletion.
I have no idea. I guess so.
--
Few things are harder to put up with than the annoyance of a good example.
-- "Mark Twain, Pudd'nhead Wilson's Calendar"
I turn off indexing on Vista as I have no use for it. I know where my
files are.
--
Go to a movie tonight. Darkness becomes you.
You're quite some guy Tom. You know where your files are. You never do a
typo in a root shell. I wonder why they invent these things at
times! :-;
It took between 5 and 6 minutes to create the files and 2 min 5 sec to
delete on my Win7RC box. Please note that my computer is pretty high-
end so I don't know how much of the better performance I'm seeing vs.
your Vista box is due to the hardware and how much is due to Win7
improvements over Vista. I did notice that neither the CPU nor the hard
disk were getting hit very hard during the run so Win7 may have had some
significant improvements achieved in this area. Wasn't disk I/O
performance one of the big areas of complaint in Vista?
>I have no idea. I guess so.
Hadron, unix expert, doesn't know what 'rm -rf' does.
I'm just great :-p
I *do* make typos in root shells by the way. I even gave an example of
such an error in that particular thread.
But I really don't need indexing. I use 'find' occasionally, but most
of the time I have no problems in finding what I'm looking for.
--
Q: What do you call a WASP who doesn't work for his father, isn't a
lawyer, and believes in social causes?
A: A failure.
> On 2009-08-08, the following emerged from the brain of Hadron:
>> TomB <tommy.b...@gmail.com> writes:
>>> Thanks for the script. It took about 25 minutes to run it on Vista.
>>> Deleting the files took about 50 minutes.
>>>
>>> On my Debian box creating the files took about 45 minutes. Deleting
>>> them took about 1'20".
>>>
>>> Here are the screenshots:
>>> http://www.drumscum.be/mio_win.png
>>> http://www.drumscum.be/mio_lin.png
>>>
>>>> If someone knows what the DOS command equivalent for the Unix command
>>>> 'time' is, I'd be happy to know what it is. :)
>>>
>>> Uh? 'time' *is* a DOS command.
>>
>> Is deleting a directory the same as deleting the contents and then the
>> directory with fat32/ntfs?
>
> I have no idea. I guess so.
>
On FAT it certainly is, as you have to traverse the FAT structure (both of
them) to delete the linked list entries and replace them with "not-
allocated"
On NTFS I have to look, but I can't be arsed to. Not important enough.
--
Never promise to complete any project within six months of the end of
the year -- in either direction.
What the hell are you talking about?
1) I was referring to Windows
2) I was asking about a single deletion of a directory as opposed to a
recursive delete.
Once more you demonstrate your cluelessness.
Lying for Linux hurts the reputation of all of us who use Linux. Lying
for Windows just hurts the reputation of the person telling the lie.
It's not fair, but that's how it is in general when you have a big
majority group and a small minority group.
--
--Tim Smith
nice non-answer.
--
Rick
Intel Pentium Dual CPU E2140 @ 1.60GHz.
1GB RAM.
7200 RMP drive.
Vista Business 32 bit.
Much lower specs indeed.
My Debian box is an AMD X2 4400 with 2 GB of RAM.
>> On my Debian box creating the files took about 45 minutes. Deleting
>> them took about 1'20".
>
>> Here are the screenshots:
>> http://www.drumscum.be/mio_win.png
>> http://www.drumscum.be/mio_lin.png
>
> When I did my initial test the domain controllers and file server were
> freaking out causing problems on our workstations. Now that
> everything is copacetic up in the server room once again, I get even
> better results.
>
> 6.5 minutes to create a million files and 2.5 minutes to delete them.
>
> http://bdhi.ath.cx/images/millwin.jpg
>
>>> If someone knows what the DOS command equivalent for the Unix command
>>> 'time' is, I'd be happy to know what it is. :)
>
>> Uh? 'time' *is* a DOS command.
>
> Unix 'time' times command execution. Unix 'date' is the equivalent of
> DOS 'time', but has many more options.
>
> Now, do you know a DOS command equivalent to the Unix 'time' command or
> not? LOL
Ah, I was just confused there.
New answer: not that I know of.
--
There is a great discovery still to be made in Literature: that of
paying literary men by the quantity they do NOT write.
> I *do* make typos in root shells by the way. I even gave an example of
> such an error in that particular thread.
I haven't made such an error in quite awhile. The last one I did has made
me vewwy vewwy carweful.
--
The very ink with which all history is written is merely fluid prejudice.
-- Mark Twain
> In article <FsudndWF74xvf-HX...@supernews.com>,
Okay, time for Baye's Theorem.
P(A|B) = P(B|A) P(A) / P(B)
A = Hurting the reputation of X users.
B = An X user lying about X.
P(A|B) is the probability that the reputation of an X user will be hurt,
given that the X user is lying about X.
So we need the rest of the numbers:
P(B|A) is the probability of an X user lying about X given
that it will hurt the reputation of X.
1. How likely is it that a Windows user will lie, knowing it will hurt
the reputation of Windows users?
2. How likely is it that a Linux user will lie, knowing it will hurt
the reputation of Linux users?
P(A) is the probability of hurting the reputation of X users.
1. How likely is the hurting of the reputation of Windows users?
2. How likely is the hurting of the reputation of Linux users?
P(B) is the probability of of an X user lying about X.
1. How likely is a Windows user to lie about X?
2. How likely is a Linux user to lie about X?
If you can seriously pin a real number on any of those values, I'd be very
surprised.
So, let's try a different tack, and restate the problem using another
minority group.
1. Lying for Black people hurts the reputation of all who are
Black people.
2. Lying for White people just hurts the reputation of the person
telling the lie.
If you believe statement 1, that would be because /you/ think that the lying
hurts the reputation of all Black people, or because /you/ think that
everyone else thinks the same way. In other words, /you/ are stereotyping
Black people.
Just as you are stereotyping Linux people.
--
Q: What is printed on the bottom of beer bottles in Minnesota?
A: Open other end.
> <rant>
>
> I too disable indexing on my Windows boxen. I was fucking livid, but not
> surprised considering the source of the software, when I took Chrome for
> a test drive and found it turned indexing back on. I turned it back off
> and Chrome turned it back on once again. <grrrrrrr>
>
> Fuck Google!
> </rant>
>
> Yes, I feel better now. :)
Interesting...
A friend of mine said Chrome did the exact same thing on his
system.
Idiot.
Under Windows, which is what he is speaking of, it does nothing.
> Timmy wrote:
>>
>> Lying for Linux hurts the reputation of all of us who use Linux. Lying
>> for Windows just hurts the reputation of the person telling the lie.
>> It's not fair, but that's how it is in general when you have a big
>> majority group and a small minority group.
>
>Okay, time for Baye's Theorem.
>
>(snip)
>
>If you believe statement 1, that would be because /you/ think that the lying
>hurts the reputation of all Black people, or because /you/ think that
>everyone else thinks the same way. In other words, /you/ are stereotyping
>Black people.
>
>Just as you are stereotyping Linux people.
Right. Timmy is wrong. Again.
How is that stereotyping black people? It would be making a comment on
what *everyone else* thinks. Everyone else involves every other race,
does it not? If anything, it's stereotyping the entire human
population, which of course doesn't really qualify as stereotyping.
Now everyone else could be stereotyping black people, but not the /
you/ you refer to in that statement.
>
> Just as you are stereotyping Linux people.
>
See above. Tim said nothing about "Linux people." He's not doing the
stereotyping, just commenting that it exists (sort of, I mean it's not
really stereotyping).
> On Aug 8, 8:01?pm, Chris Ahlstrom <ahlstr...@launchmodem.com> wrote:
>>
>> If you believe statement 1, that would be because /you/ think that the lying
>> hurts the reputation of all Black people, or because /you/ think that
>> everyone else thinks the same way. ?In other words, /you/ are stereotyping
>> Black people.
>
> How is that stereotyping black people? It would be making a comment on
> what *everyone else* thinks. Everyone else involves every other race,
> does it not? If anything, it's stereotyping the entire human
> population, which of course doesn't really qualify as stereotyping.
> Now everyone else could be stereotyping black people, but not the /
> you/ you refer to in that statement.
You have a point, sort of. The fact is that people tend to think that
others think along much the same lines they do.
For example, that is why "Hadron" is always shouting "liar" and "suck up".
To be still more clear: If you think that one person lying tars a whole
group, then you are /definitely/ guilty of stereotyping. Just as if you
think one Pole doing something stupid indicates some fact about the rest of
them.
>> Just as you are stereotyping Linux people.
>
> See above. Tim said nothing about "Linux people." He's not doing the
> stereotyping, just commenting that it exists (sort of, I mean it's not
> really stereotyping).
Nah, it is still stereotyping. Unless he were to eventually find solid
probabilities to fill into the equation, then his stereotyping would
actualize.
--
Generosity and perfection are your everlasting goals.
> cc belched:
>>
>> See above. Tim said nothing about "Linux people." He's not doing the
>> stereotyping, just commenting that it exists (sort of, I mean it's not
>> really stereotyping).
>
>Nah, it is still stereotyping. Unless he were to eventually find solid
>probabilities to fill into the equation, then his stereotyping would
>actualize.
It's just another way for the Timmy troll to attack Linux advocates.
"You're not good enough", Timmy claims.
I suppose Timmy's "logic" justifies him attacking advocates, while
being silent regarding the filthy liars who Wintroll in here.
What the fuck does that even mean? Where did he claim that someone
wasn't good enough? You're just making up points to argue against now.
> I suppose Timmy's "logic" justifies him attacking advocates, while
> being silent regarding the filthy liars who Wintroll in here.
Uhh, isn't that exactly what Tim was trying to justify (although not
exactly in those words)?
> #!/bin/sh
> cnt=0
> while [ "$cnt" != "1000001" ]; do
> touch $cnt.tmp
> cnt=`expr $cnt + "1"`
> done
I bet you would cut your time considerably if you changed line 5 to:
cnt=$((cnt+1))
Try it.
A quick test with 10,000 files, cut my time down from 30 to 15 seconds.
> If someone can provide me with a Windows batch script alternative
Pass.
--
K.
http://slated.org
.----
| "The shepherd drives the wolf from the sheep's throat, for which
| the sheep thanks the shepherd as his liberator, while the wolf
| denounces him for the same act, as the destroyer of liberty.
| Plainly the sheep and the wolf are not agreed upon a definition of
| the word liberty; and precisely the same difference prevails today
| among human creatures." ~ Abraham Lincoln
`----
Fedora release 8 (Werewolf) on sky, running kernel 2.6.26.8-57.fc8
18:40:33 up 75 days, 22:38, 5 users, load average: 0.00, 0.00, 0.06
Probably. It was just a quick and dirty script anyway.
--
You will be a winner today. Pick a fight with a four-year-old.
> On 2009-08-12, the following emerged from the brain of Homer:
>> Verily I say unto thee, that TomB spake thusly:
>>
>>> #!/bin/sh
>>> cnt=0
>>> while [ "$cnt" != "1000001" ]; do
>>> touch $cnt.tmp
>>> cnt=`expr $cnt + "1"`
>>> done
>>
>> I bet you would cut your time considerably if you changed line 5 to:
>>
>> cnt=$((cnt+1))
>>
>> Try it.
>
> Probably. It was just a quick and dirty script anyway.
The point is not that it could be done but that the OP made a
stink about Vista *Creating* these millions of files and then
being unable to *delete them* in a reasonable amount of time.
When pressed for proof he ran away with his tail between his
legs.