« first day (1139 days earlier)      last day (4177 days later) » 

13:00
OK. Looking forward to another 30 years or so with x86 then. :)
What is that. Celeron 1 sockets to... look sort-of like slot A
PGA370
@Hennes i think it is one of the dual adapter things, hard to find pics of them. I never tried that 3rd party adaptions things.
Aye. I used that. Hmm, can also run P3's
or P2's
I used Asus P2B-D board IIRC.
Dual socket boards for two pentiums or two celrons
With Old slot based pentiums to PGA370 convertors
And two celeron 366/466's hacked to allow dual usage
(The celeron and Pentium core were identical. Just some soldering and a resistor would make a celeron-1 dual CPU capable)
13:08
Is that a slotket?
@Hennes What OS were you running on that?
damn my taskbar is stuck again, i should proably try win8 mabey they just discarded taskbars altogether :-) moved the icons from the destop to the front app, where they are really accessable.
Linux
@Hennes Red Hat?
13:12
Dual seleron 366's. FSB set to 83MHz. (so a 26% OC). Running dnetc 24/7 in an attaic with a room temp of 30+ Celsius
Slackware
Cool! :)
that fixed it, mabey i just scared the computer into cooperating. I wonder if a spray bottle would work. Everytime it acts up i will just spritz it :-)
"scared the computer into cooperating" LoL! :)
@Hennes The soft clock so much fun They said it was impossible if the bios did not support it. . Then they said my board was not supported. (impossible again). so i just tried every profile in the whole list, and got it.
Bob
Bob
@Psycogeek Remember that most x86 is now compiled down to microcode similar to RISC instruction sets.
13:21
-- Now some genius has decided to put VRMs on the die, right after they put the cap on the die in the worst way (haswell aka hasbeen to some)
@Overflowh Welcome to Root Access chat for Super Users! Please don't ask if you can ask or if anyone's around; just ask your question, and if anyone's free and interested they'll help.
Bob
Bob
@allquixotic ^
grabs popcorn
@Bob uhh no, i dont know what an instruction even is barely , that is in the matrix, the oracle knows, but she talks in cryptic abbreviations.
Automatic messages, nice.
Hi guys
Bob
Bob
3 mins ago, by Bob
@Psycogeek Remember that most x86 is now compiled down to microcode similar to RISC instruction sets.
@Psycogeek Whoops, replied to the wrong post. Sorry.
@Overflowh Hi.
13:27
If my processor is running at 1.3Volts and it is (max) using about 150Watts of juice how many Amps is that?
Bob
Bob
115 A?
Though I haven't seen a CPU with a TDP of 15 W in a while. If at all.
Some of the P4s went up to about 115 W TDP.
Wow, Pentium Ds went up to 130...
> Core 2 Extreme QX9775 Yorkfield XE (45 nm) 3.2 GHz 150 W
O.O
> Core i7 3970X Extreme Edition Sandybridge-E (32 nm) 3.5 GHz @ 6 Cores 150 W
O.O
(And of course there's a bunch of Xeons, though that's not entirely unexpected.)
@Bob This is when i beat the heck out of it overclocking it. at 115Amps pumped into the chip, the On-die VRMs have a loss of ~0.3v so is only ~34watts of extra heat or ~ 1/3rd more.
Hi
Bob
Bob
@Kwaio Hi
My question may be dumb, but how can you make "echo" or any other i/o program to wait before sending EOF ?
Maybe I would have to write the program myself...
Bob
Bob
13:38
@Kwaio Context?
@Kwaio have it not send one at all, then add one yourself.
Bob
Bob
I suppose they could simply keep the stream open (you don't 'send' EOF - it's not a character, it's an identifier to mark a closed stream, e.g. with C's getchar())
@Psycogeek EOF is not a character
1
Q: Verifying which SMTP server software is used by remote server

KwaioI am trying to catalog which smtp servers are used in an heterogenou pool of servers, and so far i was doing : echo "HELO localhost" | telnet localhost 25 2>/dev/null | grep 220 #process the output of command My problem is the server doesn't always have time to answer "220 ..." before the soc...

Bob
Bob
Ouch.
I've struggled a lot with netcat in a similar situation.
I'll be back in ~1 hour - I'll see if I can help then.
@Bob Dont tell me, i am not the one who asked to have it wait before "sending" it :-)
Bob
Bob
13:42
Aha!
@Kwaio Try this:
14
A: Creating a script for a Telnet session?

VatineI've used various methods for scripting telnet sessions under unix, but the simplest one is probably a sequence of echo and sleep commands, with their output piped into telnet. Piping the output into another command is also a possibility. Silly example (echo password; echo "show ip route"; sle...

Basically you want to run a sleep after the echo in a subshell.
@WakanTanka Forcing the echos and sleeps to run in a sub-shell, sharing a stdout and actually have the sleep pause the output before the "quit" command is echoed, to allow the full routing table to print, before the connection dies. — Vatine Jul 17 at 10:49
Good luck!
@Bob I am trying to work this out in my head, if there is no "item" used like 00 to define the EOF how does the "process" know it has reached the last item? Mabey it waits like its on the phone to it, and when it isnt talking anymore it just Knows. The AI .
Bob
Bob
@Psycogeek the stream is closed
@Bob oh, how does the stream know to close? Throw me a bone here how does a stupid computer know things ?
Bob
Bob
then it depends how the program is reading it. I'll elaborate when I get back to a computer
like if it was only a specific chunk size it was reading from to begin with, the chunk is a known.
@Kwaio heck if you can spell heterogenou , there is no way i am going to be any help :-)
In computing, end of file (commonly abbreviated EOF) is a condition in a computer operating system where no more data can be read from a data source. The data source is usually called a file or stream. In the C Standard Library, the character reading functions such as getchar return a value equal to the symbolic value (macro) EOF to indicate that an end-of-file condition has occurred. The actual value of EOF is system-dependent (but is commonly -1, such as in glibc) and is unequal to any valid character code. Block-reading functions return the number of bytes read, and if this is fewer th...
13:58
woop gotta phone call
so
gotta love the wiki, ""and is unequal to any valid character code"" uhh no it is human definable, and machine readable, it is valid, it is recognised as the EOF , it might not be valid for printing . It might not have a glyph assigned, even if it was a recognisable black hole, it is there.
@Bob hum... intresting
Bobobobob
i'll try the subshell method now
14:23
@Kwaio You'll want to use netcat, if you want to do something more complex then use expect; there's no need to rely on subshell and sleep functionality.
I can't
Bob
Bob
@allquixotic hmhmhmhm?
@Psycogeek That's just for C
it's very library dependent
again, EOF is not an actual character
it's more the state the stream is in
either the stream has reached the end (or closed), or it has not reached the end (or is still open)
@Bob yes you guys are referring to keeping a connection open. like you said , context.
Hey guys... just need a reality check here:
2
Q: What are all these "ad.doubleclick.net" pages I'm secretly visiting?

KronoSI'm using IE 10 and a few pages that I visit appear to be constantly visiting ad sites behind the scenes. For example after going to a pastebin post I went to select "Back" to only find I was returning to the same site. After right clicking on the back button I get the following trail of sites ...

Is this really all that confusing of a question?
re:
@Kronos, the point is, you're not doing any research, nor are you answering questions that make it possible for us to actually give you an answer that solves your problem. Anyhow, good luck! — techie007 4 mins ago
Yay ii just forgot grep "-e" options were doing OR with each other not AND
i killed all ssh sessions and half the applications \o/
14:35
@KronoS gotta love it, the answer is so simple and so known it is easier to type that were not going to tell you, than to answer it :-)
@Psycogeek Ya that appears to be the attitude
Bob
Bob
@Kwaio to do and, pipe it through another grep
@Bob probably not gonna update to the new dedi server model. they regressed their disk space offering
Bob
Bob
@allquixotic ah. by how much?
@KronoS: I don't think its particularly simple - I'd guess, entirely speculatively, its a way to tell how long you're on a site.
14:41
>
> How much would it cost monthly to have an EX40 server with 2 x 3 TB
> SATA disks?

3 TB enterprise disk costs 25 Euros per month.

 Would it be possible to remove the default 2 TB disks and
> replace them with 3 TB disks?

Yes, but we won't refund the removed 2 TB enterprise HDDs.

Enterprise class is not necessary, but
> if it is required to use Enterprise class, then please calculate the
> price for Enterprise class.

We don't put non enterpirse disks in EX40 any longer.

>
> The configuration I would like to have is:
There is a way to restore my Windows 7 installation from linux? I'm really frustrated, it's the third time in two months that Windows decides to have worked enough.. And suddenly it doesn't start anymore..
Bob
Bob
@Overflowh "restore"?
@JourneymanGeek Perhaps... the main thing I'm trying to gauge here is whether or not I'm being ridiculous with my original post and helpful to those commenting
you'll probably want something that will monitor what your web browser and the server are saying to each other.
(though I feel that they aren't being that helpful)
14:42
so I'd be paying more for disks than for the base server itself (mobo, PSU, CPU, RAM, network uplink, chassis and energy draw)
that makes a lot of sense
@KronoS:If its that simple, they can say why it is in the comments ;p
@JourneymanGeek exactly. I'm just trying to get a reality check here, and making sure I'm not being a jerk basically :)
Bob
Bob
@allquixotic hm. weird.
@Bob Sounds like it's time to email Mr. Martin Hetzner again and say, in a more respectful tone, basically that I've been a customer for a long long time, want to remain so, but I wish they'd credit me for not having the two 2TB disks installed, that way I could just pay the difference between the cost of 2TB disks and the cost of 3TB disks
last time I mailed him I got them to waive an upgrade fee and process a cancellation early
@Bob Yeah.. restore, recover, repair.. I tried the utility in the Windows installation disk (the one in the left corner: "Repair your PC"), but half the options in that utility give back errors, the other half took almost all the day to say "I cannot repair your pc nor restore from a early point"..
14:45
@Overflowh: would backing up from windows, then restoring using a third party disk be ok?
failing which, probably clonezilla (which backs up and restores from a livecd)
@Overflowh there's no general-purpose "FIX MY WINDOWS NAO" button in Linux, so you'll need to actually understand the exact nature of the problem you're experiencing to even begin to resolve it using Linux
Bob
Bob
@allquixotic knowing the owner helps :P
@Overflowh it would help a lot if you could tell us what problem you're hacing
saying something is "broken"... there's pretty much an infinite number of ways something could be not working.
@JourneymanGeek I'm just coping the important directories from my disk to an extern one using Mint from usb (yeah, I choose mint just because it has wifi drivers pre-installed.. debian don't)
@Overflowh: I'm trying to get what exactly you're ok with ;p
@allquixotic Is the reason why I'm asking for your help. It's the third time that the problem appear, and I still cannot figure out what causes it..
14:51
SHUT UP CLIPPY.
@JourneymanGeek Yeah yeah, it's ok for me to back up :) since it's the third time, I found a quite fast way for doing this xD
@Overflowh: clonezilla.org is what I'd use for a livedisk - that uses this partclone.org
me, I probably end up using DD ;p
(well, ddrescue, more idiot proof)
@JourneymanGeek A live disk..?
yup
pop in a cd, back up the whole system to a external drive
Bob
Bob
@Overflowh unless I missed something, you still haven't mentioned what the "problem" actually is...
14:53
or an additional internal drive
@Bob Yeah, give me just a second to describe it
@JourneymanGeek Oh.. well, in this moment I'm backing up not the whole disk, but just the important files for me (to an external hard drive). But leaving intact the actual status of the disk "damaged". I'm using Mint 15, installed with persistence on an USB key
Well, now the problem
Generally, those three times, this is what happened:
1. Windows initially have some problems at start up: taking too much in loading all the boot components (just dropbox, intel drivers and the antivirus), taking too much in start programs, freezing with the simplest actions like opening a directory or a new tab in firefox. (and this for one and half day)
2. this morning, the pc stopped working. It reaches the point with the black background and the Windows blinking logo, and stopped there, for hours.
I turned off the pc, and tried to use the "Repair your PC"
Windows has a blinking logo?
I think 7 has a logo thats animated.
do you mean blinking cursor in the top left/
But it cannot find out a system restore point the first time (giving back an error). The second time I tried it found one, but after the restoring process, nothing changed.
15:00
?
I can't remember 8's logo
@allquixotic mabey blinking No-go , some of the symtoms are similar to a drive failure.
@allquixotic Well, not really blinking.. I don't know an adapt adjective, sorry..
Bob
Bob
@allquixotic I think Win7 kinda flashes bright now and then.
Bob
Bob
15:02
@Overflowh Have you given safe mode a shot? Helps with dodgy drivers.
@Bob Yeah, I tried to start the safe mode, but it's appeared again the utility of "Repair your PC"
Bob
Bob
Skip that one.
The startup repair appears after a failed attempt to boot.
Just skip it and select safe mode.
@jokerdino
lol
@KronoS: might play with it when I find it/and the time to but thirdtier.net/2012/06/… wfetch might be useful for your issue
Bob
Bob
@JourneymanGeek While you're looking at capturing traffic, I'd recommend Fiddler2.
@Bob I have that running right now ;)
Bob
Bob
15:11
lol
currently using it to serve replacement JS for testing
don't mind me, just injecting iframes into text fields
I don't want to overload my actual systems with too many testing tools so, I'll probably need to find time to spin up a new VM
Bob
Bob
> don't want to overload my actual systems
> spin up a new VM
O.o
@Bob I was about to quote the exact same thing lol
it's easy to fall into that trap though -- Biofuels seem to fall into the same mode of thinking: "I don't want to run out of crude oil mined out of the ground, so I'm going to make my own oil!" "...hmm... wait... this actually takes MORE oil out of the ground to make this biofuel than it'd take to just mine the oil out of the ground and burn it! damn... oh well... don't tell the President, this is making me good money!"
15:27
@Bob: I mean, I don't want to install a bunch of stuff I will use once, then forget ;p
my systems can run VMs fine, and, I can just dispose of the whole VM once I'm done
@AzkerMohamed Welcome to Root Access chat for Super Users! Please don't ask if you can ask or if anyone's around; just ask your question, and if anyone's free and interested they'll help.
Bob
Bob
@JourneymanGeek I wouldn't even bother with Fiddler. It only runs when you start it; it doesn't run in the background.
You can do the following:
1) Install a working copy of windows on one partition (lets call this C:\)
2) Create a second partition to hold your data (The usually D:\)
3) Make sure your my documents etc uses D:
4) Boot Linux and make a raw disk backup of C:

Whenever a problem occurs which is not trivial to fix restore that raw disk image, boot windows (now working) and make sure you run windows update
"Error reading from file: Input/output error"
Someone knows why I'm getting back this error in copying from the internal to the external drive? It appear only with few files..
Bob
Bob
@Overflowh Such as?
Give an example of one of the files.
You might be hitting one of the reparse points.
15:32
@Bob A .jpg image, some strange Spotify files, a Firefox's bookmark .json file
One way to make that backup is with dd (optionally pied though gzip)
Bob
Bob
o.O
Or you could just use clonezilla, symantec ghost, acronis trueimage, ...
Bob
Bob
@Overflowh @Psycogeek's guess may be right - you might have a failing HDD.
@Hennes Wow.. it's possible to tell Windows to use another Partition for the, say, User directory?
15:33
SMART values should show failing disks
Bob
Bob
@Hennes No. SMART is one indicator of a failing disk.
A 'good' SMART report does not necessarily mean that a disk is fine.
Not sure if it works for C:/user/. I know it works for my documents (which is my case points to d:)
Aye
It just means it detected problems. Not the absence of problems
Bob
Bob
It's important to keep that kind of thing in mind - and quite dangerous to assume that the disk is fine if SMART reports no errors.
However most of the time it is correct, so it is a good place to start with
Bob
Bob
Ya.
15:34
@Bob Oh god no..
Bob
Bob
@Hennes Just keep in mind that that won't capture newly installed programs, changed settings, etc. :P
Which is why it's good to have some scheduled system for taking the images.
Correct. Including newly installed malware ;)
Bob
Bob
Update it monthly, or weekly.
And keep a base image for emergencies
Bob
Bob
Yep.
With incremental images, you could even have a 'known good' point, and weekly/daily increments on top, while being fairly space efficient..
@Overflowh As @Hennes said, SMART is a good place to start.
There are also some tools that purport to do a surface scan of the disk, but it's a good idea to get your data off it (ASAP).
15:38
@Bob I'm sorry, but I really don't have any idea about what SMART is..
@Bob wat?
I know what the L1 lagrange point is. but. that an asteroid?
almost kills the moon at one point o_O
Bob
Bob
@Overflowh It's basically information about the drive. It'll tell you if, for example, the drive had to reallocate sectors at some point.
you should be able to read it with smartctl in Linux
@allquixotic Part of a Saturn rocket, actually :P
__NOTOC__ J002E3 is the designation given to a supposed asteroid discovered by amateur astronomer Bill Yeung on September 3, 2002. Further examination revealed the object was not a rock asteroid but instead the S-IVB third stage of the Apollo 12 Saturn V rocket (serial S-IVB-507). When it was first discovered it was quickly found that the object was in an orbit around Earth. Astronomers were surprised at this as the Moon is the only large object in orbit around the Earth and anything else would have been ejected long ago due to perturbations with the Earth, the Moon and the Sun. Therefo...
@Bob It's part of smartmontools? Mind does not recognize smartctl
Bob
Bob
@Overflowh I'm not sure tbh
I'm normally in Windows for this kind of thing.
@Bob Ok thanks.. it's installing..
15:43
> It is thought that J002E3 left Earth orbit in June 2003, and that it may return to orbit the Earth in about 2032.
lol... it likes us
Mmmh.. any help? smartctl seems quite complicated..
Bob
Bob
@Overflowh smartctl --all /dev/sdX
where /dev/sdX is the drive
@Bob Tried, but tell "Permission denied"
15:48
@Psycogeek I wonder if they sell any full size pages of that. It would be lovely to wrap that around an ordinary newspaper
With both sda and sdc
As root? Or with sudo ?
Since you are using raw disk access, so you would either need uid 0, effective uid 0, or be in the disk group.
Bob
Bob
Weird that Verizon apparently needs the device to be certified...
Here, you just buy a SIM. Generally they don't give a shit where/how you use it.
@Hennes Thanks, with sudo it worked.. and gave back an amount of unreadable (for me) informations..
Bob
Bob
Please copy it into here. Or put it on pastebin.com and link here.
15:50
Some background: you are sending direct commands (raw) to the device. Some of those commands can be "wipe yourself". You do not want every single user to be able to do that. :)
@Bob @Hennes pastebin.com/ReRRFFrp this is the report. Thanks even just for opening it..
@Hennes Yeah I know.. I always forget the power of sudo :/
 5 Reallocated_Sector_Ct   0x0033   100   100   010    Pre-fail  Always       -       5
Bob
Bob
lots of pre-fails there
not a good sign
@Hennes hm. actually, the numbers look fine
except the raw value. hrm.
What does this means?
During its life/usage this happened 5 times:
* Try to read a sector -> Fail or get incorrect data
* Retry until you succeed
* Quicly make a backup of that data
15:55
@JourneymanGeek Can't use chrome due to work restrictions
COmpare it to reading a book. You fail to read a page, go though quite some ffort to decipher the page beneath a coffee stain and make a copy at the end of the book. Then you write on the stained page: Do not try to read this, use page 100 instead.
That is a re-allocated sector
Reallocated sectors happen. No disk drive is perfect
However they should not increase.
If this nunmber increases then that is a good indicator of pending failure.
So.. 5 times in.. I think two/three years is not so bad, right?
Agree.
However there is
197 Current_Pending_Sector  0x0012   100   100   000    Old_age   Always       -       2042
Those are pages which did not succeed in getting rewritten to a spare page yet.
If that is 2042 unreadable sectors then DOOM is upon us.
I am not 100% sure that is is 2042 sectors though. Most manufacturers use custom smart values. But I really do not like that number.
Upon me.. :/

« first day (1139 days earlier)      last day (4177 days later) »