If I would take a wild guess, I would say, disable XGL ! ! !

It's not running right now either. It is installed, but not started... and for what it's worth.... I have no issues running XGL. My system hums along fine with it running or using regular X. C. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On Wednesday 11 April 2007 12:20, Dylan wrote:

On Wednesday 11 April 2007, Clayton wrote:

...

Amarok hums along just fine with Ruby dead and gone.. no functionality is lost in Amarok - that I can see.

...

Has anyone else noticed this issue? Any suggestions or ideas what is going on?

C.

Is Amarok running a script you don't use which is triggering ruby to start in the first place?

Undoubtedly. Perhaps one of those things that proclaims to the world (via a Web page, blog, email or whatnot) what music you're listening to at the moment.

Dylan

Randall Schulz -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On Wed, Apr 11, 2007 at 09:28:24PM +0200, Clayton wrote:

A little experimenting... it's the Fetch Lyrics sript that is spawning Ruby - starting the script starts Ruby, and stopping the script stops Ruby. Ok, so now I know what is likely starting Ruby... I wonder what is going wrong that makes Ruby spike the CPU like that. Weird...

A few options: either start reading the code and looking for something monumentally stupid :) or strace the process and try to find the monumentally stupid thing _that_ way.

From the look of it, it was coincidental that Ruby was involved. Ruby seemed to be involved because whatever script it was handling was attempting read or write into the bad area. The result... the system

I traced the problem.... a failing IDE hard drive. Here's a little info to close off this thread. paused briefly while it tried to sort out the read/write error. I had ignored the error because of a long term battle I've had with Maxtor IDE drives and Linux. I've never had a Maxtor drive that didn't spew out DriveSeek errors from time to time... despite the fact the drives doing this never actually failed, and passed all diagnostics. This one was doing it too... so I ignored it.. assuming it was more of the same. On a whim I enabled the SMART daemon, and immediately it popped up a warning the the drive was failing. I ran smartctl, and it started throwing out messages about checksum errors on the drive, and loads of unrecoverable errors. As soon as I moved all the data on that drive to another drive and unmounted the drive... problem solved. I only lost one easily replaceable file on the dying drive. I'm assuming the drive is unrecoverable (ie it's not worth risking my data to try and run the Maxtor tools on it and "repair" the drive)... so it's up for replacement in the next week or so. C. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On Tuesday 17 April 2007 03:15, Carlos E. R. wrote:

...

There is an applet that might have helped. You know those kinds of applets that show a cpu usage graph? But there is one type of "busy" that doesn't usually show, and it is the percent of time the cpu is waiting for I/O. Typicals are "user, system, nice, idle", but IOwait is usually neglected. The gnome cpu monitor does show it, in black over black, and I always change it to some dark color that I can see: thus I know when my system becomes unresponsive but cpu is low that disk usage is high; ie, that the cpu is waiting for the disk.

The KDE panel applet "System Monitor" shows I/O wait time in the CPU column as green. To my knowledge, GKrellM does not offer this distinction.

...

-- Cheers, Carlos E. R.

Randall Schulz -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Hello, On Tue, 17 Apr 2007 22:27:36 -0500 "Rajko M." wrote:

On Tuesday 17 April 2007 04:15, Clayton wrote: ...

...

I'm assuming the drive is unrecoverable (ie it's not worth risking my data to try and run the Maxtor tools on it and "repair" the drive)... so it's up for replacement in the next week or so.

You may try to slow down the drive. I got one that was about to fail, but decreasing the speed, using Maxtor utilities, from 66 to 33 helped and it still runs.

Probably, the decreasing speed problem is because the HDD controller changed a bad sector into a spare sector. and the spare sector possibility don't optimize to seek, I guess. AFAIK, A modern HDD has a number of spare sectors. and when happened a bad sector, that can replace a bad sector into a spare sector by using the hard drive tools, etc. *But*, there is a limitation in the number of spare sectors (and depend on HDD). For example, the following results by a smartctl on my HDD. # smartctl -A /dev/hda | egrep -e 'ATTRIBUTE_NAME|Reallocated_Sector_Ct' ID# ATTRIBUTE_NAME FLAG VALUE WORST THRESH TYPE UPDATED WHEN_FAILED RAW_VALUE 5 Reallocated_Sector_Ct 0x0033 100 100 005 Pre-fail Always - 2 The HDD used two spare sectors because the HDD had two bad sectors. And if the raw_value exceed the threshold, then it is necessary to change it with a new HDD, I think. So, there is a one way to defend the data from the bad sector. It is a RAID (Excluding RAID0). I would recommend the Linux Software RAID. In addition, you would want to read the following documents. http://help.com/wiki/Bad_sector http://smartmontools.sourceforge.net/BadBlockHowTo.txt Thanks, eshsf -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Randall R Schulz wrote:

On Wednesday 18 April 2007 09:01, eshsf wrote:

...

...

...

You may try to slow down the drive. I got one that was about to fail, but decreasing the speed, using Maxtor utilities, from 66 to 33 helped and it still runs.

Probably, the decreasing speed problem is because the HDD controller changed a bad sector into a spare sector. and the spare sector possibility don't optimize to seek, I guess.

What would be the relationship between bus transfer rate and seek time? None that I can think of.

Maybe he's talking about how fast the disks spin. ;-) -- Use OpenOffice.org http://www.openoffice.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Ok, I am a post to the list because I think that's no problem. On Thu, 19 Apr 2007 23:24:23 +0400 Aaron Kulkis wrote:

eshsf wrote:

...

Hello,

On Tue, 17 Apr 2007 22:27:36 -0500 "Rajko M." wrote:

...

On Tuesday 17 April 2007 04:15, Clayton wrote: ...

...

I'm assuming the drive is unrecoverable (ie it's not worth risking my data to try and run the Maxtor tools on it and "repair" the drive)... so it's up for replacement in the next week or so.

You may try to slow down the drive. I got one that was about to fail, but decreasing the speed, using Maxtor utilities, from 66 to 33 helped and it still runs.

Probably, the decreasing speed problem is because the HDD controller changed a bad sector into a spare sector. and the spare sector possibility don't optimize to seek, I guess.

AFAIK, A modern HDD has a number of spare sectors. and when happened a bad sector, that can replace a bad sector into a spare sector by using the hard drive tools, etc. *But*, there is a limitation in the number of spare sectors (and depend on HDD).

Assuming a 15ms seek time, I sincerely doubt that the original poster would have noticed the effects of this... that would account for a whopping 30ms (i.e. 0.03 seconds)...and once the data is read into memory, it's done ... that disk block is now in the I/O buffers, and there is no need to revisit it.

exactly. when I did post, I didn't consider it at all, almost all the HDD have the I/O buffers on drive and the Linux have a buffer cache though. Yes, so, only when the first accessing it or flushed a cache, it might influence but the influence is a little bit. Thanks, eshsf -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On Wednesday 18 April 2007 22:41, Randall R Schulz wrote:

I think the only hypothesis supported by this observation is something related to the electronics. Possibly a damaged cable or out-of-spec bus driver circuits on the mainboard or in the disk drive.

It is. The board runs fine with another Western Digital that is at maximum transfer rate that board supports, cable is the same and the connector is now used for WD drive, so culprit is controller on the disk drive. Is that manufacturing or design flaw I can't tell, but disk is still alive. -- Regards, Rajko. http://en.opensuse.org/Portal -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On Thu, 2007-04-19 at 23:24 +0400, Aaron Kulkis wrote:

Clayton wrote:

...

I've been trying to chase down an extremely annoying problem I've been having. On what you could almost call heartbeat regularity, roughly once ever 2 to 3 seconds, my computer has been doing a micor-pause. One of the CPUs (AMD X2 3800+) spikes to 100% and everything freezes for a fraction of a second. You don't notice it unless you're watching a video file or playing a game... then it becomes annoying to the extreme.

I don't have zmd installed.. or beagle... so I started looking into the processes running... stopping or killing one and seeing if it made a difference. I traced it down to ruby. If ruby is running I get the regular CPU spike. Kill it... problem is gone. I can't uninstall ruby without pulling down Amarok as well. Not much of an option.

I know that Ruby is a scripting language... beyond that not a lot... question is, how is it that Ruby is killing my performance so much. If it's a scripting language.. what is it doing that spikes my CPU to 100%... or what would be using it and spiking the CPU with whatever it"s doing.

Amarok hums along just fine with Ruby dead and gone.. no functionality is lost in Amarok - that I can see.

Has anyone else noticed this issue? Any suggestions or ideas what is going on?

All I can say is.. if it were a "micro" pause, then how would you even notice it in the first place?

Macro = that which is descernable by the body/mind. Micro = that which is too small to be seen/felt/etc without instrumentation (such as a microscope..)

I'm a computer systems engineer, I've been programming since 1980, I've used and programmed on more operating systems than I can remember... what in the hell is a "micro pause" because I have never seen that term used before...ever...for anything.

Wow! And despite that, you don't know how to google? I got plenty of hits on "micro pause" (but in your defence, I didn't come up with any hits on what it means in hell).

Sounds to me like you're just making up terms to ascribe to the

Do you really mean to say that people need to be as qualified as you in order to ask a question here? He's obviously describing the problem as he's experiencing it. So what if the "terms" he's using might not be scientifically correct?

microprocessor what is probably a hardware fault (probably related to I/O), or a programming bottleneck (like say, an exclusive

And as far as he's concerned, he traced it down to a programming bottleneck in Ruby. He therefor asked a question about that.

lock on 10,000 rows on a database, causing all other processes which try to access those rows to be put in a sleep state until the rows are unlocked -- yes, I've seen this happen -- a 4-CPU machine running 96% idle was upgraded to an 8-CPU machine running at 98% idle).

Cheers, Magnus -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Carlos E. R. wrote:

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1

The Thursday 2007-04-19 at 23:24 +0400, Aaron Kulkis wrote:

...

Date: Thu, 19 Apr 2007 23:24:03 +0400

You have a problem with your email provider.

...

All I can say is.. if it were a "micro" pause, then how would you even notice it in the first place?

Macro = that which is descernable by the body/mind. Micro = that which is too small to be seen/felt/etc without instrumentation (such as a microscope..)

I'm a computer systems engineer, I've been programming since 1980, I've used and programmed on more operating systems than I can remember... what in the hell is a "micro pause" because I have never seen that term used before...ever...for anything.

Wow, you are picky!

Yes, I am. Anyone in the field of computing, or any field of engineering, needs to learn to good communication skills. Precise problems DEMAND precise communications. You know.. the difference between "The car doesn't work" and "it stalls when you press the accelerator real hard." Learn to communicate EFFECTIVELY, people!

Obviously, you haven't read his explanations, because he explained his meaning very well.

I'm not questioning his explanation, I'm questioning the sloppy terminology. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On May 01, 2007 07:19 PM, Aaron Kulkis wrote:

...

Wow, you are picky!

Yes, I am.

Anyone in the field of computing, or any field of engineering, needs to learn to good communication skills.

  When you are in a boxing ring ... learn to keep your chin low, or else you'll get a right hook, and a knockout.  You just left your chin wide open up there, with the extra "to" between learn and good. :-) ----------------------------------------- Aegroto, dum anima est, spes esse dicitur Powered by Open-Xchange.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Aaron Kulkis wrote:

I'm a computer systems engineer, I've been programming since 1980, I've used and programmed on more operating systems than I can remember... what in the hell is a "micro pause" because I have never seen that term used before...ever...for anything.

I also have no idea what a "micro pause" is. The only thing that comes to mind, was that on some mini-computers, such as the Data General Eclipse, you could microstep through the microcode. There was a pause in there, but I don't think that's what he had in mind. -- Use OpenOffice.org http://www.openoffice.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Aaron Kulkis wrote:

James Knott wrote:

...

Aaron Kulkis wrote:

...

I'm a computer systems engineer, I've been programming since 1980, I've used and programmed on more operating systems than I can remember... what in the hell is a "micro pause" because I have never seen that term used before...ever...for anything.

I also have no idea what a "micro pause" is. The only thing that comes to mind, was that on some mini-computers, such as the Data General Eclipse, you could microstep through the microcode. There was a pause in there, but I don't think that's what he had in mind.

Wow... DG... Haven't heard mention of them in a LONG time.

I went to Purdue...we had some VAX-11/780's that were "homemade" dual-VAX machines (George Goble's pioneering work on dual CPU UNIX)...and then some Goulds (Powernode 9080's, and then NP-1's), and in electrical engineering, ONE Data General... but it was running System V wherease everything else was running 4.2/4.3 BSD and since it wasn't networked, either, nobody used it.

But I hear they were nice machines to program on.

I wasn't a programmer. I was a technician, who looked after the hardware. However, even at that, I noticed a lot of nice things about the Eclipse (and even Nova) instruction set. I also worked on the VAX 11/780, some PDP-11's, a PDP-8i, a few different PR1ME models and some Collins 8500C computers. There was even one, count 'em one XT clone. ;-) My company ran VAX/VMS on the 11/780, which I used when doing my Fortran homework. One thing I liked about the VAX was the disk pack drives could be arranged in array, shared by several computers, whereas with Data General, only two computers could share a drive. This was back in the days before ethernet, though the Collins computers had a network based on time slots over a triaxial cable. My first ethernet experience was with those Vaxes, though it was coming in, just as I was leaving that position. -- Use OpenOffice.org http://www.openoffice.org -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Sigh.. why is there always someone....

All I can say is.. if it were a "micro" pause, then how would you even notice it in the first place?

If you bloody read my original post you would have seen that I DIDN'T notice the pause unless I was playing a game or watching a video. The pause was so small that it did not interfere with normal day-to-day stuff.

Macro = that which is descernable by the body/mind. Micro = that which is too small to be seen/felt/etc without instrumentation (such as a microscope..)

Yes Mr. Pedantic those are definitions of the words (probably taken from that highly unreliable source everyone quotes as gospel truth, Wikipedia), but you have also demonstrated despite the fact that you've been programming since 1980 (what this has to do with micro and macro has yet to be stated) you have no understanding of the actual concepts. Sigh Since you're too busy with your diatribe to actually bother with something as mundane as understanding the concepts, I'll help you. Micro comes from a Greek root word that simply means small. In English it is often used to mean exactly that... small or very small. If you go look up the word micro in a dictionary... lets take Merriam Webster for example... micro is defined as "very small, involving minute quantities or variations" which was exactly the context I used the word in. But.. since you're obviously right... what with programming since 1980 and all, you had better go correct the people who use the words micro-computer since their computers are obviously too small to be seen without instrumentation. Oh, and the people who do micro-economics... and those poor people who drive micro-cars in the Netherlands... oiy, that must be a challenge! Cars so small you need a microscope to see them... wait... a microscope.. .aaaaahhh.. I'm caught in a spiral of illogic.

remember... what in the hell is a "micro pause" because I have never seen that term used before...ever...for anything.

I made it up... .so sue me. Don't bother replying Mr Kulkis. I filter pedantic twits. Back on topic... I resolved the problem. I discovered that the micro pauses were being caused by a failing hard drive. There was a bad sector on the drive, and the computer was (obviously) struggling with reading the bad sectors. It was co-incidental that the problem happened when Ruby was running... most likely because the script that was being processed needed to access some data on the bad part of the drive. I backed up the data, and took the failing drive out of service. Problem solved. No more micro pauses. C. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

Sorry to barge in on your very well defined arguments ... but this was one was easy enough for even my damaged brain not to barf at :-)

Ha ha ha.. no problem :-)

Micro in a micro-computer,was referring to the technology used in creating the chips which used to be micro technology,

True enough... my intent was simply to point out the absurdity of the argument. I worked in the semiconductor industry for a long time. I know the subject well. Funny though, you never seem to see SUSE in the industry... loads of Solaris based software though (running the wafer steppers and wafer scanners). C -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org

On May 02, 2007 09:27 AM, Clayton wrote:

True enough... my intent was simply to point out the absurdity of the argument.

I worked in the semiconductor industry for a long time. I know the subject well. Funny though, you never seem to see SUSE in the industry... loads of Solaris based software though (running the wafer steppers and wafer scanners).

  Wafer scanners?   I've never worked with the technology myself, but as far as I can understand, it is a laser that is blowing bubbles on a silicon dye?  The micro or nano technology referring to the size of the dye used?  Where do you use the scanners? ----------------------------------------- Aegroto, dum anima est, spes esse dicitur Powered by Open-Xchange.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org