background disk write problem

[vc]

Please refer to the image attached below, for the comparison results.  Tests were conducted in the following manner, after an initial reboot:

1)  Launch GKrellm;
2)  Launch KSnapshot, click 'New Snapshot' button;
3)  Wait several minutes, while observing indicated system activity;
4)  Perform the screen capture when indicated activity constant, and remaining at a minimum;
5)  Save screencap, close apps, Logout;
6)  Log in to next environment, repeat test procedures.

All screencaps were then edited in GIMP, re-assembled into a new image, labeled, and Saved.

Reviewing the results now cause me to wonder:  what do AfterStep, KDE4, and XFCE4 have in common?

[Joble]

That is very interesting.  Can't answer the question but I appreciate the amount of effort and it's very interesting indeed.

[Texstar]

It you notice a little tick to your hd then it is probably Gamin.

Gamin is a file and directory monitoring system defined to be a subset of the FAM (File Alteration Monitor) system. This is a service provided by a library which allows to detect when a file or a directory has been modified.

[vc]

It you notice a little tick to your hd then it is probably Gamin.

Gamin is a file and directory monitoring system defined to be a subset of the FAM (File Alteration Monitor) system. This is a service provided by a library which allows to detect when a file or a directory has been modified.

Does that then mean that the environments which do not suffer the background disk write behaviour are also not activating or making use of gamin, somehow?  I had assumed gamin to be a system service; a daemon perhaps, running constantly in the background regardless of selected window manager environment - am I mistaken in this assumption?  Additionally; why does top not indicate the gamin activity?  Current capture of top output follows:

Code: [Select]

top - 14:12:05 up  1:51,  3 users,  load average: 0.38, 0.45, 0.29
Tasks: 106 total,   1 running, 104 sleeping,   0 stopped,   1 zombie
Cpu(s): 15.4%us,  2.1%sy,  0.0%ni, 81.6%id,  0.9%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:   1260928k total,   610164k used,   650764k free,    36156k buffers
Swap:        0k total,        0k used,        0k free,   299436k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                            
14672 me         7   0  2480  948  716 R  2.0  0.1   0:00.01 top                                                
    1 root       1   0  1796  564  496 S  0.0  0.0   0:00.86 init                                                
    2 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kthreadd                                            
    3 root       1   0     0    0    0 S  0.0  0.0   0:00.01 ksoftirqd/0                                        
    4 root       1   0     0    0    0 S  0.0  0.0   0:00.09 events/0                                            
    5 root       1   0     0    0    0 S  0.0  0.0   0:00.00 khelper                                            
    6 root       1   0     0    0    0 S  0.0  0.0   0:00.00 netns                                              
    7 root       1   0     0    0    0 S  0.0  0.0   0:00.00 async/mgr                                          
    8 root       1   0     0    0    0 S  0.0  0.0   0:00.00 pm                                                  
    9 root       1   0     0    0    0 S  0.0  0.0   0:00.29 sync_supers                                        
   10 root       1   0     0    0    0 S  0.0  0.0   0:00.00 bdi-default                                        
   11 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kintegrityd/0                                      
   12 root       1   0     0    0    0 S  0.0  0.0   0:00.06 kblockd/0                                          
   13 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kacpid                                              
   14 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kacpi_notify                                        
   15 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kacpi_hotplug                                      
   16 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kseriod                                            
   18 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kswapd0                                            
   19 root       1   5     0    0    0 S  0.0  0.0   0:00.00 ksmd                                                
   20 root       1   0     0    0    0 S  0.0  0.0   0:00.00 aio/0                                              
   21 root       1   0     0    0    0 S  0.0  0.0   0:00.00 crypto/0                                            
   24 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kpsmoused                                          
   40 root       1   0     0    0    0 S  0.0  0.0   0:00.21 ata/0                                              
   41 root       1   0     0    0    0 S  0.0  0.0   0:00.00 ata_aux                                            
   43 root       1   0     0    0    0 S  0.0  0.0   0:00.00 scsi_eh_0                                          
   44 root       1   0     0    0    0 S  0.0  0.0   0:00.00 scsi_eh_1                                          
   55 root       1   0     0    0    0 S  0.0  0.0   0:00.00 reiserfs/0                                          
   65 root       1   0     0    0    0 S  0.0  0.0   0:00.09 flush-8:0                                          
  141 root       1  -4  2176  836  412 S  0.0  0.1   0:00.11 udevd                                              
  306 root       1   0     0    0    0 S  0.0  0.0   0:00.00 ksuspend_usbd                                      
  308 root       1   0     0    0    0 S  0.0  0.0   0:00.00 khubd                                              
  341 root       1   0     0    0    0 S  0.0  0.0   0:00.00 usbhid_resumer                                      
  355 root       1   0     0    0    0 S  0.0  0.0   0:00.00 kstriped                                            
  386 root       1   0     0    0    0 S  0.0  0.0   0:00.00 khpsbpkt                                            
[me@localhost ~]$

[vc]

That is very interesting.  Can't answer the question but I appreciate the amount of effort and it's very interesting indeed.

The matter is of concern (or should be) to those whom may wish to install PCLinuxOS into any computer employing SSDs for storage, as the backround disk write activities do add up over time.  What these tests have revealed is that three desktop environments do manifest this behaviour pattern, and so should be avoided for SSD-based installs - whereas the remaining available desktop environment choices do not, and therefore should be significantly less wearsome upon SSD devices.

I have avoided installing 2010 into my EeePCs for this very reason; however, I now see there are at least some alternatives that may be considered.

[Joble]

I would recommend E-17.  The settings menu take a little getting used to, much different looking than the kde, but it is a very capable desktop environment once you get used to it.  I've got the wood theme and I like it a lot!  

I do really appreciate the research you've done.  I have a netbook with a mechanical hard drive not ssd.  I'm not worried about the live of an ssd drive, but about the amount of time I can get out of my battery.  So far I have lasted 6 hours at the coffee shop before the battery beep started.  If background writes are a problem, I might stretch that out to 6+ hrs by switching desktops, thx to your research.  Of course there is a downside, that's a lot of coffee.   

[vc]

I would recommend E-17.  The settings menu take a little getting used to, much different looking than the kde, but it is a very capable desktop environment once you get used to it.  I've got the wood theme and I like it a lot!

I agree, Joble - Enlightenment is an attractive environment, and when the 2010 E17 Final is released I shall install same and explore.

I do really appreciate the research you've done.  I have a netbook with a mechanical hard drive not ssd.  I'm not worried about the live of an ssd drive, but about the amount of time I can get out of my battery.  So far I have lasted 6 hours at the coffee shop before the battery beep started.  If background writes are a problem, I might stretch that out to 6+ hrs by switching desktops, thx to your research.  Of course there is a downside, that's a lot of coffee.  

You're welcome.  I did encounter some difficulties with capturing the results from awesome, as I had initially found that particular window manager somewhat confusing - yet the effort in total took only a couple of hours, and seems to illustrate the point adequately.  Additional data obtained, such as the memory utilisation figures, could also perhaps be of interest to those whom may be considering installation into old P2/K6-2 laptops or somesuch.

Regarding the battery discharge duration of the laptop unit you referred to - have you considered the possibility of obtaining an 'aftermarket' replacement battery of higher capacity?  There were such listed at the DealExtreme website; the listings indicated some replacement battery packs for the Asus EeePC 700 and 900 series models as being available for reasonable price levels - maybe there could be a battery pack for your particular unit, as well?

If not, then another option would be to purchase a 12-Volt automotive charging adapter for your laptop instead, and to connect that adapter to an external 12-Volt 'gel cell' battery, of the type that would usually be incorporated into a UPS unit.  A typical 12-Volt 20-Amp/hr 'gel cell' battery weighs only a few kilos, fits readily into a small strapped cloth 'lunch' bag or a plastic container with a carrying-handle, and would power your laptop continuously for an additional eight or ten hours.  A larger 'marine' sort of battery (with integral carrying handles) would last through an entire weekend - I could envision some circumstances where such could be useful, if perhaps not quite so portable as the former.

[Joble]

Uh.... What?  6 hours at the coffee shop is not enough?  If I ever get a cabin off the grid in the hills with copper wire mesh (not that I've ever considered it) I will consult.  

BTW:  I have considered it.

[vc]

BTW:  I have considered it.

I've already lived it... hence the prior '12-Volt lifestyle' experience, and previous accompanying flashback....

Irrelevant.  Returning now to the thread topic; I'm still curious - what is causing this background disk write behaviour, anyway?  Why did a Minime 09 install suffer this behaviour initially, yet now no longer?  Would the 2010 KDE4 background disk write behaviour also cease, eventually?  Why should some window manager environments be associated with this behaviour, whereas others are not - on the same system?  As well; why do some environments cause GKrellM to indicate the 'presence' of two users logged into the system, whereas again, others are indicated as having only one user logged in, instead?

These are all clues; they are each nodes and vertices belonging to a particular pattern - a full outline of which has not yet been found, nor successfully fitted.  My opinion of this situation is that finding the underlying cause and arriving at a satisfactory resolution of the problem could be of significant benefit to PCLinuxOS - would that be a reasonable line of thought to pursue?  Or, as with the KDE4 'cashew', must we resign ourselves to acceptance, instead?