openbox on older computer ?

[Muppet]

Hi Melodie,

did you see yesterday something weird in the pastebin ?

i have just for my own interest another dsmeg command in terminal and something i noticed.

at the end of the pastebin and the ones before i posted somewhere near the line nr.800 there is the message;

localhost gdmgreeter[1897]: WARNING: Theme broken: must have pam-message label!

localhost net_applet[2055]: ### Program is starting ###

but just now for the first time i see this in addition;

localhost ntpd[1903]: synchronized to LOCAL(0), stratum 10
localhost ntpd[1903]: kernel time sync status change 0001
localhost ntpd[1903]: synchronized to 131.211.84.189, stratum 1

are these 3 lines saying anything important perhaps ? i find it so curious that i have never seen this before....................

regards,

Willem

[melodie]

I looked just before the restart message:

Oct 17 15:13:21 localhost ntpd[1982]: kernel time sync status 0040
Oct 17 15:13:22 localhost ntpd[1982]: frequency initialized -37.484 PPM from /var/lib/ntp/drift
Oct 17 15:13:22 localhost klogd: usbcore: registered new interface driver snd-usb-audio
Oct 17 15:14:05 localhost gdmgreeter[1866]: WARNING: Theme broken: must have pam-message label!
Oct 17 15:14:17 localhost net_applet[2135]: ### Program is starting ###
Oct 17 15:16:36 localhost ntpd[1982]: synchronized to 213.239.154.12, stratum 2
Oct 17 15:16:36 localhost ntpd[1982]: kernel time sync status change 0001
Oct 17 15:23:08 localhost syslogd 1.5.0: restart.

Have you configured your timezone ?

Have you tried older kernels ?

[Muppet]

I looked just before the restart message:

Oct 17 15:13:21 localhost ntpd[1982]: kernel time sync status 0040
Oct 17 15:13:22 localhost ntpd[1982]: frequency initialized -37.484 PPM from /var/lib/ntp/drift
Oct 17 15:13:22 localhost klogd: usbcore: registered new interface driver snd-usb-audio
Oct 17 15:14:05 localhost gdmgreeter[1866]: WARNING: Theme broken: must have pam-message label!
Oct 17 15:14:17 localhost net_applet[2135]: ### Program is starting ###
Oct 17 15:16:36 localhost ntpd[1982]: synchronized to 213.239.154.12, stratum 2
Oct 17 15:16:36 localhost ntpd[1982]: kernel time sync status change 0001
Oct 17 15:23:08 localhost syslogd 1.5.0: restart.

Have you configured your timezone ?

Have you tried older kernels ?

Hi,

yes, i have configured my timezone.

no, i have not yet tried the older kernels. i am gonna try first to use one hdd instead of the 2 hdd that are connected. if that does not work i will go to try the older kernels.

i have noticed that the flatcable is connected from the motherboard to the first hdd then a other 20cm is going thru into the second hdd. i do not remember if this flatcable was bought afterwards while installing the second hdd or that it was default there when delivered from factory.

the psu is 200w on this machine.

come back later.................

edit;

had the freeze now installed with one hdd only. occurred around 20 minutes after startup. i had hoped that with one hdd installed it would worked ok. to fix that was probably to change the psu. but it didn't unfortunate.............oefff....what is this a persistant item.

next is trying older kernels

regards,

Willem

[Muppet]

Hello Muppet again,

i have now installed kernelversion 2.6.37.4 for i686 computers.

when i did the reboot after downloading from synaptic i saw in the grub that the kernelversion of openbox is 2.6.38.8.bfs
in synaptic there is also a 2.6.38.8 for i686 computers.

what does the bfs mean ?

should i have better downloaded the kernelversion 2.6.37.4.bfs instead of the current as mentioned above to try out and better focus only on the 'bfs' versions if i have to download another older kernelversion ?

Willem

[melodie]

Hi,

Openbox have bfs to follow the line given by Texstar with his official version.

This is not important : the important in regard of the problem you meet with is testing older kernels, whatever version. Non bfs (brain f*** schedular) are alright as well for this purpose. Please let us know how it goes ?

And about the PSU : if you could also provide the brand. Is it what is called a "no name" ? 200watts is not really powerful and I have also been told that they often have a value written and can provide even lower power than what they are supposed to. So let's see with one hard drive instead of two and with an older kernel.

I would also like to remind you that you have provided lines from "messages" but I would also have liked to see what a dmesg after a reboot following a freeze looks like : a full dmesg.

Regards,
Mélodie

[Muppet]

Hello,

have installed the kernelversion 2.6.37.4 i686 on one hdd.

after i had installed this kernelversion i saw these kind of lines when startingup saying;

unable to locate IOAPIC for GSI...

you can see at pastebin here; http://pastebin.com/FYu68DxD

i have corrected this changing in bootkernelline at boot/grub/menu.lst the acpi=off instead of 'on'

also in general i have added 'lapic' in the bootkernelline.

i hope i have done this right. after doing that the current pastebin is; http://pastebin.com/2aUSbfws

for the psu; the brandname is ASTEC type SA202-3545-2288A, 200W

today i have done several things on the computer and untill now no freezes.......................even when i had terminal constantly open with htop running in it.

regards,

Willem

[melodie]

Hello,

have installed the kernelversion 2.6.37.4 i686 on one hdd.

after i had installed this kernelversion i saw these kind of lines when startingup saying;

unable to locate IOAPIC for GSI...

you can see at pastebin here; http://pastebin.com/FYu68DxD

Hi,
You can see a detailed explanation here : http://forums.fedoraforum.org/showthread.php?p=1426830

I did a research on the web with your error message, found it, first at a Fedora French forum which led me there, which is detailed, and in English. As I can see no need to worry.

i have corrected this changing in bootkernelline at boot/grub/menu.lst the acpi=off instead of 'on'
also in general i have added 'lapic' in the bootkernelline.

Ok, as long as the system works better than it did...

i hope i have done this right. after doing that the current pastebin is; http://pastebin.com/2aUSbfws

You seem to have rebooted to the older kernel, and I see zram loaded. I am astonished, I thought I remembered it was another setting which had to be used while using that one kernel. Maybe was I wrong then... or maybe I don't remember well.

for the psu; the brandname is ASTEC type SA202-3545-2288A, 200W

Ok, let's see on the web what people say about it ?

today i have done several things on the computer and untill now no freezes.......................even when i had terminal constantly open with htop running in it.

regards,

Willem

This is encouraging news !

Regards,
Mélodie

[melodie]

Hi,

About the PSU:
http://forums.france-hardware.com/discussion/petite-annonce/pieces-detachees/sujet-25072.html

Posté le 01 juin 2003 - 17 h 44 m 45 s
   

Astec Modem SA202-3545-2288A
maximum output power is 200w
output :
+12v : 6A max
+5v 22A
+3.3v 14A
-5 0.3A
-12 0.3A
+5v SB 2A max

max power...

The price would be nowadays 7.95 €.

I continue to believe that 2 hard drives are too much for it.

[Muppet]

Hello,

thanks for the all the info again.

about the message 'unable to locate IOAPIC for GSI...' i have set currently the acpi back to 'on'. just to see how this goes.

i had seen that website at fedora aswell but found later on a site that gave suggestions what you could do to 'fix' it. options were trying; noapic, noacpi and acpi=off
the last one did not let see the message when startingup the computer, but have no idea what it does under the hood or that it is only making the message disappear from screen and anythingelse is left as it was.

adding ´ lapic 'in the bootkernel i think is a basic to do as it is otherwise suggested in the /var/log/message

for the moment i am doing manually turning off the zram when the program has started up. i had so many crashes and reinstalls behind me that i feel better doing it this way for the moment.
also i will try it when it is left to 'on ' to see how this goes.

to replace the current psu is perhaps something that will be in budget for this 'old 'computer.
since i have no idea where to look for, i came this across, could the 370w version a good replacement ? ( i know it should have 20pins connector for the motherboard and in the spec. it is saying that it is detachable to 20pin)
it is sold for around 30.00Euro
http://trust.com/products/group.aspx?grp=PSU

i would like to ask something about the grubmenu at startup;

i can make choices for;  linux
                                 linux-non fb
                                 failsafe
                                 2.6.38.8-pclos3.bfs
                                 2.6.37.4-pclos1

i have installed the text version off the grubmenu

should i go down and select there the kernelversion that i want to startup  and give enter or could i leave it on the first 'linux' and give enter there.

should i also try the kernelversion 2.6.37.4-pclos1-bfs ?

Willem

[melodie]

Hi,

I told you I think that in regard of that machine having a bfs kernel or not is prabably not very important. I suppose if your kernel is non bfs, and the machine works ok with it, leave it.

The PSU you are showing has:
4 peripheral, 1 floppy drive and 2 S-ATA connectors

Can it connect to IDE hard drives as well ? I suppose your hard drives are of the IDE type ?

about the message 'unable to locate IOAPIC for GSI...' i have set currently the acpi back to 'on'. just to see how this goes.

IO usually means Input Output, and apic is not the same thing as acpi.

APIC : http://en.wikipedia.org/wiki/Advanced_Programmable_Interrupt_Controller

ACPI : http://en.wikipedia.org/wiki/Advanced_Configuration_and_Power_Interface

LAPIC : Local APIC : http://en.wikipedia.org/wiki/LAPIC#Local_APIC

just to give you an idea... I have not gone very deep in these definitions, just enough to know they must not be confused, as having each a different, specific function.

i can make choices for;  linux
                                 linux-non fb
                                 failsafe
                                 2.6.38.8-pclos3.bfs
                                 2.6.37.4-pclos1

If you have booted to 2.6.37.4-pclos1 once, then the first default choice boots to 2.6.37.4-pclos1 each time after. You can check what kernel you are running with the command "uname -a", or "uname -r" in a console as simple user.

i had seen that website at fedora aswell but found later on a site that gave suggestions what you could do to 'fix' it. options were trying; noapic, noacpi and acpi=off

noacpi and acpi=off means the same thing.

I tend to think that if the people at Fedora consider the message as harmless, (and in that one case, the machine of the guy who was posting was working well in spite of the message), then the message is in fact harmless. Whereas the options you found on another place are exactly the one I gave you to try at the beginning, (almost : it was noapic nolapic... ) which didn't help stopping the freezes.

What seems to have calmed them down so far is removing one hard drive, and booting to an older kernel. What happens with zram : I am not sure at all that zram would bother in any way. Testing with and without, keeping the same 2.6.37.4-pclos1 kernel and running with just one hard drive plugged in, could be interesting, because the purpose of having the zram activated is enhancing the potential memory. You can use your applications as if you had a bit more RAM, and this makes the use more comfortable on machines having low memory, embedded systems, and so on...

[Just17]

i can make choices for;  linux
                                 linux-non fb
                                 failsafe
                                 2.6.38.8-pclos3.bfs
                                 2.6.37.4-pclos1

i have installed the text version off the grubmenu

should i go down and select there the kernelversion that i want to startup  and give enter or could i leave it on the first 'linux' and give enter there.

The first entry boots the last kernel you installed.
If you wish to boot a different kernel then select it from the list.

So according to the above the first and last entries will boot the .37.4 kernel.

[Muppet]

Hi Melodie and Just18,

thank you for the answers.

about the psu i have not yet the info if the psu i referred to does IDE drives. i will look into this.

i had the idea about the nvidia card to try the 'nouveau' driver from synaptic with this kernel and was very curious of this would now work with this current 'older'kernelversion.

i have made a new pastebin to look into but i think it looks very promising. little error messages for this driversetting as i can see.........

link pastebin; http://pastebin.com/1GLsSaid

http://pastebin.com/egj20fpJ

if it is stable now than i am very curious what could have made it worse in the latest kernel 2.6.38.8..........

regards,

Willem

Edit;

bad, bad Muppet idea to try 'nouveau' driver. also unstable in this kernelversion for my computer anyway.
have switced back for now to the 'nv' driver.

Edit;Edit;

i am testing this week. so not much of muppet questions these days. i come back later with status update