On 2014-08-13 18:58, Vagrant Cascadian wrote:
> On 2014-08-13, richard kweskin wrote:
>> On 2014-08-13 02:04, Vagrant Cascadian wrote:
>>> On 2014-08-12, richard kweskin wrote:
>
>>>> 5 edit /etc/ltsp/update-kernels.conf to have BOOT_METHODS=NBD,
>>>> ARCH="i386" and IPAPPEND=3 uncommented
>>>> 8 patched /usr/share/ltsp/update-kernels
>>>> 9 aptitude install pxelinux
>>> steps 5, 8 and 9, except in the chroot, followed by:
>>>
>>
>> Thank you once again for all your help. In your above remark
>> concerning
>> steps 5, 8 and 9 do you mean that these steps are to be done both on
>> the
>> server's setup and on the client's chroot or to be done only on the
>> client's chroot.
>
> Only in the client's chroot, for a setup using ltsp-build-client and
> a
> separate chroot, and all the client's files come from that chroot.
Ah, ok. I will eliminate some steps then.
> I think your earlier install was an amd64 using the ltsp-pnp method,
> which needs a largely different set of instructions, since all the
> client's files come from the server install.
It was with i386 and ltsp-pnp.
> In general, I would recommend using ltsp-pnp at this point for fat
> clients or localapps; it's a much simpler configuration, with only
> one
> place to edit all files.
>
> In my opinion, ltsp-pnp is probably the best innovation in LTSP since
> the switch to LTSP5.
I agree. However, there are cases in which the above method is used. I
am only testing different cases and because you have been so forthcoming
I can now write up some documentation both in Greek and in English.
>
> live well,
> vagrant
fyi : when this same box
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 2
On-line CPU(s) list: 0,1
Thread(s) per core: 2
Core(s) per socket: 1
Socket(s): 1
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 15
Model: 4
Stepping: 3
CPU MHz: 2800.000
BogoMIPS: 6400.16
L1d cache: 16K
L2 cache: 2048K
NUMA node0 CPU(s): 0,1
was creating the squashfs image in amd64 but for clients in i386 it was
using 2 processors and of course took less time, but in the earlier
configuration in i386 it used only one processor.
In contrast at several schools where I have gone to help them setup
they have had i3 or i5 cpu's and although we setup the server in i386 so
that ltsp-pnp was used the server utilized all 4 processors to create
the image as expected.
Is it possible that some hardware only has access to both processors in
64 bit mode where other hardware can also do it in i386? Is it the
choice of kernel? 3.14-2-amd64 now and 3.14-2-486 before?
Richard
------------------------------------------------------------------------------
_____________________________________________________________________
Ltsp-discuss mailing list. To un-subscribe, or change prefs, goto:
https://lists.sourceforge.net/lists/listinfo/ltsp-discuss
For additional LTSP help, try #ltsp channel on irc.freenode.net
