Puppy Linux Discussion Forum

[emil]

It works with win 7, I am pretty sure it also works with XP and vista.
I wouldn't be surprised if Win 2000 would work.

The Windows Virtual Box installer is for both 32/64 bit.
During installation it detects which version has to be installed.

I have posted a more detailed "howto create your own Virtual Box exe installer"
http://murga-linux.com/puppy/viewtopic.php?t=73921
It just came to my mind that a similar setup could be used like the venerable "qemu puppy".

[emil]

I am working on a new version which should have some improved functionality:
Download sagelive-511-53.iso

Changelog sagelive-511-53.iso
+ updated sage version to 5.3
+ added Gnutls (for server certificate)
+ added GNU less (replace busybox less, which misses a feature and breaks sage inline help in console mode)
+ added new Personalize settings (from shinobar, replace Countrywizard: cut down the number of pop-up-windows though)
+ updated and fixed system file permissions so this lupu511 version can handle users (in console mode) - still no multiuser puppy
+ included ssh Daemon, so you can log in over the net
+ included preconfigured sage server (with 20 Worker accounts - potential simultanious users over the network)
+ new theme and worked to sort the menus&Desktop (JWM)
- removed Abiword and some Puppy applications (some are still there but not in the menu like gparted)

I dropped Abiword because everytime I need a word processor I ended up installing Libre Office or something (I have an sfs mirrored here).

Most work has gone into the setup of the preconfigured server.
In principle it should be possible to start it and then connect from other computers to the running sage server (with the browser and the IP adress of the server on port 8080).

The Sage server runs as an unprivileged user. It is preconfigured to support a maximum number of 30 simultanious users (using "worker accounts"). For multi user setups it might be necessary to edit the file /home/sageadmin/start_server.sage.

passwords:
Default password for the sage notebook admin account is mathematic.
If you start the sage server it will prompt you for a new password.
Default system passwords are not published (not necessary, because there is autologin)
However it is recommended that you change passwords, especially before you start the ssh daemon (server).
1) open console (as root), type passwd
2) in console (as root) , type passwd sageadmin

hostname
If you missed to set an unique hostname during firstboot:
change hostname in file /etc/hostname (needs reboot)

change ssh keys
you might feel it appropriate to change keys from time to time.
There is an automated script: fix-ssh-keys.sh (run it from terminal)

[ICPUG]

Nearly two years on and this superb specalist puplet is still an active project.

Do you have any feedback from the Sage community, Emil? Is Sage Live used a lot?

[emil]

Hi ICPUG,

thanks for feedback from an old time follower of this thread.

I have access to the google analytics page of the sage mirrors and there were around 500 downloads of the live cd last month.

I had responses from one university course in Austria, where the Live CD was used as "last stop solution", when students failed to get sage installed otherways.

That was also the reason why I hid some system tools in the current release (like gparted), because not every teacher and sysadmin mightbe happy with handing out those tools. still you can start them from commandline. And it is not the primary scope of this pupplet.

Sage Live CD was used even more, especially there was a time when no good Windows Solution was available for sage. There was a lot of struggle with a port to Cygwin, but ultimatly it was settled that a virtual machine is good enough. With the server you can work in the browser of Windows the same as nativly in Windows. And for programming you can use a ssh shell and work in in the vm directly. Speed loss is not an issue (around 5%). The official sage VM is now based on Fedora and is rather big. I took part in that discussion and I think that is the right solution for sage on Windows (or other exotic OSes).

I am happy that I can manage to sustain the support for this pupplet, maybe I will have also a release next year with an up to date sage version. However probably I will have to update my heavily customized Lupu 511 base then, if time allows ...

I would be very happy to have some feedback if the sage server works - I have not really a big enough LAN to test that properly. It would be even possible to attach the Iso to VirtualBox or so (maybe it is necessary to change network settings to "bridged adapter")

kind regards ,
emil

known bugs:
menu entry for mg (editor) fails. works from comandline

[saintless]

Hi, Emil,

just some information for people with old computers like mine
(Tested on Gateway E-1400, 600 Mhz, 256Mb RAM).
If you can't start sage console from the start menu or from the desktop icon and when you type sage in urxtv you get this message:

[rasul]

Hi emil and thanks for your great job, I have used your previous sagelive and now I am managing to use your new one for my dissertation, 50 MB usage of RAM is very fantastic! However, I have also a question. I'm going to run my code on a quad-core machine with 18 GB of RAM, fortunately SageLive supports multicore but does it support also 18 GB of RAM? If not then do you have any suggestion? My code is so heavy that it failed on 4GB Mac Machine, now I want to use my professor's machine and I want to make sure it will work before to tell him I want to work on his machine.

Thanks again.


PS. your previous Sagelive was opening sage notebook directly but the new SageLive first starts a terminal then opens the notebook, is there anyway to bypass the terminal opening in the new sage as well?

[emil]

Hi thanks for feedback,

Saintless, the problem with the old instruction set "should" not happen. can you confirm that sage is running fine (for example run all the included tests with sage --testall.

I compiled sage with the optional flag "SAGE_FAT_BINARIES". This should avoid using the sse2 instruction set. If this error appears I have to look into it (but I have no old machine to test, and ... hrmm, no time). If the tests are fine, maybe there is just a bug throwing the error message.

Hi rasul - thank you for your kind words, it is a honor that you can work with sagelive on your thesis.
I am sorry, but sagelive can not use your modern hardware to full extend, because it is a real 32 bit system and it also uses the original lupu 511'kernel which does not have the PAE feature.
As mentioned above, I see the target more in the low end segment.

In any case to make full use of the hardware of your professor you should install a 64 bit Linux system of your choice and then compile sage natively there. Especially the optimized numeric libaries will run much faster if they can be optimized for multicore and heaps of RAM (e.g. ATLAS library). Compiling sage from source should be mostly painless because they have really good build script and include most of the necessary libraries.

Which OS is already installed on this machine?

about the autostart feature of sage and the terminal - this has escaped me. I thought I had copied the startscript "as is" from the previous version. Again I have to look into it, though it might need some time because I am not "into" it at the moment.

kind regards
emil

all the best
emil

[saintless]

[emil]

Hi,

the command would be from the terminal.

[rasul]

Hi emil,

Thanks for the info, it's a pitty i cannot use your distro on that Machine, it's extremely light and as the name suggests your distro can be looked at as only the very sage program in live, too much that the footprint of the distro on the system is small and negligible!

Also thanks for your suggestion, I will miss your distro then for sometimes although i will hopefully follow your great jobs.

Best regards

[saintless]

[emil]

Hi Toni-

yep, we can ignore the timeout. However if you leave the testing running it should go on for endless time (about 3 hours on my machine-if it is a very slow computer it coul as well take overnight to run all tests. If there is a real problem with the instruction set, then there should be some wrong results. If it is no problem for you I kindly ask you to try this, The sage folks release new versions in a frenzy and sometimes there are regressions which are not immediately covered.

One of the new features I used in this build was a new compile flag SAGE_ATLAS=base. I am not completly sure how this interacts with SAGE_FAT_BINARIES=yes and as I mentioned I cannot test this myself. cheers emil

[saintless]

Hi, Emil,
No problem at all. I will wait for sage testing to the end and post the result in text file attached to my next post.
Cheers, Toni

[saintless]

Hi, Emil,
here is the errors report::

[emil]

The Killed/crashed errors don't look well, I will post something on the sage channels about the compile flags and maybe I get some explanation. In the meanwhile everybody with a very old chipset should use one of the earlier versions, where the "Fat Binaries" flag is correctly used.