nacho 

a brave new world

i got my first job little more than 1 and a half month, and i'm really happy with it, although the salary is not so great, i get to play with linux and solaris and i only work 25 hours per week so i still have time to study and everything. it's impressive what you can learn when youre asked to do something and you lack the knowledge to complete the task, and internet is a database that was unthinkable 15 years ago. And with google all you have to do is ask the right question.

yesterday, i installed a new nevada build at home, nv42a and i'm quite impressed, gnome 2.14 is really cool. Of course i had problems, the library that takes care of address resolution was broken, thankfully the people at #opensolaris in freenode pointed me to a workaround; everything is working nicelly now
the new nv42 also comes with deer park (firefox pre 2) with the flash and realplayer plugins installed, obviously it is not as stable as the good old gnome 2.6 but it is definitely much nicer. I also love the new wallpapers and splashscreen

well, i'm off to read about solaris projects, this os has so many features that after using it for a few months already as my desktop i feel i'm just scratching the surface.

and one year went by

Opensolaris is one year old today, here we are, 365 days after sun released most of the solaris source code to the community
the days where people said opensolaris would fade away, that it was impossible for sun to create a community around it are over, people that predicted a dark future for the os have been slowly disappearing and are now being replaced by some who actually believe there is something in the road ahead.
i think there is no one true way to measure the success of the project but rather many little and no so little details:
The project generated so much interest that there are 14000 people subscribed to the opensolaris community, including myself. I have to get more involved in the Contents project and i promise i will as soon as i have some free time, i look forward to writing some more documentation
There are at least 5 opensolaris distributions out there (that was unthinkable 2 years ago)
people started contributing code to opensolaris, 100 put backs already
and many other things I will not mention mainly because other people have
I do want to highlight something, people are now talking about solaris, a world that was previously unknown for many people like me has opened its doors and is free for all, open for all.
2 years ago, solaris was only a dream for me
thank you, thanks to all the people involved in the community, to all those who made this thing happen

Updates

with the university and everything I never seem to find time to write anything :(
I'm happy to write that the mozilla foundation have donated 10,000 us dollars to the OpenBSD project, so as much as i hated the OpenBSD guys went public about their financial problems, it seems to have worked
Also, it seems like I will have to buy a new book earlier than expected this year, hopefully, Solaris Internals will be released soon: the bad news is that with the rapid developement of Opensolaris these days, some of the info available in the book will be out of date. I would love to see this book released under Creative Commons as documentation for the community but I also know that the authors worked really really hhard to get this book done and they deserve to profit with the sales, i just hope the book is not expensive. Lots and lots of great things are happening in the Opensolaris community
On a side note, a guy in a mailing list told me solaris is 10 years behind linux, i didnt know if I should start laughing or answer, so i did a bit of both, my answer just said, "do you know what you're talking about?" Solaris is certainly lagging behind in some areas, specially drivers and sound but is far ahead from linux in API stability and kernel debugging among other things. Check this for example and you will know what I mean. After the same guy replied telling me he knew what he was talking about, I decided I would not respond since the thread had turned into a flamewar already (about Slackware's merits)

varias cosas

Esta semana lei un mail acerca de los problemas financieros que estan afectando a OpenBSD, a pesar de que no uso OpenBSD, se por referencias que es un gran sistema operativo; ademas, la fundacion OpenBSD no se encarga solamente del sistema operativo del mismo nombre, ellos tambien son los responsables del OpenSSH y seria muy triste que algo le pasara.

Ademas probablemente el mas grande installfest del mundo va a ocurrir esta semana por segunda vez, se llama flisol y basicamente ocurre en varias ciudades de latinoamerica, lo bonito del caso es que no solo se instala linux, sino que tambien instalan BSDs y tambien software libre para Windows y MacOS

Finalmente, ya se puede preordenar Slackware 11!!

my birthday

I was born about 22 years ago in Quilmes,Buenos Aires, Argentina, seems like God really hated the world back then ;)

And the university starts again....

Summer is over boys and girls, well, not technically but clases have started so no more vacations at least till next year. The only thing I regret is not having updated my vmware translation, bleme it on the ZFS guys for making such a great file system ;)
I'm off to class again

more ZFS

Yesterday I got really excited with ZFS, I really wanted to know how the self healing capabilities of the damn thing worked. I found an old 3 gb disk, not huge but big enough for my purposes; it was also the first time I added a second disk to a solaris install, first created an empty configure file then turned the PC off and connected the disk, my main disk was c0d1 and the new one is c0d0.
I will repeat the process and paste most of the output and explanations or well... what I understood
root@darkstar:/export # format
Searching for disks...done


AVAILABLE DISK SELECTIONS:
0. c0d0
/pci@0,0/pci-ide@f/ide@0/cmdk@0,0
1. c0d1
/pci@0,0/pci-ide@f/ide@0/cmdk@1,0
Specify disk (enter its number): 0
[..] /*useless output*/
format> fdisk
Total disk size is 1530 cylinders
Cylinder size is 4096 (512 byte) blocks

Cylinders
Partition Status Type Start End Length %
========= ====== ============ ===== === ====== ===




WARNING: no partitions are defined!

SELECT ONE OF THE FOLLOWING:
1. Create a partition
2. Specify the active partition
3. Delete a partition
4. Change between Solaris and Solaris2 Partition IDs
5. Exit (update disk configuration and exit)
6. Cancel (exit without updating disk configuration)
Enter Selection:
I pressed 1 to create a new solaris2 type partition, this one will use the whole disk and will be the active partition (I'm not going to boot from it so it's not important anyway)
Partition Status Type Start End Length %
========= ====== ============ ===== === ====== ===
1 Active Solaris2 1 1529 1529 100
well, I updated the configuration and left (pressing 5)
the next step is creating slices
format> partition
[..]/*useless output*/
partition> modify
Select partitioning base:
0. Current partition table (original)
1. All Free Hog
Choose base (enter number) [0]?1
Time to explain what I think I understood, slices are like partitions inside a solaris2 type partiton and the configuration I selected makes a huge slice taking all of the disk. Now this slice is only usefull when configuring other slices since they will borrow or lend space to it, for example, when I create c0d0s0, the space will be borrowed from c0d0s6, my free hog slice, the same will happen with c0d0s1 and c0d0s3, c0d0s2 represents the whole disk, that's why it occupies 2.98 gb

Part Tag Flag Cylinders Size Blocks
0 root wm 0 0 (0/0/0) 0
1 swap wu 0 0 (0/0/0) 0
2 backup wu 0 - 1526 2.98GB (1527/0/0) 6254592
3 unassigned wm 0 0 (0/0/0) 0
4 unassigned wm 0 0 (0/0/0) 0
5 unassigned wm 0 0 (0/0/0) 0
6 usr wm 0 0 (0/0/0) 0
7 unassigned wm 0 0 (0/0/0) 0
8 boot wu 0 - 0 2.00MB (1/0/0) 4096
9 alternates wm 1 - 2 4.00MB (2/0/0) 8192

Do you wish to continue creating a new partition
table based on above table[yes]? yes
Free Hog partition[6]?
Enter size of partition '0' [0b, 0c, 0.00mb, 0.00gb]: 1.4gb
Enter size of partition '1' [0b, 0c, 0.00mb, 0.00gb]: 1.4gb
Enter size of partition '3' [0b, 0c, 0.00mb, 0.00gb]: 180m
Warning: no space available for '4' from Free Hog partition
Warning: no space available for '5' from Free Hog partition
Warning: no space available for '7' from Free Hog partition

Part Tag Flag Cylinders Size Blocks
0 root wm 3 - 719 1.40GB (717/0/0) 2936832
1 swap wu 720 - 1436 1.40GB (717/0/0) 2936832
2 backup wu 0 - 1526 2.98GB (1527/0/0) 6254592
3 unassigned wm 1437 - 1526 180.00MB (90/0/0) 368640
4 unassigned wm 0 0 (0/0/0) 0
5 unassigned wm 0 0 (0/0/0) 0
6 usr wm 0 0 (0/0/0) 0
7 unassigned wm 0 0 (0/0/0) 0
8 boot wu 0 - 0 2.00MB (1/0/0) 4096
9 alternates wm 1 - 2 4.00MB (2/0/0) 8192

Okay to make this the current partition table[yes]?
Enter table name (remember quotes): "test"

Ready to label disk, continue? yes
partition> q
[..] /*useless output, it's the format menu actually*/
format> q
we're done creating partitions and slices, now, let's create our ZFS mirror like we did last time, but now with a real disk. and moint it in /export/test, I will turn compression on too
root@darkstar:/export # zpool create test mirror c0d0s0 c0d0s1
root@darkstar:/export # zfs set compression=on test
root@darkstar:/export # zfs set mountpoint=/export/test test
root@darkstar:/export # zpool status
pool: test
state: ONLINE
scrub: none requested
config:

NAME STATE READ WRITE CKSUM
test ONLINE 0 0 0
mirror ONLINE 0 0 0
c0d0s0 ONLINE 0 0 0
c0d0s1 ONLINE 0 0 0
everything seems ok right now, I will copy a small file into my new ZFS filesystem nad get the md5 checksum of it
root@darkstar:/export # cp home/nacho/NVIDIA-Solaris-x86-1.0-8178.run test
root@darkstar:/export/test # digest -a md5 NVIDIA-Solaris-x86-1.0-8178.run
10c346b1dcfb5d3ee9312c1ce1c6bd24
now I will test the ZFS self healing capabilities the same way it was done in one of the ZFS introductory videos but using /dev/zero instead of /dev/random, I really just want to corrupt data and I dont really care if it is with random data or pain zeros
root@darkstar:/export/test # dd if=/dev/zero of=/dev/rdsk/c0d0s0 bs=1024 count=20480
20480+0 records in
20480+0 records out
20971520 bytes transferred in 13.288751 seconds (1578141 bytes/sec)
root@darkstar:/export # zpool export test
root@darkstar:/export # zpool import test
ZFS uses a cache, needed it to flush it and read the data from the disk to notice any changes, hence I exported and imported my ZFS pool
root@darkstar:/export # digest -a md5 test/NVIDIA-Solaris-x86-1.0-8178.run
10c346b1dcfb5d3ee9312c1ce1c6bd24
sounds like the data is still ok, let's check the status of my pool
root@darkstar:/export # zpool status
pool: test
state: ONLINE
status: One or more devices has experienced an unrecoverable error. An
attempt was made to correct the error. Applications are unaffected.
action: Determine if the device needs to be replaced, and clear the errors
using 'zpool online' or replace the device with 'zpool replace'.
see: http://www.sun.com/msg/ZFS-8000-9P
scrub: none requested
config:

NAME STATE READ WRITE CKSUM
test ONLINE 0 0 0
mirror ONLINE 0 0 0
c0d0s0 ONLINE 0 0 1
c0d0s1 ONLINE 0 0 0
I really look forward to being able to boot from ZFS, it's just so easy to do these things. I think i will play some more with it

Trying ZFS

well, today was my first time with ZFS, since i dont have a biig disk available or any free slices I just used regular files, I used 200mb files in fact and here is how i created my first ZFS mirror
first, let's create the files
root@darkstar:/export # mkfile 200m testfs1
root@darkstar:/export # mkfile 200m testfs2
now let's create the pool and the mirror
zpool create test mirror /export/testfs1 /export/testfs2
finally let's change the mount point and turn comression on
root@darkstar:/export # zfs set compression=on test
root@darkstar:/export # zfs set mountpoint=/export/test test
root@darkstar:/export # zpool status
pool: test
state: ONLINE
scrub: none requested
config:

NAME STATE READ WRITE CKSUM
test ONLINE 0 0 0
mirror ONLINE 0 0 0
/export/testfs1 ONLINE 0 0 0
/export/testfs2 ONLINE 0 0 0
all looks good but let's check the value of the properties of the zfs pool
root@darkstar:/export # zfs get all test
NAME PROPERTY VALUE SOURCE
test type filesystem -
test creation Fri Mar 10 18:24 2006 -
test used 422K -
test available 179M -
test referenced 98.5K -
test compressratio 1.00x -
test mounted yes -
test quota none default
test reservation none default
test recordsize 128K default
test mountpoint /export/test local
test sharenfs off default
test checksum on default
test compression on local
test atime on default
test devices on default
test exec on default
test setuid on default
test readonly off default
test zoned off default
test snapdir visible default
test aclmode groupmask default
test aclinherit secure default

that was easy wasnt it? ;)
I will play more with it later
I will tell you about my failed experience using ZFS and lofi exported file, hint: it didn' result quite as expected, specially when i tried to cat /dev/random the lofi exported file

Linus y GPLv3

Era hora de que escribiera algo en español en el blog ¿no? ;)
hace poco salio publicada una nota acerca de Linus y su opinión acerca de el borrador de la nueva version de la licencia GPL (GPLv3) y debo admitir que me impresionó, especialmente despues de los últimos comentarios de su parte que hicieron ruido (acerca de Gnome).
Creo que esta vez Linus tiene razón; no me parece que las restricciones en las licencias sean la manera en la que se debe afrontar tecnologias como DRM, me parece que mucho mas importante es informar a los usuarios acerca de los derechos que están resignando al aceptar contenido que usa DRM y deberian ser estos los que al final decidan si estan dispuestos o no a pagar por ese contenido. La aceptación es lo que va a decidir el futuro de la tecnologia de la misma manera que fue la aceptación lo que decidió el futuro de la licencia GPLv2
Mas alla de las restricciones a las tecnologias como DRM que impondria la licencia si su borrador fera aprobado asi como esta, éste incluye algunas cosas interesantes sobre todo en lo que refiere a la compatibilidad con otras licencias que fueron aprobadas por OSF, aun asi creo que falta mucho para ver el resultado final en forma de licencia.
Volviendo al artículo, me pregunto si se podran conseguir los tiburones con lasers en la cabeza :)

nexenta

nexenta is a non-sun developed opensolaris distribution, it is basically debian around an opensolaris kernel and it is what I was testing today, I must say I'm overall impressed with it, except for a few minor details it's incredibly usable despite being alpha.
The installation went without a hickup, not many questions asked, It was a long process, luckily they provide you with a little tetris like game in another window so you can play while the install process does its job, nice touch
Incredibly most of my hardware was detected and working except for the sata on board controller, even the via rhine nic and the sound blaster cads were working (I had to install extra packages in my nv_28 to get those two working).
The only hardware problem I had was my nvidia card which wouldn't work despite having installed the nvidia drivers, the ones in their package repository are old and present a known issue with nevada builds older than 33 that will cause some problems with the frame buffer driver, nexenta is based around nv_34 I think; luckily that has already been fixed in a new version of their driver but the package repository has not been updated.
The other problem I had was X, seems like nexenta does not ship the regular xorgcfg, xorgconfig and scanpci and instead guesses the config via a getconfig perl script. it does not even need a /etc/X11/xorg.conf file but it'll use it if it's there, I had problems changing the default keyboard layout even using an xorg.conf file
The one thing that looked weird to me was the fact that while the desktop is a really good looking gnome, most of the programs are KDE programs, it's a really odd package selection but thanks to synaptic, the graphical package manager, it is easy to solve, and there are tons and tons of packages ready to download and install through it as well
All in all I really like what I saw, and I really look forward to seeing a more stable and user friendly nexenta for new users to try, good job guys
off to sleep now, this is probably full of typos