Virtualization hardware - was (Re: Recommendation for software/hardware for virtualization)

2007-04-17 Thread Tzahi Fadida
Specifically regarding available hardware.
What hardware (CPU - INTEL/AMD) today comes with virtualization features and 
is most compatible to vmware/xen/etc...?
Can someone recommend a good spec for that?
How is xen compared to vmware (ESX?) performance wise and utilizing special 
hardware features? What about graphics, i recall they had problems utilizing 
performance in the past for windows, is it solved?
10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Virtualization hardware - was (Re: Recommendation for software/hardware for virtualization)

2007-04-17 Thread Tzahi Fadida
On Tuesday 17 April 2007 16:33:31 Jacob Broido wrote:
 Hi,

 CPU's: I assume you're refering to VT-x(Intel) and Pacifica(AMD) cpu
 extentions. Well, you can find a list of CPU models supporting these
 extentions here: http://wiki.xensource.com/xenwiki/IntelVT  and AMD: Athlon
 64 F* , All current Opterons

 Having said that, using hardware aided vm extentions does not neceserly
 boost one's performance (in many cases it worsens it), what it does allow
 though, is running unmodified Guest OS's under a hypervisor(i.e windows
 under xen).

 As for xen VS vmware(ESX) , there are many mixed benchmarks(vmware
 prohibits publishing benchmark results of its products) around the compared
 perfomance of these two, so I wont go directly into that. Having said that
 Vmware seems to be  a more mature and polished product. Keep in mind that
 ESX is not a freeware and quite expensive , moreover it must have SAN or
 iSCSI as its storage.

 All in all, regarding specs it really depends on the type of load you're
 planning to do. But one thing is for sure, you can never have too much RAM
 and double importance  goes for VM's.

Is it worth using xen/vmware esx for a development workstation?
Are the VT extentions are in effect when using the host/guests version of 
vmware (i.e. not esx). Is it fair to say that intel VT features are supported 
and more stable in xen and vmware products?
10x.

 On 4/17/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  Specifically regarding available hardware.
  What hardware (CPU - INTEL/AMD) today comes with virtualization features
  and
  is most compatible to vmware/xen/etc...?
  Can someone recommend a good spec for that?
  How is xen compared to vmware (ESX?) performance wise and utilizing
  special
  hardware features? What about graphics, i recall they had problems
  utilizing
  performance in the past for windows, is it solved?
  10x.
 
  --
  Regards,
  Tzahi.
  --
  Tzahi Fadida
  Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
  WARNING TO SPAMMERS: see at
  http://members.lycos.co.uk/my2nis/spamwarning.html
 
  To unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]



-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Recommendation for software/hardware for virtualization

2007-04-16 Thread Tzahi Fadida
Hi,
As part of my new job, i am going to develop also windows kernel/drivers.
Therefore i will have to (naturally) run windows, however, i got comfortable 
with linux desktop and i am reluctant to move back to windows as my main 
desktop environment.
What are my options, considering i am willing to buy ANY (within reason) new 
hardware to accommodate the possibility of running the two OSs (linux and 
windows) in tandem without interfering with the work flow.
What is the most optimal solution money can buy software wise.
10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Recommendation for software/hardware for virtualization

2007-04-16 Thread Tzahi Fadida
On Monday 16 April 2007 22:16:00 Oleg Goldshmidt wrote:
 5) A virtual machine such as VMware. You will need a decent computer,
enough memory, etc., but the requirements are modest by today's
standards. I run Linux on a T43 Thinkpad and there are some things
(being nice to co-workers mainly) that I need to do in Windows, so
I have an XP in a VMware Player, allocated 368MB of RAM out of the
total GB to it, and it works just fine. Depending on what kernel
work you will be doing, you may need to run Windows on real HW and
Linux in a VM, which may be less than absolutely perfect (but
probably decent) if you spend the vast majority of your time in the
Linux desktop.

 The above assumes that we are talking about desktop computers and you
 will be working at your desk.

There are 2 workplaces i am looking at. 
At work i am getting 1 computer with who knows what on it. 99% it is windows. 
It would be stupid to develop drivers on your main OS, thus i am guessing 
vmware would be the other solution there anyway, so i will also run linux on 
a separate vmware session or run cygwin solutions. However, i got the feeling 
it won't play nice if i have 1 main OS + 2 guests at the same time. As a 
kernel developer, however, i may get a new computer with those new CPUs that 
can handle VT. Do you think they will be able to handle 2 guests?

The other place is at home which is here i am referring to the hw solution. 
Here the host os will obviously remain Linux and thus, the virtualized OS 
would be windows. I guess VMWARE here too? What about XEN? I hear that there 
are CPUs which are better at virtualization , what should i purchase? 
currently my computer won't be able to handle another OS since it is p1.6.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Recommendation for software/hardware for virtualization

2007-04-16 Thread Tzahi Fadida
On Tuesday 17 April 2007 01:29:44 guy keren wrote:
 Tzahi Fadida wrote:
  On Monday 16 April 2007 22:16:00 Oleg Goldshmidt wrote:
  5) A virtual machine such as VMware. You will need a decent computer,
 enough memory, etc., but the requirements are modest by today's
 standards. I run Linux on a T43 Thinkpad and there are some things
 (being nice to co-workers mainly) that I need to do in Windows, so
 I have an XP in a VMware Player, allocated 368MB of RAM out of the
 total GB to it, and it works just fine. Depending on what kernel
 work you will be doing, you may need to run Windows on real HW and
 Linux in a VM, which may be less than absolutely perfect (but
 probably decent) if you spend the vast majority of your time in the
 Linux desktop.
 
  The above assumes that we are talking about desktop computers and you
  will be working at your desk.
 
  There are 2 workplaces i am looking at.
  At work i am getting 1 computer with who knows what on it. 99% it is
  windows. It would be stupid to develop drivers on your main OS, thus i am
  guessing vmware would be the other solution there anyway, so i will also
  run linux on a separate vmware session or run cygwin solutions. However,
  i got the feeling it won't play nice if i have 1 main OS + 2 guests at
  the same time. As a kernel developer, however, i may get a new computer
  with those new CPUs that can handle VT. Do you think they will be able to
  handle 2 guests?
 
  The other place is at home which is here i am referring to the hw
  solution. Here the host os will obviously remain Linux and thus, the
  virtualized OS would be windows. I guess VMWARE here too? What about XEN?
  I hear that there are CPUs which are better at virtualization , what
  should i purchase? currently my computer won't be able to handle another
  OS since it is p1.6.

 normally, when developing drivers for windows, especially if they are
 hardware drivers - you can't do that on a guest OS - you need to do this
 on an OS running directly on the hardware. this is because the guest
 only sees virtualized hardware - and your company's hardware is not
 supported by the virtualization engine ;)

No hardware. It is all software (security etc...).

 thus, in most driver-related work places, you will get 2 computers
 anyway - one on which to develop the code and compile it. one on which
 to run the driver (and crash it every once in a while). thus, you will
 be able to run the virtualization system on the first PC. note that in
 windows, you often connect the debugger, from the development machine,
 via a serial (COM) cable to the target machine. thus, if you'll run
 windows as a guest - it'll need access to the COM port - you'll have to
 figure out how to configure it, and be ready to handle problems on your
 own - no one in the company will help you with it - unless there's
 another linux freak there ;)

Well, the product is cross-platforms and we have a few linux developers there. 
Hopefully, i'll get some help with that if necessary.


-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Help with sound card

2007-04-12 Thread Tzahi Fadida
Just FYI,
It is a known issue that many times skype will not work without Microphone 
capture on.

On Thursday 12 April 2007 01:00:31 Orr Dunkelman wrote:
 Long story short: Laptop, sound card, microphone doesn't work. Windows -
 work.

 Laptop: Compaq Evo N800V (four years old)
 FC6. Standard. Kernel - 2.6.20.4 - manually compiled.

 modules (courtesy of lsmod):

 snd_intel8x0   31644  4
 snd_ac97_codec 89508  1 snd_intel8x0
 ac97_bus2560  1 snd_ac97_codec
 snd_seq_dummy   3972  0
 snd_seq_oss29312  0
 snd_seq_midi_event  7296  1 snd_seq_oss
 snd_seq45808  5
 snd_seq_dummy,snd_seq_oss,snd_seq_midi_event snd_seq_device  7948 
 3 snd_seq_dummy,snd_seq_oss,snd_seq
 snd_pcm_oss39552  0
 snd_mixer_oss  15616  3 snd_pcm_oss
 parport_pc 26020  1
 snd_pcm71172  3 snd_intel8x0,snd_ac97_codec,snd_pcm_oss
 snd_timer  21252  2 snd_seq,snd_pcm
 pcmcia_core39716  3 pcmcia,yenta_socket,rsrc_nonstatic
 snd49668  13
 snd_intel8x0,snd_ac97_codec,snd_seq_oss,snd_seq,snd_seq_device,snd_pcm_oss,
snd_mixer_oss,snd_pcm,snd_timer soundcore   7776  3 snd
 snd_page_alloc  9992  2 snd_intel8x0,snd_pcm
 pcspkr  3328  0

 playes with alsamixer.

 results:

 when microphone is on, switch: Microphone Capture, I hear from the speakers
 whatever I say into the mic.
 when it's off - nothing.

 Tried all combinations of
 Line-in capture, microphone capture, +20 db Boost, and IEC 958.
 Tried both mic1 and mic2 for inputs (surprisingly only mic1 works).

 No other sound device (especially not a mic).

 Permissions:
 /dev/dsp and /dev/snd/*
 were user.root and user.user, permission 600.
 none worked.

 anyone with some guess why this happens? google tells me a lot about two
 sound cards, but I have only one, and alsamixer does not solve the problem.

 Thanks!



-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Performance monitoring for selected process

2007-04-04 Thread Tzahi Fadida
On Wednesday 04 April 2007 18:59:17 Maxim Veksler wrote:
 Hi,

 Except from stracing -f executables is there some way I can monitor
 the process for performance? I would like to debug process delayed
 response activity and need to know when it's doing heavy IO / when
 it's CPU intensive and when it's all too busy waiting for IRQ.

 A graphical display would be preferred, textual will do as well obviously.

 For general system statistics, I've tried the following :
 1. sysstat + kSar
 2. gnome-system-monitor
 3. ksysguard
 4. ntop

 All work great but are too general for my needs, I'm looking for tools
 that could display single process statistics.

Regarding per I/O stats:
I don't believe there is a tool that can help you with that. However, there 
are ways to mitigate that. For example:
1. Bring the system to a minimal I/O work and minimal applications in memory. 
2. Turn off swap.
3. Run applications and use top, etc... apps to view changes.
also use: vmstat, cat /proc/interrupts, cat /proc/stat for interrupts usage if 
you know the interrupt your respective driver (ur app uses) uses.
Also you can enable echo 1 /proc/sys/vm/block_dump and tail /proc/kmsg to 
see application disk usage.

More than that i can indicate for you of a way to use kprobe/jprobe to get a 
bit more stat accuracy but you will have to create a module (and maybe also 
the kernel but i am not sure).

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: linux-il moderation and spam filtering

2007-03-25 Thread Tzahi Fadida
I think that there should be a way, like using a signing key signature that 
you register with the list in advance.
Or simply using a text signature to enhance the accuracy of identifying you 
correctly. I suggest you start adding a distinct signature to your emails and 
wait a while for the spam filters to get this new info into the system.
Of course, this is only a guess as to why they are not getting it that they 
emails are from you.

On Sunday 25 March 2007 09:40, Peter wrote:
 Hello all,

 last week the spam filters on the list ate two of my messages. They were
 large messages and I put a lot of effort into them. In the past few
 weeks this has happened a few times. Obviously I have been wasting my
 time writing them. Subsequent reporting of a message by an admin did not
 make it appear.

 The time has come to ask the question whether ilnux-il admins will
 consider giving up spam filtering the mailing list. I am subscribed to a
 number of mailing lists and none filter or moderate postings without
 explicitly saying so and especially without offering recourse. linux-il
 is the only one.

 I consider that subscribed and verified (at least by email
 confirmation and captcha) members of a mailing list can be trusted to
 not flood the list with spam (which would result in their being banned).
 Electronic chaperones are not welcome. Especially chaperones which have
 proven that they will cause data loss as they did for me.

 I would like to have a clear answer about why the linux-il list has a
 spam filter although it is a closed list as any other, who decided o
 install it, and whether removing (not just for me) it is an option that
 will be considered. If the answer is no, I may consider unsubscribing.
 There is no point in posting to a Russian Roulette list that may delete
 those postings I put most effort into (and has done so repeatedly so
 far). I can follow the list from an archive without wasting any time
 composing messages that will be deleted.

 I am awaiting response,

 thanks,
 Peter

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: list police ate my message

2007-03-25 Thread Tzahi Fadida
On Sunday 25 March 2007 12:08, Dotan Cohen wrote:
 On 24/03/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  Perhaps there is a way to register a key to the list engine and you can
  sign your email with the key. Then, the list engine will be sure to know
  it is you. Everyone else and your not signed emails will have to pass the
  spam filters.

 It's called pgp:
 http://en.wikipedia.org/wiki/Pretty_Good_Privacy

I know pgp, however, it is not enough that you sign it, you have to tell the 
list to register you somehow for later processing otherwise your email will 
be considered like any other.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: linux-il moderation and spam filtering

2007-03-25 Thread Tzahi Fadida
On Sunday 25 March 2007 12:01, Peter wrote:
 On Sun, 25 Mar 2007, Tzahi Fadida wrote:
  I think that there should be a way, like using a signing key signature
  that you register with the list in advance. Or simply using a text
  signature to enhance the accuracy of identifying you correctly. I
  suggest you start adding a distinct signature to your emails and wait
  a while for the spam filters to get this new info into the system. Of
  course, this is only a guess as to why they are not getting it that
  they emails are from you.

 Mailing list members are members because they signed in and were
 confirmed. There is no need for furhter id. That's the way it works all
 over the world.

You yourself pointed out in an earlier emails some insights about x86 work all 
over the world, that doesn't make it the right way :).
Anyway, your header can be hijacked by anyone and they can impersonate you, 
this is SOP for spammers. You have to make more effort into proving who you 
are in order for the list to trust you more, thus, i suggested to add some 
kind of signature to improve.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: linux-il moderation and spam filtering

2007-03-25 Thread Tzahi Fadida
On Sunday 25 March 2007 13:51, Peter wrote:
 On Sun, 25 Mar 2007, Tzahi Fadida wrote:
  Mailing list members are members because they signed in and were
  confirmed. There is no need for furhter id. That's the way it works all
  over the world.
 
  You yourself pointed out in an earlier emails some insights about x86
  work all over the world, that doesn't make it the right way :).
  Anyway, your header can be hijacked by anyone and they can impersonate
  you, this is SOP for spammers. You have to make more effort into proving
  who you are in order for the list to trust you more, thus, i suggested to
  add some kind of signature to improve.

 The way these things work is, 'don't fix it if it ain't broken'. While
 there is a way to abuse the list, it is unlikely to be used. Other lists
 with larger circulation are better targets for spam and the process is
 still not (or very seldomly) used. Thus using it 'preventively' is in

Security in obscurity. You know, i think most of the unsubcribers to this 
list, did so because of a high volume of non-relevant emails (to them of 
course), imagine how a few spam messages could affect others.

 the same class as requiring biometric authentication for an entire
 country's adult population to allow use of the internet, so as to
 prevent a potential teenager from potentially accessing a potentially
 damaging website once or twice (usually it's disgusting enough the first
 time).

Ok, now you are exaggerating. Aside from the scales and completely 
inappropriate censorship attempts which is not the case here, i did not 
suggest blocking people who did not provide a signature but rather not run 
the spam filters on them. You can still send unsigned emails.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: list police ate my message

2007-03-24 Thread Tzahi Fadida
Perhaps there is a way to register a key to the list engine and you can sign 
your email with the key. Then, the list engine will be sure to know it is 
you. Everyone else and your not signed emails will have to pass the spam 
filters.

On Saturday 24 March 2007 22:15, Dotan Cohen wrote:
 On 24/03/07, Dotan Cohen [EMAIL PROTECTED] wrote:
   Would it be possible to whitelist addresses that have X posts? I'd
   imagine any X larger than 10 should suffice.
 
  I like this idea.
 
  (silly test)
 
  --
  Tzafrir Cohen | [EMAIL PROTECTED] | VIM is
  http://tzafrir.org.il || a Mutt's
  [EMAIL PROTECTED] ||  best
  ICQ# 16849755 || friend
  t

 Ah, understood. I did know that it is so easy to forge From addresses.
 I just did not think of it.

 Dotan Cohen

 http://lyricslist.com/
 http://what-is-what.com/

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Driver for Infrared Device

2007-03-24 Thread Tzahi Fadida
I got an ancient infrared device
PACKARED BELL - BPCS 146542 FCC ID: FODFMIR.
It has a serial connector (DB9 pins).
I don't have a remote for it.

Is there a driver in linux that will support it? 
Also, is there a program to record a tv remote buttons and use it to operate, 
for example, amarok, etc...
10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Driver for Infrared Device

2007-03-24 Thread Tzahi Fadida
I am trying to install lirc, however, i am having trouble compiling the 
modules.
I have kubuntu edgy. I installed the linux-source package which is the same 
version of the kernel i am using. I copied the debian/config to .config and 
make oldconfig  make modules

then i downloaded from sourceforge lirc 0.8.1 and ./configure
left the default serial and let it run the configure.
Then, make  make install.
However, when i try to modprobe the lirc_serial i get:
WARNING: Error inserting lirc_dev 
(/lib/modules/2.6.17-11-386/misc/lirc_dev.ko): Invalid module format
FATAL: Error inserting lirc_serial 
(/lib/modules/2.6.17-11-386/misc/lirc_serial.ko): Invalid module format

I don't understand, i used the same kernel version sources with their ubuntu 
patched. Not a generic kernel.
What am i doing wrong.
10x.

On Sunday 25 March 2007 01:21, Matan Ziv-Av wrote:
 On Sat, 24 Mar 2007, Tzahi Fadida wrote:
  I got an ancient infrared device
  PACKARED BELL - BPCS 146542 FCC ID: FODFMIR.
  It has a serial connector (DB9 pins).
  I don't have a remote for it.
 
  Is there a driver in linux that will support it?

 It works with lirc.

  Also, is there a program to record a tv remote buttons and use it to
  operate, for example, amarok, etc...
  10x.

 If your remote control is not already in lirc database, you need to
 record it with irrecord. Then, any program that supports lirc will work
 with the remote. For other programs you might need some tricks.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Blooming Filters (was: Re: [Job] MySQL consultation)

2007-03-22 Thread Tzahi Fadida
On Thursday 22 March 2007 11:25, Muli Ben-Yehuda wrote:
 On Thu, Mar 22, 2007 at 10:55:11AM +0200, Peter wrote:
  I am not an expert on this, but any algorithm that runs in O(1) or
  close to that for the data size you use is a candidate. The data
  size should be obviously less than 2^32 for x86 at least in any
  indexable dimension if you want a reasonably smooth ride. Larger
  things (like 30 million record databases to be used in real-time)
  move up to better things like 64-bit x86 and up. F.ex. 30 million
  records will require less than 7 bits per entry just to keep a
  complete linear index in 3GB of RAM (the maximum usable you can put
  in a x86 32 bit machine).

 With PAE you can stick in a lot more RAM on x86-32, but then your
 bottleneck becomes the 3GB of virtual address space for a single
 process.

Perhaps some kind of a hardware solution can be used. I.e. attaching a pci 
with memory and addressing it's 64gb memory either directly (if you have 64 
bit bus) or in two phases. It can also be any RAM space size you choose, but 
it will cost you (each additional bit doubles your costs :) ).

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Blooming Filters (was: Re: [Job] MySQL consultation)

2007-03-22 Thread Tzahi Fadida
On Thursday 22 March 2007 11:49, Muli Ben-Yehuda wrote:
 On Thu, Mar 22, 2007 at 11:46:36AM +0200, Tzahi Fadida wrote:
   With PAE you can stick in a lot more RAM on x86-32, but then your
   bottleneck becomes the 3GB of virtual address space for a single
   process.
 
  Perhaps some kind of a hardware solution can be used. I.e. attaching
  a pci with memory and addressing it's 64gb memory either directly
  (if you have 64 bit bus) or in two phases. It can also be any RAM
  space size you choose, but it will cost you (each additional bit
  doubles your costs :) ).

 Accessing memory on an external PCI adapter is *slow*. What you want
 to do is go 64 bit on the host, or barring that, use 32 bit + PAE and
 multiple address spaces (i.e., multiple processes).

Yes of course, you are constrained by the bus speed, however, the cost in $ is 
much lower if you use some kind of slower but adequate storage solution. 
In addition, if you plan to use 128gb,256gb of these things you cannot do it 
in main memory (since you don't have one large enough) and there are times 
you cannot dedicate all of memory to this filter since you need some for your 
server, otherwise you will have to have either two computers which will have 
slow networking response which defeats the purpose or NUMA (which i know very 
little about and i am not sure this is doing what i think it is doing, i.e. 
having a separate address bus per cpu) but you again constrained by bus speed 
(i think) between the processors or you will have to construct it smartly 
that each processor handles half the filter or it's section of the filter.


 Cheers,
 muli

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Blooming Filters (was: Re: [Job] MySQL consultation)

2007-03-22 Thread Tzahi Fadida
On Thursday 22 March 2007 14:58, Peter wrote:
 On Thu, 22 Mar 2007, Tzahi Fadida wrote:
  Perhaps some kind of a hardware solution can be used. I.e. attaching a
  pci with memory and addressing it's 64gb memory either directly (if you
  have 64 bit bus) or in two phases. It can also be any RAM space size you
  choose, but it will cost you (each additional bit doubles your costs :)
  ).

 So what you are advocating is an expensive RAM disk ? The process limit
 is still there. 32 bit cpu registers cannot easily manipulate more than
 3 GB of data. Any index etc must be limited to this size. The runtime
 for accessing a 'ramdisk' will explode wrt running a single large
 process.

Advocating is a strong word, i was suggesting. 
How exactly would you address 128gb,256gb?
Unless of course your system board and CPU supports such sizes...
As for 3GB, As i understand you must either have 2gb,4gb,... for this blooming 
filters, i.e. you need 4gb which does not leave much room for your kernel and 
apps in 32bit systems (and btw swapping is not really an option with this 
hash func).
As for expensive, some memories would not be adequate as CPU main memory, 
however will be adequate for storing this blooming filters but will be a lot 
cheaper than buying 64gb of ram or even 4gb. For example, even if it take 100 
micro seconds (probably much much less) to retrieve the value you still have 
1 such attempts in 1 second which is more than adequate for the spell 
check app they suggested.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Blooming Filters (was: Re: [Job] MySQL consultation)

2007-03-22 Thread Tzahi Fadida
On Thursday 22 March 2007 16:18, Peter wrote:
 On Thu, 22 Mar 2007, Tzahi Fadida wrote:
  Advocating is a strong word, i was suggesting. How exactly would you
  address 128gb,256gb? Unless of course your system board and CPU
  supports such sizes...

 The board does not care about sizes. Disk requests are serialized and
 they can be any lengths. Implementing a 1024 bit wide address counter to
 be pushed out serially to hardware is trivial even with an 8 bit cpu
 from 20 years ago. The problem is speed and size. Anything that fits in

And where prey tell, you can go to a store today and get a computer that 
support addressing of 1024 bits RAM?
Being realistic, you have a 32bit system in place and all you need is to 
implement the 2 or 4 gb blooming filter, why buy an insanely expansive new 
computer instead of just adding some PCI with some memory that would be good 
enough for your needs? Like an Asus battery backed ram drive in 50$ + as much 
ddrs you need. I think 1gb~=500nis *4 = 2000. ANS + 50$ = 2250nis.

New computer = 5000nis + 2250nis at least.

 1 register can be manipulated in 1 clock cycle or less, that is fast.
 Thus jumping around in an index or a tree using 32 bit integers on 32
 bit hardware with ~3-4GB of RAM is not a problem. When more than 32 bits
 are needed things slow down a lot. It can go from 1 clock cycle to 4-5.
 When the 'local' data size is larger than the cache, things slow down to
 main memory speed. When that is not large enough, you start using disk
 seeks and VM swapping. Strictly speaking, a 32bit machine could handle

Swapping is not an option with hashing like that since the seeks can be all 
over the place thus, rendering the whole thing as if you were using only the 
disk and practically worse.

 100TB or more of data, working as a Turing machine on the 100TB 'tape'
 (or tapes), but you really wouldn't want that (insert memories of
 recompiling Linux on i386 with 8MB ram here). One of the reasons RDBMSs
 'like' to run on 'bare' partitions is exectly this: they prefer to use
 their own seek, hash, and striping algorythms instead of relying on the
 OS. So by the time any dimension of the problem touches on 2^32 things
 can slow down 10-1000 and worse times (even without script kiddies using
 SQL and PHP4 scripts to handle the output).

From what i understand from PostgreSQL dev team which some were also involved 
in oracle and knows how the raw partition usage works, the gains, i 
understand, over using a common linux FS, is no more than 10-20%.

  As for 3GB, As i understand you must either have 2gb,4gb,... for this
  blooming filters, i.e. you need 4gb which does not leave much room for
  your kernel and apps in 32bit systems (and btw swapping is not really
  an option with this hash func). As for expensive, some memories

 There is no fixed size for Bloom filters, they are probabilistic. You
 can make a 10 bit Blooming filter, it depends on your hash algorythm.

Hence i said 2gb,4gb,...

 bogofilter) are also closely related to this afaik. The point is that
 there are ways to build very fast speculative indexes over huge data
 sets without actually storing the data. This can reduce the number of

I doubt that applies here since the hash function actually tries to spread the 
keys as much as possible to avoid false positives as much as possible.

 actual (expensive) lookups by orders of magnitude. I am not sure what
 Google uses for algorythms internally but from my adventures with web
 publishing and so on I would say that they are using similar principles.

They are not checking for existence they are locating actual data, thus, they 
would not use a hash function that can return false positives.
Btw, i was actually considering using blooming filters in my thesis for the 
full disjunctions algorithm, however, i decided that a simple hash table 
would be better. The reason was that if your application requires accuracy 
then, you cannot use a good enough answer. I.e. after getting a true value i 
would have to actually verify existence rendering it useless if i can just 
use a hash table (but not useless if the database does not support a proper 
hash table).

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Blooming Filters (was: Re: [Job] MySQL consultation)

2007-03-22 Thread Tzahi Fadida
On Thursday 22 March 2007 17:45, Peter wrote:
 On Thu, 22 Mar 2007, Tzahi Fadida wrote:
  On Thursday 22 March 2007 16:18, Peter wrote:
  On Thu, 22 Mar 2007, Tzahi Fadida wrote:
  Advocating is a strong word, i was suggesting. How exactly would you
  address 128gb,256gb? Unless of course your system board and CPU
  supports such sizes...
 
  The board does not care about sizes. Disk requests are serialized and
  they can be any lengths. Implementing a 1024 bit wide address counter to
  be pushed out serially to hardware is trivial even with an 8 bit cpu
  from 20 years ago. The problem is speed and size. Anything that fits in
 
  And where prey tell, you can go to a store today and get a computer that
  support addressing of 1024 bits RAM?

 Wht are you assuming that the computer needs to be able to address 1024
 bits of RAM if the address counter is made that wide (in software). You
 can easily map this 1024 bit address space so one part covers actual
 ram, another the video ram, another is mapped to a network drive,
 another ... it's called virtual memory, and it does not say anywhere
 that it is limited to one level. Of course this costs time. But ...

Listen, i did not suggest to map 1024 bits, i was using your example.
What you are talking about is PCI and other buses. On the same 32bit address 
bus you can address many data buses using bridges, which is exactly what i 
said from the beginning and yes, as muli said earlier, it is slow. 


  Being realistic, you have a 32bit system in place and all you need is to
  implement the 2 or 4 gb blooming filter, why buy an insanely expansive
  new computer instead of just adding some PCI with some memory that would
  be good enough for your needs? Like an Asus battery backed ram drive in
  50$ + as much ddrs you need. I think 1gb~=500nis *4 = 2000. ANS + 50$ =
  2250nis.

 Because you don't need a Blooming filter 4GB in size. You need one
 Blooming filter 500MB in size, two 256MB, four 128MB, and the last two
 fit into the second 800NIS PC (the first 500MB fits into the first).
 Anyway the B. filter is not good for storing data but it could be good
 to check f.ex. hash keys present/absent in a cache quickly and cheaply.

That was not the initial argument i brought. If you only need 128mb then there 
is no problem. I was talking about 2^32 potential entries. If i understand 
correctly, the wider the key the less false positives there are. Everything 
comes at a cost.


 The idea is that there is no canned solution. There is hard work to be
 done to make something support 30E6 records in real time. A Bloom filter
 may be a part of it. Maybe not. But SQL is almost certainly a part of
 the problem and not of the solution ... again I'm not an expert here.

SQL is not a problem, it is a generic solution. Obviously a specific solution 
will always be better than SQL. SQL provides you with a quick, generic way to 
store and retrieve data using relational algebra. If you know in advance that 
throughout the application life you will only need a hash table then there is 
no need to use SQL. Mitigating the generic solution is the RDBMS, which 
provides past the obvious retrieval and execution mechanisms, query 
optimizers, planners, etc... to get you as near as possible to the specific 
solution.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: [Job] MySQL consultation

2007-03-21 Thread Tzahi Fadida
On Wednesday 21 March 2007 16:26, Elazar Leibovich wrote:
 This is not the problem. The problem is that running Queries like:
 SELECT SQL_CALC_FOUND_ROWS c.cat_ID FROM
   posts AS l,post2cat AS a2c,
   categories AS c
 WHERE (date2006-01-01 AND date2006-01-02) AND(c.cat_ID=a2c.category_id
 AND l.article_id=a2c.post_id)
   AND (c.category_name=cat1 OR category_name=cat2)
 GROUP BY l.id
 HAVING count(a2c.rel_id)=%s
 LIMIT 10;

Ok, what i see here (again without knowing the internals of MYSQL), is that 
you have 3 tables posts and post2cat and categories. you perform a join:
(categories join post2cat) join posts. Naturally all keys mentioned in the 
join should have PRIMARY index otherwise it will be real slow.
Next, i see a further constraining clause of date..
So put a primary index there on the date clause and on the table the date 
attribute belongs to you will have to drop the old index on the id field and 
make it a secondary index (no choice but it is better).

Another thing you can do is to split some tables by date. If most queries are 
for certain dates then there is no reason to do queries on whole tables.

If you want to further make it run a lot faster you will have to say what is 
the rate of insertions into the tables.
What you can do in situations where you have huge data is to use a technique 
called Materialized Views. I expect you know what a view is. Well, a 
materialized view is just a view that is saved to a table and is updated each 
time there is an insertion (or selection, depends on your needs). This way, 
even if it takes 10 minutes to update the view. If the insertion only happens 
3 times a day, it is worth it. Other than that you will have to redesign your 
database/your software logic to work only on the relevant sets of data.

 Takes about 10 minutes to complete. I want to run many such queries, and
 I'm not sure if I can optimize it, or that since the DB is huge, I've met
 the SQL limit and I should start looking in other directions (reducing logs
 resolution etc.).
 Will the fact I'll have no TEXT columns speed things up? Will removing all
 columns except of id from the giant posts table throttle things up
 significantly?


-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Can I disable kernel switching?

2007-03-13 Thread Tzahi Fadida
Little top posting here since i am not completely sure i understand what 
Shlomi wants to do but i think i have an idea. In 2.6, the kernel views 
network addresses as SYSTEM addresses thus, pings will never hit the drivers 
it will be done in the kernel.
Here is an excerpt from LDD3:
Unfortunately, this effect can't be accomplished through IP number 
assignments alone, because the kernel wouldn't send out a packet through 
interface A that was directed to its own interface B. Instead, it would use 
the loopback channel without passing through snull. To be able to establish a 
communication through the snull interfaces, the source and destination 
addresses need to be modified during data transmission. In other words, 
packets sent through one of the interfaces should be received by the other, 
but the receiver of the outgoing packet shouldn't be recognized as the local 
host. The same applies to the source address of received packets.

To achieve this kind of hidden loopback, the snull interface toggles the 
least significant bit of the third octet of both the source and destination 
addresses; that is, it changes both the network number and the host number of 
class C IP numbers. The net effect is that packets sent to network A 
(connected to sn0, the first interface) appear on the sn1 interface as 
packets belonging to network B.

I.e., go and write/modify a driver to do that or find a way to do that in the 
packet mid-flight.

On Tuesday 13 March 2007 08:08, Shachar Shemesh wrote:
 Amos Shapira wrote:
  I'm still curious, though - will the Linux kernel in your situation go
  through the hassle to assemble and transmit Ethernet frames or will it
  notice that it's actually sending to its own address and just say sod
  it - I'll just copy the buffer in memory (which is what I half expect)?
 
  --Amos

 I haven't checked the kernel's code, but from what I know, Oron's
 solution will not work. I don't think there is a way to do what Shlomi
 is looking to do.

 The kernel views all IP addresses of all interfaces as his addresses.
 I know for a fact that, at least in the past, it would answer ARP
 requests destined for one of its interfaces, even if the request arrived
 on another interface. I don't remember at the moment for sure, but I
 think even the proxy arp would not affect this setting. One thing is
 for sure, though. If any kernel option will affect this case, it is
 proxy arp, and not IP forwarding. It's the layer 2 you wish to bypass,
 not the layer 3.

 One thing I would try is to send it to the broadcast address, rather
 than to the specific address for the other interface. This will, for
 sure, cause the packet to go out on the wire. You can disconnect the
 cable to find out whether it will also prevent the kernel from
 short-looping the packet.

 Shachar

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Big Off-Topic: Finding a linux related job.

2007-03-09 Thread Tzahi Fadida
On Friday 09 March 2007 10:58, Shamir Udi wrote:
 try IBM labs, they got big offices in Haifa.

Since you brought it up, that was the position that did not say no, but rather 
that the position was filled a day before my second interview. Also, this is 
the only company that bothered me in the what if department. It is also a 
bit strange since just a few years ago, i never thought i will be wanting to 
work for a very big company.
To their credit, i can say that they were the only company that knew what FOSS 
is, actually considered it a plus, and what PostgreSQL is :).
Cool company to work for, for sure.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Big Off-Topic: Finding a linux related job.

2007-03-09 Thread Tzahi Fadida
On Friday 09 March 2007 08:43, Geoffrey S. Mendelson wrote:
 BTW, after reading some of your other emails, what exactly is your degree
 in? Not the title, but what do you know?

Well, you are correct, it is not clear. The title goes like this:
Faculty: Industrial Engineering. Area: Information Systems. 
Degree: Information Management Engineering. 
I myself, still don't understand what that means aside of course of what that 
actually means :). What the faculty wanted to do is to change the title of 
the degree from just Information System because they wanted to stress that 
this degree is mainly about research and not courses like many universities 
does.
Research at the faculty can mean one of two things. A big project + small 
research thesis or a big research thesis. I choose the later as the 
definition for my thesis and also took the former :). Which means i have 200 
pages of no 1.5 spaced lines :) and the project of programming the research 
subject into PostgreSQL internals. Plus some researchers and i got out a 
paper to VLDB which is an implementation oriented conference, about our 
research and implementation.

In the work department, that means i programmed in C in PostgreSQL (Linux goes 
with out saying) for two years. Gave TA sessions for 2 years about 
ERD/SQL/XML/XQUERY/XPATH/VB/ACCESS(yes i know, access).
Basicly i gave the students a basic knowledge of how to analyze problems from 
a given story and make ERD diagrams, SQL queries, etc... and build it all in 
ACCESS. Made a project in java using the jade framework for building agent 
based, cooperating agents for driving down prices in english auctions.

Aside from that, i have 3 years experience in VB + windows Performance 
Counters(what can you do), and some specific experience in security. 
Experience in VXML. Experience with ColdFusion though no one seem to work 
with it a lot anymore.

My knowledge pretty much extends since i started my high school studies at 
bosmat thru 2 yeas engineering in bosmat, then BSc and now MSc.
However, in recent time i am trying to learn kernel internals and device 
drivers writing, this combined with the electronics knowledge from bosmat 
will perhaps give me some leverage.

The rest, if anyone is truly bored, can be found in my resume which is on my 
home page (written below).

 Personaly I would not waste any time on a FOSS project. Most managers
 won't care, it has no relevance to their world, and many startup managers
 will take it that you are more interested in the work than the money
 and if they do hire you, will take advantage of you (to be polite).

You are probably right, thing is, most don't know what FOSS is about so it is 
less of a problem :).

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Big Off-Topic: Finding a linux related job.

2007-03-08 Thread Tzahi Fadida
Hi,

I contemplated whether to send this email to this list and i usually would not 
be so public about this kind of thing, however, i am a bit at a loss and in 
need of some tips so i hope you'd bear with me. 
I have recently finished my MSc studies and wanted to find a linux related 
job. Not necessarily kernel hacking/device drivers though this is what most 
attracts. I can say without a doubt that it is difficult to convince 
potential employers of your capabilities without a proven working experience. 
It is the old chicken and the egg problem. I would have thought that 3 
separate degrees and my 2 years experience with PostgreSQL internals would 
have some leverage, however, i came to believe that this is a myth. Usually 
it goes like this, i get 2-3 professional interviews and then get a negative 
answer. I get the feeling these big companies i go to, just fill their 
interview quotas and in the end takes people with the most working 
experience. Especially in the north, where there are something like 30 people 
contending per job.
Can someone throw some tips as to how i should proceed?
10x.

P.s.: As to some who would have paranoid thoughts that this is self 
advertising, let me assure you that you are only partly correct, but there is 
no other way to ask the question. So i apologize in advance for that part.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Big Off-Topic: Finding a linux related job.

2007-03-08 Thread Tzahi Fadida
On Thursday 08 March 2007 23:24, Orna Agmon Ben-Yehuda wrote:
 On Thu, 8 Mar 2007, Tzahi Fadida wrote:
  Date: Thu, 8 Mar 2007 20:27:00 +0200
  From: Tzahi Fadida [EMAIL PROTECTED]
  To: IL Linux linux-il@linux.org.il
  Subject: Big Off-Topic: Finding a linux related job.
 
  Hi,
 
  I contemplated whether to send this email to this list and i usually
  would not be so public about this kind of thing, however, i am a bit at a
  loss and in need of some tips so i hope you'd bear with me.
  I have recently finished my MSc studies and wanted to find a linux
  related job. Not necessarily kernel hacking/device drivers though this is
  what most attracts. I can say without a doubt that it is difficult to
  convince potential employers of your capabilities without a proven
  working experience. It is the old chicken and the egg problem. I would
  have thought that 3 separate degrees and my 2 years experience with
  PostgreSQL internals would have some leverage, however, i came to believe
  that this is a myth. Usually it goes like this, i get 2-3 professional
  interviews and then get a negative answer. I get the feeling these big
  companies i go to, just fill their interview quotas and in the end takes
  people with the most working experience. Especially in the north, where
  there are something like 30 people contending per job.
  Can someone throw some tips as to how i should proceed?
  10x.
 
  P.s.: As to some who would have paranoid thoughts that this is self
  advertising, let me assure you that you are only partly correct, but
  there is no other way to ask the question. So i apologize in advance for
  that part.

 1. As an interviewer in the north (how uniform are those people?) I do not
 prefer experience only. Zest, intelligence, flexibility and the
 ability to learn on your own count a lot as well.

Obviously, i can't talk about all possible jobs in the north, i only recounted 
my experience.


 2.I see Aviram has not answered yet, so I can remind you of his regular
 answer: go get some proven experience working on a big FOSS project.

That is what i tried to do. To date, the companies i went save one (which 
surprisingly did not give me a negative answer, but rather before the second 
interview informed me that the position has been filled), did not care about 
FOSS projects. 99% did not even know what PostgreSQL is, even though some 
were in the storage business which is not necessarily related but should at 
least be acknowledged. Starting from scratch again now in another FOSS 
project, just to maybe get a job a year from now is problematic from obvious 
reasons. I think that at some point the employer needs to give you a chance.


 Orna.
 --
 Orna Agmon Ben-Yehuda http://ladypine.org/
 ICQ: 348759096


 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Big Off-Topic: Finding a linux related job.

2007-03-08 Thread Tzahi Fadida
On Thursday 08 March 2007 21:16, Geoffrey S. Mendelson wrote:
 On Thu, Mar 08, 2007 at 08:27:00PM +0200, Tzahi Fadida wrote:
  I have recently finished my MSc studies and wanted to find a linux
  related job. Not necessarily kernel hacking/device drivers though this is
  what most attracts. I can say without a doubt that it is difficult to
  convince potential employers of your capabilities without a proven
  working experience. It is the old chicken and the egg problem. I would
  have thought that 3 separate degrees and my 2 years experience with
  PostgreSQL internals would have some leverage, however, i came to believe
  that this is a myth. Usually it goes like this, i get 2-3 professional
  interviews and then get a negative answer. I get the feeling these big
  companies i go to, just fill their interview quotas and in the end takes
  people with the most working experience. Especially in the north, where
  there are something like 30 people contending per job.
  Can someone throw some tips as to how i should proceed?

 After years of seeing very few jobs that were Linux related, I've seen
 three this week. The one that was not posted here was for Kayote in
 Jerusalem.

 IMHO you have a problem a friend of mine had in 1970. He had a degree from
 Tel Aviv Univeristy and worked his way up to a high position in the
 data processing department of the police.

 He resigned his position and went to the Univeristy of Pennsylvania,
 which at the time was the TOP school for computr sicence to get a Master's
 degree. When he was done, he came back here.

 No one would hire him. He was far too qualified and wanted far too much
 money to start at the bottom, and as he put it no one would move over
 for me. He went back to the U.S. where there were plenty of new openings
 everyday and AFAIK is still there.

 The way I see it is you have several options:

 1. Go back to school and get a Phd.

I can't, i am academically exhausted :), if there is such a thing. Perhaps in 
10 years from now.

 2. Keep looking around for a startup that needs someone and has no one
to promote.

If you know of a one tell me. I was involved in several startups in the past. 
I am not averse but then again, not highly inclined to go that path with only 
options :)

 3. Convince someone that you really are willing to work at the bottom
and work your way up.

Again, if you know someone, please tell :)

 4. Hope for a boom, where people will be calling you asking to work for
them.

I think that instead of a boom, in the last few days there were corrections 
downwards so the boom in this context means, some small bubble just burst in 
wall street. I don't think we will ever going to see a climbing like before 
2000. I think this is as good as it gets.

 5. Go get an MBA and start your own startup.

To be honest, i tried something a few months ago but as you say, i needed an 
MBA to understand what not to do/start. Not a total waste though, at least i 
learned to Python pretty well.

 6. Go where the jobs are. How is your Chinese? :-)

Like my Japanese. :)

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Big Off-Topic: Finding a linux related job.

2007-03-08 Thread Tzahi Fadida
On Friday 09 March 2007 00:55, guy keren wrote:
 sorry for top-posting - but my answer is not related to what you write
 here, rather to things i've learned about you via other means.

10x for the input, i appreciate it.

 your first problem is quite simple - you went for an M.SC in industrial
 engineering, in something that people perceive as information systems.
 when people look for programmers, they have a tendency to prefer
 computer science/computer engineering/electrical engineering graduates.
 so your degree, in this case, is sometimes a barden. i don't know if and
 how you can remedy this. maybe preparing a resume that emphasizes on
 your programming knowledge, and on the fact that you're looking for a
 programming job, could help.

It tried recently to give more emphasis for programming. Perhaps i should make 
it more understood. You are right, perhaps people do get sidetracked by the 
other things on the resume and have a bit of a prejudice because of the 
information systems thing. 
Btw, just to clarify, information systems is a big field. It involves network 
algorithms, databases and obviously methodologies, etc... it is not so far 
from computer science. I think you are right that it is a wide misconception.

One thing i am not prepared to do, which someone advised me, and that is to 
remove diplomas/qualifications etc... since this is not morally right to your 
employer. I know some does that at the beginning to get started but it is 
still not right.

 your second problem comes from something about what you reflect. as far
 as i understood, you have fixed notions of what you want, and that's not
 neessarily programming. you talked about system analysis, and about
 configuration management, and then about programming. when you're not
 focused on the profession you want to work in (and these are 3 quite
 different things) - it reflects in the interviews. it passes the message
 of instability. i think you should work on deciding about what you want
 to do - and focus on it. mind you, not focus about i want to program
 databases. just about i want to program. if you're too peculiar, at
 least in this stage, regarding what you tell people - you'll be in a
 problem.

I admit that when i just started looking i casted a wide net, unfocused. I 
started at first to look for the obvious design and analysis but i quickly 
understood what they mean in the jobs descriptions and that is, for the real 
industrial engineer profession itself that relates to, well, production lines 
etc... Which, i know nothing about. 
Hopefully, what i say next won't overqualify me to some potential employer.
I was later offered a very highly (you would never believe how high) payed 
salary at 3 certain (often competing) gov related companies, which i can't 
say publicly their names. The jobs were a mix of low level C programming + an 
arcane configuration management. The problem with that is that after a 3 
months wait period for certain procedures to pass thru, divisions 
closed/reorganized, positions frozen. I can't believe it myself. Anyway, i am 
not doing this again. Maybe in 10 years like my plans for Phd :). 
Since then (about 2 months), nothing. I personally do not expect a very high 
salaries like they offered. I only expect to have a good, challenging and 
interesting job. I am completely willing to start at the beginning.
Basically, i am looking for a job :).

As i said above regarding the CV, in recent 2 months i have become very 
specific in saying out right that i wish to program and nothing else and that 
is what i say in interviews. 

 finally, the amount of _linux_ jobs in the north is rather small,
 relative to the total amount of jobs. consider compromising on this for
 a few years. i was unemployed for two month because i did not want to
 compromise - and that was after about 10 years of (formal) work
 experience on unix and linux - and living in haifa, like you.

I am willing to commute but most companies wants you to relocate. Problem is, 
i just did in Haifa :) and family depends on me to stay so i can't really 
relocate again.

 by the way, the attitude of they will have to give me a chance is not
 going to help you - they do not have to give you a chance - they have to
 think of what they want and need - they don't care about you. if they
 did - hen, as you said, there are 29 other people - they could
 randomally pick any one of them to give a chance to. it doesn't work
 this way.

Of course, you are completely right. That was said out of a little 
frustration. I have taken my pills since then ;).
10x again for the input, i think it helped.


 --guy

 Tzahi Fadida wrote:
  Hi,
 
  I contemplated whether to send this email to this list and i usually
  would not be so public about this kind of thing, however, i am a bit at a
  loss and in need of some tips so i hope you'd bear with me.
  I have recently finished my MSc studies and wanted to find a linux
  related job. Not necessarily kernel hacking

Re: Visual diff tool for patches.

2007-03-06 Thread Tzahi Fadida
On Tuesday 06 March 2007 01:09, guy keren wrote:
 look at the patch file - it has a very simple general format. i think
 you can grep for the file names in it, and with little parsing get the
 list of affected files.

Well, i was curious as to why there is no (famous) script to do such a simple 
thing conveniently, therefore, i have written a script to do this. The 
script(python), displays (gtk) the list of files to change from the patch 
file with an adjacent action (new file/delete file/lanuch a diff program - 
currently meld) and you can double click to perform the action. No need to 
copy anything, this is done in /tmp and in memory automagically.

Here is an example of a call to the script:
cat ../new/patch| python ../easierpatch.py - ./ -p1

Here is the link to the script (GPL) on my homepage:
http://tzahi.webhop.info/easierpatch.py

Enjoy.
-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Visual diff tool for patches.

2007-03-05 Thread Tzahi Fadida
Hi,
I was given a patch (that refers to several files in a directory) and i wish 
to review and merge the changes one at a time visually. 
I tried xfdiff which allows me to review the changes but it have a bug when i 
am trying to apply them one by one.
I tried kcompare, but it tries to do something weird. It tries to revert the 
patch file to my original files into new files, which is weird enough, but 
also when i try to save it says it can't upload the file to destination 
folder etc.. In this instance i tried source:patch destination:folder.
If i reverse the roles i get something weird: Error - the files are identical.
kdiff3 and meld does not seem to support patch files at all.

Can anyone suggest a proper tool?

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Visual diff tool for patches.

2007-03-05 Thread Tzahi Fadida
On Monday 05 March 2007 14:06, Alexander Indenbaum wrote:
 Meld : Diff and merge tool
 http://meld.sourceforge.net/

As i wrote below, meld does not seem to support patch files.
If you know otherwise, please tell me how to do it.

 On 3/5/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  Hi,
  I was given a patch (that refers to several files in a directory) and i
  wish to review and merge the changes one at a time visually.
  I tried xfdiff which allows me to review the changes but it have a bug
  when i am trying to apply them one by one.
  I tried kcompare, but it tries to do something weird. It tries to revert
  the patch file to my original files into new files, which is weird
  enough, but also when i try to save it says it can't upload the file to
  destination folder etc.. In this instance i tried source:patch
  destination:folder. If i reverse the roles i get something weird: Error -
  the files are identical. kdiff3 and meld does not seem to support patch
  files at all.
 
  Can anyone suggest a proper tool?
 
  --
  Regards,
  Tzahi.
  --
  Tzahi Fadida
  Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
  WARNING TO SPAMMERS: see at
  http://members.lycos.co.uk/my2nis/spamwarning.html
 
  To unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Visual diff tool for patches.

2007-03-05 Thread Tzahi Fadida
On Monday 05 March 2007 14:33, Alexander Indenbaum wrote:
 Did not see you mentioned meld specifically. Anyway I could sugest the
 following procedure.

 You could prepare two copies of the source tree: pristine and
 modified, where modified is pristine with patch applied.

 Then fire up meld (or any other merge tool) and just apply/move
 changes one by one from modified to pristine.

 Hope it helps.

That is a good idea and i am going to use it, however, it is very tiring to do 
so for each patch i receive. There got to be a better way, even from the 
console but interactive so i can approve or disapprove of a certain part.
I mean, how do kernel developers work with this. You cannot cut/paste all the 
time with many many patches instead of clicking like in kdiff and friends.
There got to be a better arrangement.


 On 3/5/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  On Monday 05 March 2007 14:06, Alexander Indenbaum wrote:
   Meld : Diff and merge tool
   http://meld.sourceforge.net/
 
  As i wrote below, meld does not seem to support patch files.
  If you know otherwise, please tell me how to do it.
 
   On 3/5/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
Hi,
I was given a patch (that refers to several files in a directory) and
i wish to review and merge the changes one at a time visually. I
tried xfdiff which allows me to review the changes but it have a bug
when i am trying to apply them one by one.
I tried kcompare, but it tries to do something weird. It tries to
revert the patch file to my original files into new files, which is
weird enough, but also when i try to save it says it can't upload the
file to destination folder etc.. In this instance i tried
source:patch destination:folder. If i reverse the roles i get
something weird: Error - the files are identical. kdiff3 and meld
does not seem to support patch files at all.
   
Can anyone suggest a proper tool?
   
--
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS: see at
http://members.lycos.co.uk/my2nis/spamwarning.html
   
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
 
  --
  Regards,
  Tzahi.
  --
  Tzahi Fadida
  Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
  WARNING TO SPAMMERS: see at
  http://members.lycos.co.uk/my2nis/spamwarning.html

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Visual diff tool for patches.

2007-03-05 Thread Tzahi Fadida
On Monday 05 March 2007 19:40, guy keren wrote:
 On Mon, 5 Mar 2007, Tzahi Fadida wrote:
  Date: Mon, 5 Mar 2007 16:03:49 +0200
  From: Tzahi Fadida [EMAIL PROTECTED]
  To: Alexander Indenbaum [EMAIL PROTECTED]
  Cc: Israel Linux Mailing list linux-il@linux.org.il
  Subject: Re: Visual diff tool for patches.
 
  On Monday 05 March 2007 14:33, Alexander Indenbaum wrote:
   Did not see you mentioned meld specifically. Anyway I could sugest the
   following procedure.
  
   You could prepare two copies of the source tree: pristine and
   modified, where modified is pristine with patch applied.
  
   Then fire up meld (or any other merge tool) and just apply/move
   changes one by one from modified to pristine.
  
   Hope it helps.
 
  That is a good idea and i am going to use it, however, it is very tiring
  to do so for each patch i receive. There got to be a better way, even
  from the console but interactive so i can approve or disapprove of a
  certain part. I mean, how do kernel developers work with this. You cannot
  cut/paste all the time with manymany patches instead of clicking like in
  kdiff and friends. There got to be a better arrangement.

 who said anything about copy and paste? all you do is:

 cp -rp original/ modified
 apply patch to modified.
 run the diff tool (meld, tkdiff, whatever) between original and modified.

That is a good idea, however, i only wish to automate this for the patched 
files (since i don't want to copy 20mb of files just for the few patched 
files). How can i only copy the files to be patched, let's say to /tmp and 
then run the patch on them.

 you can write a short script to automate these operations (except the
 usage of the diff tool, ofcourse). look at this:

 http://xxdiff.sourceforge.net/local/doc/xxdiff-patch-UNFINISHED.html

Looks interesting, buggy though. If i won't find anything working, perhaps i 
will try to change this to use kdiff3 or meld.


 --guy

   On 3/5/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
On Monday 05 March 2007 14:06, Alexander Indenbaum wrote:
 Meld : Diff and merge tool
 http://meld.sourceforge.net/
   
As i wrote below, meld does not seem to support patch files.
If you know otherwise, please tell me how to do it.
   
 On 3/5/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  Hi,
  I was given a patch (that refers to several files in a directory)
  and i wish to review and merge the changes one at a time
  visually. I tried xfdiff which allows me to review the changes
  but it have a bug when i am trying to apply them one by one.
  I tried kcompare, but it tries to do something weird. It tries to
  revert the patch file to my original files into new files, which
  is weird enough, but also when i try to save it says it can't
  upload the file to destination folder etc.. In this instance i
  tried source:patch destination:folder. If i reverse the roles i
  get something weird: Error - the files are identical. kdiff3 and
  meld does not seem to support patch files at all.
 
  Can anyone suggest a proper tool?
 
  --
  Regards,
  Tzahi.
  --
  Tzahi Fadida
  Blog: http://tzahi.blogsite.org | Home Site:
  http://tzahi.webhop.info WARNING TO SPAMMERS: see at
  http://members.lycos.co.uk/my2nis/spamwarning.html
 
  To unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | [EMAIL PROTECTED]
   
--
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS: see at
http://members.lycos.co.uk/my2nis/spamwarning.html
 
  --
  Regards,
  Tzahi.
  --
  Tzahi Fadida
  Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
  WARNING TO SPAMMERS:  see at
  http://members.lycos.co.uk/my2nis/spamwarning.html
 
  To unsubscribe, send mail to
  [EMAIL PROTECTED] with the word unsubscribe in the message
  body, e.g., run the command echo unsubscribe | mail
  [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-25 Thread Tzahi Fadida
On Sunday 25 February 2007 16:09, Alex Dover wrote:
 There was a question about business perspective of MySQL vs. PostgreSQL...
 As far as I can tell , MySQL is not free for commercial distribution. It's
 only free if you use it yourself or as part of some Open Source
 distribution.
 Correct me if I'm wrong (am I reading MySQL license wrong?)

And, PostgreSQL is free in that regard since it has a BSD license which allows 
it to be incorporated into any commercial distribution or embedded in any 
closed sourced project.


 -Alex

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Per process I/O (was Decorating a kernel function).

2007-02-25 Thread Tzahi Fadida
On Friday 23 February 2007 02:11, Amos Shapira wrote:
 Assuming the above (i.e. files are mostly not written concurrently by
 multiple processes), maybe you can try to track write(2) system calls.
 Mulli's old syscalltack (http://syscalltrack.sourceforge.net/) should have
 been able to do this for you if it was still relevant for current kernels
 (is it?)

I wonder if it is still relevant? Can someone say what are the advantages of 
using (=after porting of course) syscalltrack in 2.6 kernel over existing 
solutions. Like, for example, debugging uml, using kprobes, etc...
In addition, i think that in 2.4 there was some kind of syscall table which 
was exported and which you could replace syscall functions, but in 2.6 they 
stopped this practice and made this technique obsolete if not highly 
difficult/impossible.


 --Amos

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Per process I/O (was Decorating a kernel function).

2007-02-22 Thread Tzahi Fadida
On Thursday 22 February 2007 14:51, Gilad Ben-Yossef wrote:
 Tzahi Fadida wrote:
  A bit of a followup from my previous question.
  Are there already solutions to the per process I/O stats?

 For reason already explained on this mailing list before, your request is
 meaningless.

 The read/write etc. operation of a process translate very badly to real I/O
 ops.

It may not be accurate from various reasons, namely, the I/O queue CFQ,etc...
which delays certain writes reads depending on the position of the head of the 
disk. In addition, dirtied buffers which are written to disk in unknown 
future time.
However, for the purposes of what most people needs this, it should do.
For example, most people are interested to know, what process is generating 
most/a lot of/not at all  I/O requests and they will be willing to take an 
estimate. For that it will suffice. Unless, of course, there is a facility in 
linux that does that, that i am unaware of.


 Gilad

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-22 Thread Tzahi Fadida
Regarding a multi-master clustering solution that someone asked for. Such a 
product was advertised in the pgsql-announce mailing list. Since the product 
does not seem to be free i will not advertise it here but rather redirect the 
readers to the mailing list archives which can be accessed at PostgreSQL main 
site.

On Monday 19 February 2007 05:26, Amos Shapira wrote:

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Decorating a kernel function

2007-02-21 Thread Tzahi Fadida
On Wednesday 21 February 2007 08:19, Baruch Even wrote:
 * Tzahi Fadida [EMAIL PROTECTED] [070221 02:38]:
 I'm not aware of a way to do what you ask (redirect the function to you
 and be able to call it), you can however look at kprobe which is part of
 the kernel. kprobe allows you to stick trap points into the kernel and
 be called at that point, I don't remember the exact name but they have a
 method to stick the probe at the end of the function (and also at the
 beginning or arbitrary location).

 I've used this method to trace methods and it's fairly expensive since
 it uses int3 trap points to do its work but if performance is not
 critical for you it should be fine.

kprobe is what i was looking for. Actually, jprobe so i cal also view the 
function parameters i am attaching to. The problem is that kprobe does not 
appear on the uml(user mode linux) list in menuconfig which probably means it 
is not supported there. - i need to compile a real kernel or use vmware or 
something like that. I prefer to continue to use uml.
Let's say i do wish to change the kernel code. How do i add a code that will 
refer to my module once it is loaded? I.e. i wish to add a stub in that 
function in the kernel that will call a function in my module if it is up.


 Baruch

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Decorating a kernel function

2007-02-21 Thread Tzahi Fadida
On Wednesday 21 February 2007 15:36, Baruch Even wrote:
  kprobe is what i was looking for. Actually, jprobe so i cal also view the
  function parameters i am attaching to. The problem is that kprobe does
  not appear on the uml(user mode linux) list in menuconfig which probably
  means it is not supported there. - i need to compile a real kernel or
  use vmware or something like that. I prefer to continue to use uml.
  Let's say i do wish to change the kernel code. How do i add a code that
  will refer to my module once it is loaded? I.e. i wish to add a stub in
  that function in the kernel that will call a function in my module if it
  is up.

 The way to do it is hard and fraught with peril, it will require
 assembly programming which I'm not sure I'll be able to do myself, but
 can try to pass the idea.

 What you need is to put a trampoline of sorts at the start of the
 function, this is similar to what kprobe does, the difference would be
 that kprobe places an int3 and you'll need to place a direct call to
 your replacing method. At the end of your function you need to call the
 original method, either bby temporarily restoring the original location
 or by copying it to another place and running it there.

 I think it will be easier to run under qemu/kvm than do this.

I think you misunderstood me. I did not mean something fancy like that. What i 
meant was that for this point of time i do wish to change the kernel code. 
I.e. there is no need to hack like kprobe. I just want to call my function in 
my module from the kernel function. I.e. something like:
kernel_func(parms...){
...
  If (my_module is in memory){
my_module-my_func(params...);
  }
...
}

Later, after some dev time, for the non uml kernel, i will use kprobe tactics.


 Baruch

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Per process I/O (was Decorating a kernel function).

2007-02-21 Thread Tzahi Fadida
A bit of a followup from my previous question.
Are there already solutions to the per process I/O stats?
Not that my fun exercises into implementing a better block_dump would be 
commercial grade (but rather fun grade), however, this is what drove me to 
this when i was searching for just this tool for my thesis 1-2 years ago.

P.s. got the stub into my module to work by exporting a structure that 
includes a function pointer from the kernel and in the module attach the real 
function to the pointer in the kernel structure. Later this could be either 
committed to the kernel tree (AS IF :) ) or be replaced by jprobe.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Decorating a kernel function

2007-02-21 Thread Tzahi Fadida
On Thursday 22 February 2007 08:25, Shahar Dag wrote:
 Hi

 I think that I have a dirty solution (but I must worn you that I am a
 programmer, not a Linux expert)
 - add global pointer to a function (the type of the function should be the
 type of the function you want to call). A more elegant way (and more
 complicated) would be to use non global pointer.
 - initialize that pointer to NULL
 - in the original function check the pointer, if it is not null call the
 function pointed by the pointer
 - on loading your new function, change the global pointer to the function
 address (now you can also change the function or remove it by manipulating
 the pointer)

That is what i did yesterday, but it is not that simple. When you load a 
module the kernel has to somehow know what are the symbols that were already 
defined in the kernel and what symbols are in your module. He does that 
dynamically using a symbol table. The pointer needs to be exported.
In addition, kernel functions, especially that one are re-entrant and many 
times locking is not an option. I will probably have to employ some circular 
buffer with no locking and later a background thread will crunch these.


 Have fun
 Shahar
 ___
__

 I am looking for old Vinyl record.
 If you have any that you don't need please send me mail to
 [EMAIL PROTECTED]

 Thanks
 Shahar

 - Original Message -
 From: Tzahi Fadida [EMAIL PROTECTED]
 To: Baruch Even [EMAIL PROTECTED]
 Cc: Linux-IL linux-il@linux.org.il
 Sent: Wednesday, February 21, 2007 14:46
 Subject: Re: Decorating a kernel function

 On Wednesday 21 February 2007 08:19, Baruch Even wrote:
  * Tzahi Fadida [EMAIL PROTECTED] [070221 02:38]:
  I'm not aware of a way to do what you ask (redirect the function to you
  and be able to call it), you can however look at kprobe which is part of
  the kernel. kprobe allows you to stick trap points into the kernel and
  be called at that point, I don't remember the exact name but they have a
  method to stick the probe at the end of the function (and also at the
  beginning or arbitrary location).
 
  I've used this method to trace methods and it's fairly expensive since
  it uses int3 trap points to do its work but if performance is not
  critical for you it should be fine.

 kprobe is what i was looking for. Actually, jprobe so i cal also view the
 function parameters i am attaching to. The problem is that kprobe does not
 appear on the uml(user mode linux) list in menuconfig which probably means
 it
 is not supported there. - i need to compile a real kernel or use vmware or
 something like that. I prefer to continue to use uml.
 Let's say i do wish to change the kernel code. How do i add a code that
 will refer to my module once it is loaded? I.e. i wish to add a stub in
 that function in the kernel that will call a function in my module if it is
 up.

  Baruch

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-20 Thread Tzahi Fadida
On Tuesday 20 February 2007 01:53, Amos Shapira wrote:
 On 19/02/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  On Monday 19 February 2007 05:26, Amos Shapira wrote:
   Hello,
  
   Is it possible to configure PostgresQL to use raw disk partition, like
   Oracle does?
   If not - is there any recommendation for favourite filesystem type to
 
  use?
 
  Most certainly not.
  PostgreSQL relies on the OS and FileSystems it inhabits for all I/O
  operations. Raw partition are not worth the extra effort anycase on an OS
  such as linux which is already efficient with files and extents of those
  files. If you want you can use XFS to increase efficiency at the expense
  of

 Well I didn't expect Postgres to do the actual SCSI calls to the disk, but
 I sort of guess that since all the database needs from the OS is a bunch of
 disk blocks and the database file can be usually pre-allocated, it might
 be possible to do away from the complexity (and time penalty) of a FS which
 assumes that files have to be created/resized/removed all the time. That's
 the way Oracle databases at least used to be configured for many years
 (taking the update from Ira into account).

In PostgreSQL they don't even do the caching (more or less), they let the OS 
to do it. This is a common question and the claim of the PostgreSQL dev 
community is that the added benefit of a raw fs will be negligable. I for one 
believe it because many of them uses oracle also. Not to mention PostgreSQL 
is more or less compatible in syntax with oracle. I also can tell you that 
with recent stable versions 8+ and recent kernels you can probably achieve 
almost raw fs capabilities by using certain file options (which they do). 
They also use techniques as back writings etc... This thing is very advanced.


 more chance for data loss in case of a crash since it has a large caching

  mechanisms. You can also use raid to increase performance. Additionally
  you
  may try to put the transactions log, which is the bottleneck in
  databases, on
  a ram memory that is backed by a battery. I think i saw a battery PCI
  card at
  asus for 50$. Add 128mb ram and you are good to go.

 That (battey-backed RAM) sounds like a good tip. Will keep it around.
 Thanks. If you have more specific details to help Google for such a product
 I'll appreciate to hear them.

Check out these sites, i think these are the real thing:
http://www.anandtech.com/tradeshows/showdoc.aspx?i=2431p=5
http://www.cenatek.com/

However, also remember to periodically back up your stuff to the HD. I trust 
batteries only so far. :)


 Cheers,

 --Amos

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-20 Thread Tzahi Fadida
On Tuesday 20 February 2007 01:48, Amos Shapira wrote:

 I'm still digging postgresql.org and last time I went to a (large) book
 shop I saw an entire section (about 6-7 shelves) about MySQL but not a
 single book about PostgresQL.

Did you also compare this to Oracle books? The problem is that PostgreSQL is 
an overkill for the average joe developer. MySQL is simpler, faster, for the 
common made-at-home web app. I used PostgreSQL for 3 years, mainly for 
development and i can tell you it is a magnificent peace of software. I 
suggest you use the site book, which is free, as i did. It is constantly 
updated and is very very comprehensive.


 Cheers,

 --Amos

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-20 Thread Tzahi Fadida
On Tuesday 20 February 2007 14:37, Maxim Veksler wrote:
 What about redundancy. I need an active-active cluster for databases
 to get the 5*'9's up time euphoria. Is there an open source database
 that can do that? Planing to? Tried to?

PostgreSQL can do that and i am betting MySQL can too. Though not alone. You 
need an additional product that works with them.
Check out Slony-l and friends. Usually we are talking about Master-Slave 
replication service but i think both can be active. I.e. one point of change, 
multiple points of read. For peer to peer products, you'd perhaps have to 
check out other products but i am not keeping current with new versions.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-20 Thread Tzahi Fadida
On Wednesday 21 February 2007 01:43, Amos Shapira wrote:
 On 20/02/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  In PostgreSQL they don't even do the caching (more or less), they let the
  OS
  to do it. This is a common question and the claim of the PostgreSQL dev
  community is that the added benefit of a raw fs will be negligable. I for
  one
  believe it because many of them uses oracle also. Not to mention
  PostgreSQL
  is more or less compatible in syntax with oracle. I also can tell you
  that with recent stable versions 8+ and recent kernels you can probably
  achieve almost raw fs capabilities by using certain file options (which
  they do). They also use techniques as back writings etc... This thing is
  very advanced.

 Thanks for the clarification.

  That (battey-backed RAM) sounds like a good tip. Will keep it around.
 
   Thanks. If you have more specific details to help Google for such a
 
  product
 
   I'll appreciate to hear them.
 
  Check out these sites, i think these are the real thing:
  http://www.anandtech.com/tradeshows/showdoc.aspx?i=2431p=5
  http://www.cenatek.com/
 
  However, also remember to periodically back up your stuff to the HD. I
  trust
  batteries only so far. :)

 Well - if I use it only for transaction logging then hopefully the data
 there will be useful only for fractions of seconds, wouldn't it?

Depends. Sometimes queries takes longer. However it is not the point.
The reason for backupping once in a while is for PITR - point in time recovery 
which is a new facility in PostgreSQL 8+. I am not an expert on this subject 
but IIRC, imagine you backupped the database at 8am. At 9am the database 
crashed. The data at the database is too corrupt due to various insane 
reasons you can conjure in your mind. However, the transaction log survived.
It turns out, it kept all the transactions back to 8am when you backed up your 
database. YAY!, you can basically rerun all the transactions from that time 
using the log and reconstruct the database. 


 Thanks,

 --Amos

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-20 Thread Tzahi Fadida
On Wednesday 21 February 2007 01:39, Amos Shapira wrote:
 On 20/02/07, Ira Abramov [EMAIL PROTECTED] wrote:
  Quoting Amos Shapira, from the post of Tue, 20 Feb:
   that was only a temporary solution for RAC untill OCFS came along and
  
   temporary? Back in '99 Oracle wouldn't have had it any other way.
   When
 
  we
 
  yes, temporary the way that punched cards were the best thing till
  magnetic media and interactive terminal were perfected. Get with the
  times.

 Ah, that kind of temporary...

  Anyway, now that OCFS is out and about - would it be recommended for
  other
 
   databases besides Oracle or is it too Oracle-specific?
 
  supposedly it's not Oracle specific, but it has really low I/O for
  anything else you try to do with it. it's more like raw device you can
  look at and back up via the VFS, and it's useless if you are not
  running a cluster.

 OK, thanks for the explanation. It still sounds like Oracle are trying to
 minimize the penalty for going through the file system layer, tough.

 does PostgreSQL even support running two instances of the DB on two

  separate nodes over GFS?! I don't think it does. that's available only
  from the big guns like OracleDB and DB2 probably.

 So far my searches came to a conclusion that PostgresQL doesn't support
 shared disk, at least not out of the box.

What do you mean by shared disk?
Maybe this is what looking for:
http://www.postgresql.org/docs/8.2/interactive/manage-ag-tablespaces.html


  I'm still digging postgresql.org and last time I went to a (large) book
  shop
 
   I saw an entire section (about 6-7 shelves) about MySQL but not a
   single book about PostgresQL.
 
  it's a bad bad bad statistics indicator, but I think you may find that
  on amazon.com (not co.au) the situation will be very similar.
 
  and then again, maybe MySQL is enough for the task?

 Maybe. But I generally like PostgresQL's completeness and robustness,
 especially when compared to MySQL.

 Also from talking to people who use PostgresQL to run a Very Important
 Database (TM) it looks like although they were a bit apologetic about speed
 compared to MySQL when I asked what should I use, their recommendation was
 to use MySQL for more transient data with not many inter-record relations
 and PostgresQL for data which requires many relations and has to be relied
 on for a long term.

 --Amos

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Decorating a kernel function

2007-02-20 Thread Tzahi Fadida
Hi,
Is it possible to replace a function in the kernel (using a module) without 
recompiling. This function is an exported function. Basically i wish to 
extend a functionality of a function (without asking for a stub at this 
point). Specifically the function void submit_bio(int rw, struct bio *bio)
in /block/ll_rw_blk.c 
Not that i can't recompile the kernel, it is just that i want to create a 
module that would be able to be distributed with more convenience.

The idea, of course, is to call the submit_bio using the new function but also 
do other stuff afterward.

Basically i wish to extend the functionallity of block_dump (for fun and study 
purposes) to be centralized in some /proc entry with some stats.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-19 Thread Tzahi Fadida
On Monday 19 February 2007 05:26, Amos Shapira wrote:
 Hello,

 Is it possible to configure PostgresQL to use raw disk partition, like
 Oracle does?
 If not - is there any recommendation for favourite filesystem type to use?

Most certainly not. 
PostgreSQL relies on the OS and FileSystems it inhabits for all I/O 
operations. Raw partition are not worth the extra effort anycase on an OS 
such as linux which is already efficient with files and extents of those 
files. If you want you can use XFS to increase efficiency at the expense of 
more chance for data loss in case of a crash since it has a large caching 
mechanisms. You can also use raid to increase performance. Additionally you 
may try to put the transactions log, which is the bottleneck in databases, on 
a ram memory that is backed by a battery. I think i saw a battery PCI card at 
asus for 50$. Add 128mb ram and you are good to go.


 So far google'ing around haven't revealed anything about the subject.

 And BTW - I found the following link which also mentions procedures to
 convert MS Access databases to Psql:
 http://www.postgresql.org/docs/techdocs.3

 Cheers,

 --Amos

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PostgresQL database on raw partition (and something about Access conversion)

2007-02-19 Thread Tzahi Fadida
On Monday 19 February 2007 11:58, Israel Shikler wrote:
 Hi ,

 I see your post here concerning PostgresQL.
 We are using RedHat  Mysql.
 Is there any advantages using PostgresQL over Mysql?

There are advantages both ways. 
You will need PostgreSQL when:
1) You need a database that can handle a very large load of queries per a 
small unit of time (secs,minutes).
2) You need a database that can handle transactions in an efficient manner, 
plus, you require a transactional system that has many features like save 
points in a middle of a transaction and you can role back to a savepoint 
instead of the whole transaction.
3) You need a very wide and extendable syntax and operations.
4) You need an RDBMS that can handle itself in catastrophies such as point in 
time recovery. i.e. using the transactions log to reconstract the last 
queries, etc...

All of the above requires investment of resources which are wasted if you 
don't need those advantages. MySQL will serve you better for a small load of 
queries. It has a less extensive tranactional support. It's syntax is suited 
for smaller applications. Finally, PostgreSQL exists in various forms for 20 
year so it is a beast of a thing that is very mature. MySQL has only matured 
in features in the last years. This race to features left much to be desired.

There is no reason to use PostgreSQL for small apps or small load apps, in 
fact perhaps even MySQL is an overkill and there are even more efficient 
databases.


 Regards,

   Israel Shikler
 Softkol Ltd

 -Original Message-
 From: [EMAIL PROTECTED]
 [mailto:[EMAIL PROTECTED] Behalf Of Tzahi Fadida
 Sent: Monday, February 19, 2007 11:44 AM
 To: Amos Shapira
 Cc: Linux-IL
 Subject: Re: PostgresQL database on raw partition (and something about
 Access conversion)

 On Monday 19 February 2007 05:26, Amos Shapira wrote:
  Hello,
 
  Is it possible to configure PostgresQL to use raw disk partition, like
  Oracle does?
  If not - is there any recommendation for favourite filesystem type to
  use?

 Most certainly not.
 PostgreSQL relies on the OS and FileSystems it inhabits for all I/O
 operations. Raw partition are not worth the extra effort anycase on an OS
 such as linux which is already efficient with files and extents of those
 files. If you want you can use XFS to increase efficiency at the expense of
 more chance for data loss in case of a crash since it has a large caching
 mechanisms. You can also use raid to increase performance. Additionally you
 may try to put the transactions log, which is the bottleneck in databases,
 on
 a ram memory that is backed by a battery. I think i saw a battery PCI card
 at
 asus for 50$. Add 128mb ram and you are good to go.

  So far google'ing around haven't revealed anything about the subject.
 
  And BTW - I found the following link which also mentions procedures to
  convert MS Access databases to Psql:
  http://www.postgresql.org/docs/techdocs.3
 
  Cheers,
 
  --Amos

 --
 Regards,
 Tzahi.
 --
 Tzahi Fadida
 Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
 WARNING TO SPAMMERS:  see at
 http://members.lycos.co.uk/my2nis/spamwarning.html

 To unsubscribe, send mail to
 [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: recvmsg on dgram socketpair blocks on open socket?

2007-02-19 Thread Tzahi Fadida
I did not understand the current problem, however, just giving my 2c blindly 
so...:
1. Internet sockets using tcp protocol is the same as using select with a 
timeout since this is what the tcp protocol is doing. Timeouts.
2. Can't all this discussion be solved by simply reading the /proc directory 
as a failsafe measure for whatever is the purpose of finding out if a process 
is still alive?
I.e., do it in tiers, if one thing can give you a straight answer use another 
one and another and finally use something like the /proc to make sure your 
process is among the living or a zombie. Just like doing ps (which also 
uses /proc).

On Friday 16 February 2007 03:07, Amos Shapira wrote:

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Linux servers vs. Microsoft servers

2007-02-18 Thread Tzahi Fadida
On Sunday 18 February 2007 01:57, Uri Even-Chen wrote:
 On 2/17/07, Shachar Shemesh [EMAIL PROTECTED] wrote:
  Is it a requirement that it be Access, or should it be a database
  server?

 It has to be Access - it's not only a database, but a whole program
 written in Access, with at least 50 forms, at least 100 queries and
 lots of VB code.  I wrote it myself.  A few users in the office can
 use it simultaneously, they can update tables, and Access locks and
 unlocks records and tables automatically.  If a table is locked by one
 user and another user tries to open it, he is denied.  There is also a
 database of tables with many connections, this is in Access and in
 general it can be any database, but we are happy with access and we
 don't want to convert.  But if we do decide to convert, we will
 convert only the tables (there are at least 50 of them) and not the
 entire program.

If you don't have the strength, then as was said before, you can use a samba 
share which looks and works exactly as a windows share (perhaps better).
If you want to invest a little time. just write a little script or use a 
program to convert your tables to mysql or postgresql. Once there,
you have to do mostly 3 things:
1) create a table link in access table list to the server table.
2) make sure the syntax of your queries is compatible with mysql/postgresql.
3) the locking mechanism would have to be converted to either a transaction
or simply locking again which is specific to each database.

As for the dependency on office/windows. You can greatly reduce that by:
1) If you will use a different program than access, you could only buy the 
lower costing office (IIRC) suite. What is that program that would replace 
access and run on windows? can't tell you that. You have the one on open 
office but access is fully mature plus. It would be a tough act to follow but 
remember that this is a business and not a beauty contest.
Also, if you try open office, try it directly with an sql server to avoid the 
local db bugs it might have because of it's maturity levels.
2) the windows server would be replaced by a linux server.



 It's very convenient for us that a few people can use the Access
 program simultaneously, and we don't want to lose this feature.  But
 as I said, the server will also be used for other purposes, such as a
 mail server.

 Access requires Windows, and that's one of the reasons we can't
 convert the entire office to Linux.  Each user has Microsoft Access
 installed on his computer.  It's smart on Microsoft side - we pay lots
 of money to Microsoft.  Windows and Microsoft Office for each computer
 in the office.  In theory I would like to convert the entire office to
 Linux, but in reality I don't see how it will work.

 I also think most people, who are not computer programmers, prefer
 Windows.  If they just use the computer for typing Word documents or
 using E-mail, Windows is usually easier to use or at least that's what
 they're used to.  How many companies do you know, which are not
 computer companies, who use Linux?

 Uri.

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: MySQL vs. Microsoft Access

2007-02-18 Thread Tzahi Fadida
On Sunday 18 February 2007 11:24, Uri Even-Chen wrote:
 On 2/18/07, Arieh Skliarouk [EMAIL PROTECTED] wrote:
  http://dev.mysql.com/doc/refman/5.0/en/ansi-diff-foreign-keys.html

 Thanks!  I didn't know about InnoDB and Foreign Keys.

 Is it possible to define Primary Keys and Foreign Keys with more than
 one field?  I used Access to define keys with up to 5 fields.

In postgresql you can definately do. However, i suspect you have a worse 
problem of a poor design for your tables. Read about ERD and convertion rules 
of ERD to relations.


 The Access visual interface is very convenient.  Does MySQL have a
 similar interface?

 Which text encoding should I use?  Should I use Unicode (UTF-8), or
 Hebrew Windows encoding?

 Uri.

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: MySQL vs. Microsoft Access

2007-02-18 Thread Tzahi Fadida
On Sunday 18 February 2007 10:28, Uri Even-Chen wrote:
 On 2/18/07, Arieh Skliarouk [EMAIL PROTECTED] wrote:
  I see only-tables conversion as good first step. There is howto on
  access-mysql conversion:
  http://www.kitebird.com/articles/access-migrate.html

 There is a great feature we use with the tables stored in Microsoft
 Access.  I defined one-to-many connections between tables - at least
 50 connections.  Access enforces these connections - you can't add a
 record to the table with many if there is not a key in the table
 with one connected to it.  You also can't modify the key data on
 both connected sides, and you can't delete connected records on the
 one side of the connection.  If I want to create an Access query, I
 open the tables database and add tables, and the connections are added
 automatically.

There could be a problem with that. With the cascading record updates plus 
deletes (which are dangerous mind u) and many to one restrictions.
You see, IIRC all you have to do is link the tables from the access table list 
to the tables in mysql/postgresql. However, the cascading thingie might not 
work. i.e.: if u delete a record on one table it's foreign key is redirected 
and the record it was connected to is also deleted (and vice versa).
If you use cascading thing, you should first try to see if it works on the 
server. Simple to do. The one to many connections can be enforced in any 
database (specifically postgresql which i know well) and also the cascading 
thing can be done in probably any database. The question is, will it work 
automatically without you doing anything but linking the tables.
If you are undeterred by about 1hr-2hr additional work you can also do it in 
mysql/postgresql whatever.


 Are there table connections in MySQL?  If not, I don't think we will
 want to use MySQL.  I can live without the automatic query feature
 (although it's very convenient), but without connections the database
 can get corrupted.  Access prevents us from deleting records which are
 connected to many records in other tables, and without the
 connections they can be deleted.  It can create all sorts of mess.
 For example, if you have an employee with all sorts of data, and the
 employee leaves the company.  If you delete the employee, the data
 will not be connected and will have an employee ID which is not
 valid.  The employee ID is just a number, you will not even know who
 is the employee.  Tomorrow you might add another employee and give him
 the same number, and the data will be connected to him.  For example,
 he can be listed as the person who signed orders, and not the former
 employee.  The Access connections will not let you delete an employee
 record if it has data connected to it.  In this respect, I don't think
 MySQL is compatible with Access.

 Uri.

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Linux servers vs. Microsoft servers

2007-02-17 Thread Tzahi Fadida
On Saturday 17 February 2007 10:20, Uri Even-Chen wrote:
 Dear Linux people,

 First, I want to apologize if any of you were offended or harassed by
 the messages I sent to this list about 3 months ago.  I didn't mean to
 offend anybody.

 I want to ask you something about Linux servers.  My father thinks
 about installing a new server in his company, which will be used both
 as a mail server, a server for file sharing, and a server for
 Microsoft Access.  The Access program will be executed by a few users
 simultaneously.  Currently there is a Microsoft server which runs the
 Access program, and it works well.  But it's an old server.

I am trying to understand the thing about the access. As i know access, it is
just a program that runs on your workstation and opens a file on a 
drive/network. If that is the case, then you can put the file on a samba 
share and there would be no difference. The other possibility i know is that 
the access file contains references to a microsoft sql server. This 
complicates matters because you will need someone with knowledge to port the 
tables to something like postgresql/mysql depending the application.
If there is some kind of access server i did not know about then cancel all of 
the above and just install windows. If not, then linux is the cheaper to buy/ 
cheaper to get support (nil, just ask here). Mail server can be replaced. 
especially if you do not share calendars. If you do share, there might be a 
need to do a porting work. File sharing i would not even talk about because 
it is the most easy thing here.


 My father asked an expert from a company we work with, and the expert
 recommended to use a Microsoft server (Windows 2003).  I would like to
 know if you think that a Linux server is better.

I would recommend windows 2003 too, especially if i'll be at the receiving end 
of the TCO ;).


 Best Regards,
 Uri Even-Chen

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Linux servers vs. Microsoft servers

2007-02-17 Thread Tzahi Fadida
On Sunday 18 February 2007 00:00, Geoffrey S. Mendelson wrote:
 On Sat, Feb 17, 2007 at 11:09:26PM +0200, Yedidyah Bar-David wrote:
  I am not sure I understood the OP question, and neither am I sure others
  did. If you intend to use the server, as others guessed, only as a file
  server for Access, then samba will probably be just as good. If they
  intend to have few Access clients connect to MS SQL server, then they
  might be able to make it connect and work with MySQL, and probably
  others. I personally tried and had few users that did this, with Access
  and MySQL. Google for relevant keywords, you'll find the need plugin and
  docs.

 Access works as a local database program with it's own format databases.
 If you want to share databases with Access, you really need Microsoft
 SQL server which runs on a Windows server.

 Without SQL server you don't get record locking, etc. If you try to use
 Access to update a database on a SAMBA share, you will end up with corrupt
 data.

I am not completely sure of that. Access can access (:)) VB code, i.e. it 
might handle locking to tables using a regular software code the application 
designer created (hopefully). 


 If you only have one user with write access, you can place the
 Access database on a SAMBA share and you will not have any corruption
 problems. You may get odd results because the read only clients will not
 be notified of updates and may use cached copies of records.

 Using MYSQL as a backend to Access makes a lot of sense in a Linux
 server/mixed client environment. Then you can use PHP or similar languages
 on Apache web servers while Windows clients can do simple queries or
 data base updates using Access.

 However adding a single Linux server with MySQL to avoid the cost or
 complexity of Windows Server and M/S SQL server is IMHO money poorly
 spent.  In the end you will either spend as much money or more as the
 license fees on a good consultant to install and maintain the Linux/MYSQL
 system, or end up with nothing but headaches and lost data.

I don't see it. Installing mysql is pretty much a breeze and the son can help 
him. Just exporting the tables from access to mysql is also very simple.
Finally, using some plugin/technique (can't recall right now) you just add a 
link to an external table in the table list. Thus, there is no need to change 
the application for that. The REAL problem is inside the application itself. 
If the program needs a transaction support in a multi-user environment then 
there is a need to change the application logic itself worst case or at the 
easy case just add a start/end transaction clauses where needed.

If we are talking ms sql server, then first you need to export the tables to 
mysql. However, if there are stored procedures/triggers the complexity is 
increased. If not, then it's very easy to do. In addition, you'd better hope 
the application designer did not use ms sql server specific optional sql 
commands like transactions with special syntax and options setting using sql.


 Geoff.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: recvmsg on dgram socketpair blocks on open socket?

2007-02-16 Thread Tzahi Fadida
Is there an added value in contrast of just using a simple
server that accepts on low ports but bounces the packets
to a low privileged port?
Also, again of curiosity :), is there a way to wrap the daemon
without forking and replacing the bind call with a customized bind
with a more detailed security preferences.
Another method perhaps, is to insert a module into the kernel
which decorates bind with the capability to identify a process
(for example using sysfs or something) to have a free hand with binding.

On Friday 16 February 2007 12:49, Shachar Shemesh wrote:
 Tzahi Fadida wrote:
  On Friday 16 February 2007 03:07, you wrote:
  Hi,
 
  I'm trying to help complete Shachar Shemesh' privbind project (
  http://sourceforge.net/projects/privbind) and it mostly works except
  that
 
  Just from curiosity, what does it mean privileged socket? why are they
  privileged for internal use.
  IIRC for tcp the ports under 1024 are protected to be listened
  by non-root users (i think)? is that it?

 Yes.

 The unix domain sockets are merely used to pass the TCP/UDP sockets' FD
 between non-privileged daemon and privileged privbind so that the
 later can bind them to the low port for the former.

  In addition, what would be a typical application of privbind?

 Running a daemon that requires binding to a low port, but requires no
 other privileged permission.

  10x.

 Shachar

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: entering hebrew?

2007-02-14 Thread Tzahi Fadida
You need to switch to a hebrew keyboard.
If you have xorg, then you'll probably have to edit it and add/edit 
something like this to /etc/X11/xorg.conf:

Section InputDevice
  Identifier Generic Keyboard
  Driver kbd
  option CoreKeyboard
  option XkbRules xorg
  option XkbModel pc105
  option XkbLayout us,il
  option XkbOptions grp:switch,grp:alt_shift_toggle,grp_led:scroll
EndSection

Restart X.
Then use alt-shift to switch keyboard to hebrew.

On Wednesday 14 February 2007 09:55, Dan Bar Dov wrote:
 How do I input hebrew on Linux?
 More specifically, in firefox (2.0).

 Thanks,
 Dan

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: entering hebrew?

2007-02-14 Thread Tzahi Fadida
On Wednesday 14 February 2007 11:01, Julian Daich wrote:
 El mié, 14-02-2007 a las 10:39 +0200, Tzahi Fadida escribió:
  You need to switch to a hebrew keyboard.
  If you have xorg, then you'll probably have to edit it and add/edit
  something like this to /etc/X11/xorg.conf:

 In many distros if you are using a Gnome or KDE all the keyboard layouts
 you do not need to make changes at /etc/X11/xorg.conf to switch the
 keyboard and it is not recommended to play with /etc/X11/xorg.conf if
 you have not experience with it. In these cases you can change the
 keyboard layout from the desktop settings or adding the keyboard
 applet( as in Windows or Mac) to your toolbar and /etc/X11/xorg.conf
 will always set as¨ us¨ in the XkbLayout entry.

You are right, of course, but i wish it was true for my KDE on kubuntu.
The dang thing ignores ALT-SHIFT or any combination for that matter when
group switches are enabled. Or even more weirder, sometimes it switches once
and stops. This is before and after a total reinstall 2 weeks ago. Guess KDE 
does not like my keyboard.
So, in the event KDE configuration (System settings-Regional settings in KDE) 
won't succeed for our friend here, then he can do it in xorg.conf


  Section InputDevice
Identifier Generic Keyboard
Driver kbd
option CoreKeyboard
option XkbRules xorg
option XkbModel pc105
option XkbLayout us,il
option XkbOptions grp:switch,grp:alt_shift_toggle,grp_led:scroll
  EndSection
 
  Restart X.
  Then use alt-shift to switch keyboard to hebrew.
 
  On Wednesday 14 February 2007 09:55, Dan Bar Dov wrote:
   How do I input hebrew on Linux?
   More specifically, in firefox (2.0).
  
   Thanks,
   Dan
  
   =
   To unsubscribe, send mail to [EMAIL PROTECTED] with
   the word unsubscribe in the message body, e.g., run the command
   echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Learning kernel modules development

2007-02-14 Thread Tzahi Fadida
The problem with NFS Root is that i don't have a second machine :(.
Anycase, it took a while and i got a uml to work on a debootstrap ubuntu edgy
with network support and no super user for the helpers so i won't make a 
mistake (took half a day just for that).
Got my hello world module though so that is something.
Though there is an irritating thing with ubuntu where the 6 virtual consoles
pop out to the desktop and cannot be closed. haven't figured out how to kill 
them yet. and that the console is not popping up on the shell where the uml 
loads. As for the performance, it loads pretty quick and i am not running 
servers here so it is ok for learning purposes.

The only problem i understand there is with UML is that you cannot work with 
real devices (since the kernel is talking to the host kernel. I imagine this 
is the same with vmware/etc emulation.

On Tuesday 13 February 2007 20:02, Gilboa Davara wrote:
 On Tue, 2007-02-13 at 18:54 +0200, Tzahi Fadida wrote:
  Hi,
  As you might guess from a previous thread, i am trying to learn modules
  development. I use kubuntu as my dist.
  My first goal is to be able to run the hello world module example in the
  linux device drivers 3rd edition book.
  Apparently in 2.6 you need to have the whole kernel compiled just to
  build one module, which is not a problem though i don't want to start
  spending days tuning it up to work correctly with my system.

 You don't.
 You just need the skeleton kernel-devel package. (A cut down kernel
 image that includes the kernel configuration, headers and kbuild
 environment)

  What should be my workspace so i can start learning:
  User mode linux? qemu? vmware? XEN :) ?

 In general, I use both VMWare server and an NFS-root-mounted machine.
 Xen: At least in Fedora-land, it tends to be very sensitive to kernel
 updates - a very frequent event.
 QEMU: Too slow. (Even with now-GPL'ed kernel module.)
 KVM: No hardware support. (Old[er] dual core Opterons @work/home)
 UML: it has been a while since I used it - but even with the SKAS
 host-patch, performance is less then staggering and it was -very-
 sensitive to the base (read: un-patched) kernel versions.

 NFS-root:
 The NFS-root is also an interesting option. (If you can spare a second
 machine/SBC/embedded board)
 A. You don't risk the host machine. (I did crash vmware/Xen a couple of
 times - risking the integrity of the host machine)
 B. The guest machine is booting from a network FS - no matter how
 painful your crash is (Read: doing memset with a negative size) you
 won't damage the guest OS image.
 C. Super fast boot. My NFS test machine (with a modified stripped-down
 Slackware) boot in less then 15 seconds.

  P.s.: naturally i am impatient to start coding yesterday :)

 Yep... Debugging kernel OOPs is pure joy ;)

 - Gilboa


 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]


=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Learning kernel modules development

2007-02-14 Thread Tzahi Fadida
10x for the tip. Mostly solved it by attaching the consoles to pty.

On Wednesday 14 February 2007 23:20, Yedidyah Bar-David wrote:
 On Wed, Feb 14, 2007 at 10:10:52PM +0200, Tzahi Fadida wrote:
  The problem with NFS Root is that i don't have a second machine :(.
  Anycase, it took a while and i got a uml to work on a debootstrap ubuntu
  edgy with network support and no super user for the helpers so i won't
  make a mistake (took half a day just for that).
  Got my hello world module though so that is something.
  Though there is an irritating thing with ubuntu where the 6 virtual
  consoles pop out to the desktop and cannot be closed. haven't figured out
  how to kill

 Simply comment out some (or all) of the ttys in your /etc/inittab (of
 the uml).

  them yet. and that the console is not popping up on the shell where the
  uml

 I managed to solve it, a few years ago when I played with uml, but can't
 recall details. A combination of playing with /etc/inittab, /dev/console
 and the kernel command line (console=stuff specifically). Read the uml
 howto, available in your kernel sources under Documentation/uml.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Learning kernel modules development

2007-02-13 Thread Tzahi Fadida
Hi,
As you might guess from a previous thread, i am trying to learn modules 
development. I use kubuntu as my dist.
My first goal is to be able to run the hello world module example in the linux 
device drivers 3rd edition book.
Apparently in 2.6 you need to have the whole kernel compiled just to build one 
module, which is not a problem though i don't want to start spending days 
tuning it up to work correctly with my system.

What should be my workspace so i can start learning:
User mode linux? qemu? vmware? XEN :) ? 

P.s.: naturally i am impatient to start coding yesterday :)

10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Looking for a kernel book recommendation.

2007-02-09 Thread Tzahi Fadida
On Friday 09 February 2007 00:55, you wrote:

 regarding fiber-channel communications - are you interested in
 fiber-channel, or in SCSI-over-fiber-channel? these are completely
 distinct issues, and in the storage realm, you mostly deal with SCSI.
 the HBA driver does the fiber-channel related communications and (tries
 to) hide that from you.

I'll have to say fiber-channel because i am told that the device i am 
interested in does translation into SCSI (perhaps they said iSCSI, i'll be 
told more next week) over the fiber channel but in normal operations it uses 
the fiber channel to communicate. 

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Looking for a kernel book recommendation.

2007-02-08 Thread Tzahi Fadida
Hi,
I am looking for a book that will help me get more into programming in the 
kernel. Specifically in the storage areas and kernel programming etiquette.
Perhaps something current like for kernel 2.6?
Any recommendations?
10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Looking for a kernel book recommendation.

2007-02-08 Thread Tzahi Fadida
10x.
I'll check it out.
I am also looking for the storage stuff, like how the fs is built and how to 
build one in 2.6 and some docs concerning fiber channel communications 
between linux machines.

P.s. and yes i was at your lecture on fs :)

On Thursday 08 February 2007 18:44, you wrote:
 On Thu, 8 Feb 2007, Tzahi Fadida wrote:
  I am looking for a book that will help me get more into programming in
  the kernel. Specifically in the storage areas and kernel programming
  etiquette. Perhaps something current like for kernel 2.6?
  Any recommendations?

 linux device drivers, 3rd edition. i think it's the only book that teaches
 you how to write kernel code and how to use the kernel's internal API.

 most other books are theoretical coverage or code commentary - they are
 not very useful as first books on this subject.

 after this - start reading the kernel's source ;)

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Looking for a kernel book recommendation.

2007-02-08 Thread Tzahi Fadida
Cool!. 
Assuming this is legal.
10x!

On Thursday 08 February 2007 21:03, Constantine Shulyupin wrote:
 http://linuxdriver.co.il/ldd3/
 Love, Robert. Linux Kernel Development.

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Enabling powersave for an lcd screen

2007-02-04 Thread Tzahi Fadida
The problem is not to turn the screen blank but to get it
to go to standby (powersave). 
I tried:
xset +dpms
xset dpms force off
xset dpms force standby
xset dpms force suspend

and they only blank the screen.
This is why i think acpi/apm are required to enable the lcd screen LG flatron 
l1910s to enter powersaving mode instead of just making the screen black.

On Sunday 04 February 2007 08:17, Oren Held wrote:
 I hope I'm not mistaken, but I think you don't need acpi/apm support for
 power save modes of monitors.
 Try xset dpms force off in X, and see if it works.

 xset +dpms should turn on dpms support for current X session.
 Enabling dpms  setting timeouts can be done either from xorg.conf
 (search for 'offtime' option) or from kde/gnome gui configuration tools.

 Tzahi Fadida wrote:
  Hi,
  A while ago i had a system crash and i had to reinstall kubuntu from
  scratch. The powersave before worked and it also work now in windows.
  However, i can't seem to get it to work on the new installation.
  both acpid and apmd do not work.
  sudo acpid start
  acpid: can't open /proc/acpi/event: Device or resource busy
  sudo apmd start
  No APM support in kernel
 
  I have kubuntu edgy updated.
  Bios have apm enabled.
 
  I also tried in the boot process to add acpi=force apm=off.
  and acpi=off apm=force, though this method did not load the kernel.
 
  I also tried installing powersaved.
 
  What else can i do?
  10x.
 
  p.s.: i started with dapper and upgraded to edgy with adept.
  and the kernel is
  Linux Linux 2.6.17-10-386 #2 Tue Dec 5 22:26:18 UTC 2006 i686 GNU/Linux

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Enabling powersave for an lcd screen

2007-02-04 Thread Tzahi Fadida
Cancel my last.
Adding Option DPMS to my monitor in xorg did the trick.
Section Monitor
  identifier L1910S
  Option DPMS
  ...

It seems that xorg and kde are not fully synced with their settings. The same 
thing happened, for example, when i wanted to switch keyboard maps between 
english and hebrew with alt-shift etc... In xorg.conf it works but in KDE 
regional settings the switching groups did not. Annoying.

Thanks for the help!

On Sunday 04 February 2007 11:09, Tzahi Fadida wrote:
 The problem is not to turn the screen blank but to get it
 to go to standby (powersave).
 I tried:
 xset +dpms
 xset dpms force off
 xset dpms force standby
 xset dpms force suspend

 and they only blank the screen.
 This is why i think acpi/apm are required to enable the lcd screen LG
 flatron l1910s to enter powersaving mode instead of just making the screen
 black.

 On Sunday 04 February 2007 08:17, Oren Held wrote:
  I hope I'm not mistaken, but I think you don't need acpi/apm support for
  power save modes of monitors.
  Try xset dpms force off in X, and see if it works.
 
  xset +dpms should turn on dpms support for current X session.
  Enabling dpms  setting timeouts can be done either from xorg.conf
  (search for 'offtime' option) or from kde/gnome gui configuration tools.
 
  Tzahi Fadida wrote:
   Hi,
   A while ago i had a system crash and i had to reinstall kubuntu from
   scratch. The powersave before worked and it also work now in windows.
   However, i can't seem to get it to work on the new installation.
   both acpid and apmd do not work.
   sudo acpid start
   acpid: can't open /proc/acpi/event: Device or resource busy
   sudo apmd start
   No APM support in kernel
  
   I have kubuntu edgy updated.
   Bios have apm enabled.
  
   I also tried in the boot process to add acpi=force apm=off.
   and acpi=off apm=force, though this method did not load the kernel.
  
   I also tried installing powersaved.
  
   What else can i do?
   10x.
  
   p.s.: i started with dapper and upgraded to edgy with adept.
   and the kernel is
   Linux Linux 2.6.17-10-386 #2 Tue Dec 5 22:26:18 UTC 2006 i686 GNU/Linux

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Enabling powersave for an lcd screen

2007-02-03 Thread Tzahi Fadida
Hi,
A while ago i had a system crash and i had to reinstall kubuntu from scratch.
The powersave before worked and it also work now in windows.
However, i can't seem to get it to work on the new installation.
both acpid and apmd do not work.
sudo acpid start
acpid: can't open /proc/acpi/event: Device or resource busy
sudo apmd start
No APM support in kernel

I have kubuntu edgy updated.
Bios have apm enabled.

I also tried in the boot process to add acpi=force apm=off.
and acpi=off apm=force, though this method did not load the kernel.

I also tried installing powersaved.

What else can i do?
10x.

p.s.: i started with dapper and upgraded to edgy with adept.
and the kernel is
Linux Linux 2.6.17-10-386 #2 Tue Dec 5 22:26:18 UTC 2006 i686 GNU/Linux

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: archiving parts in linux.

2007-02-01 Thread Tzahi Fadida
On Tuesday 30 January 2007 16:19, Yehoshua (Shay) O'Hayon Suchar wrote:
 Tzahi Fadida said:
  I tried 7z but after about 12% of archiving it gives a weird error:
  Error:
 
 
  System error:
  Too many open files 
  (and this file was never opened :).
 
  What can i do?

 Could be that the error has something to do with the limit of file
 descriptors in the system?

Sorry for not responding earlier. My system crashed and i had to restore from 
backups. 
I did not really understand this. Are you saying there is a limit on the 
number of files i can tar gz?
I also make backups of my emails and there can be as much as a million files, 
pherhaps more because of the mbox style(i think) of holding mails.


 Check out this: http://bcr2.uwaterloo.ca/~brecht/servers/openfiles.html



 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: archiving parts in linux.

2007-02-01 Thread Tzahi Fadida
On Thursday 01 February 2007 14:55, Dotan Cohen wrote:
 On 01/02/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  On Tuesday 30 January 2007 16:19, Yehoshua (Shay) O'Hayon Suchar wrote:
   Tzahi Fadida said:
I tried 7z but after about 12% of archiving it gives a weird error:
Error:
   
   
System error:
Too many open files 
(and this file was never opened :).
   
What can i do?
  
   Could be that the error has something to do with the limit of file
   descriptors in the system?
 
  Sorry for not responding earlier. My system crashed and i had to restore
  from backups.
  I did not really understand this. Are you saying there is a limit on the
  number of files i can tar gz?
  I also make backups of my emails and there can be as much as a million
  files, pherhaps more because of the mbox style(i think) of holding mails.

 Somebody please correct me if I'm wrong, but so far as I know mbox
 saves all the emails of a particular folder in one big file. So if you
 have 200 emails in 5 folders, then you will have 5 mbox files.

 Kmail has another two or three files per mbox folder, so you really
 will have 5*n folders, but you get the idea. It's not 200*n.

Sorry, of course maildir.


 Dotan Cohen

 http://what-is-what.com/what_is/firefox.html
 http://technology-sleuth.com/question/what_is_a_firewall.html

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



archiving parts in linux.

2007-01-28 Thread Tzahi Fadida
Hi,
I wish to archive a 50mb file into several parts of 5mb files in linux.
I also need to be able to open the files in windows.
For some unknown reason, i can't find a way (other than using a non-free rar) 
to do this.
I tried zipsplit but it says something about the file being too large or 
something like this. 

I tried 7z but after about 12% of archiving it gives a weird error:
Error:


System error:
Too many open files 
(and this file was never opened :).

What can i do?

Also, i am trying to backup my files (many thousands of them). is there a way 
to cause tar or something else not to use a local tmp directory (since i 
don't have enough room) and also, it seems to eat too much memory till my kde 
programs starts to close.
10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Why the Internet sucks

2007-01-18 Thread Tzahi Fadida
Just a little correction: IIRC YES moved StarWorld to channel 270.
(Just a little FYI to all the Trekkies out there :) ).

On Thursday 18 January 2007 12:13, Geoffrey S. Mendelson wrote:
 In case you haven't been following the English press, both HOT and YES
 have dropped English language channels. This has caused an uproar among
 the English speaking (aka Anglo) community here.

 I was interviewd by a reporter for Haaretz, and also commented in my blog,
 that now that people were angry at both they would be changing the way
 they spent their entertainment shequels. I also commented that anyone
 who had an Internet connection would quickly find out that they
 could bootleg the programs for free without YES or HOT and this would
 really clog the Internet.

 While I did not tell her how to bootleg the programs, as assuming I
 did know how, it would not be appropriate for me to tell anyone.
 If someone was interested they would have to find out on their own.

 The Jerusalem Post was far more forthcomming, they listed URLs in their
 articles.

 I have heard that 70-80 percent of the Internet's traffic was BitTorrent
 files, and most of them were bootlegs. After this latest increase in
 downloading to replace what was free, but is no longer available,
 I expect that almost all of the traffic in Israel is downloading bootleg
 files, most of it TV programs that people used to watch on BBC Prime
 (dropped by HOT) or Star World (dropped by YES).

 In plain English, there goes the neighborhood. :-(

 Geoff.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Why the Internet sucks

2007-01-18 Thread Tzahi Fadida
On Thursday 18 January 2007 19:59, Diego Iastrubni wrote:
 ביום חמישי 18 ינואר 2007, 12:38, נכתב על ידי Tzahi Fadida:
  Just a little correction: IIRC YES moved StarWorld to channel 270.
  (Just a little FYI to all the Trekkies out there :) ).

Sorry, did not know that.
I guess it is lucky i just moved to HOT.


 Which days and hours? :)

Saturday at 19:00 there is the last (probably ever) season of Enterprise.


 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Constraining an via rhine to 10mb full duplex

2007-01-07 Thread Tzahi Fadida
10x, it worked and you are right, i looked at the updated
source code and it says they removed the options in some version.
I probably looked at an older version of the code using google.

Anyway, just for future refernce, the line should be something like this:
iface eth0 inet dhcp 
pre-up /usr/sbin/ethtool -s $IFACE speed 10 duplex full
...

10x.

On Sunday 07 January 2007 10:34, you wrote:
 On 1/6/07, Tzahi Fadida [EMAIL PROTECTED] wrote:
  Can someone enlighten me to how to constrain my on board ethernet
  card to 10Mb full duplex (due to cable problems only this speed works).
  I can use mii-tool to force it but i prefer some kind of options in
  modules.conf or conf.modules whichever works.
  I am using kubuntu edgy and via-rhine.ko is a module in my kernel.

 From 'modinfo via-rhine' , I can see no module parameter to influence it.

 You'd have to use mii-tool or ethtool to configure it - this is the
 proper way. You may add it to your /etc/network/interfaces:
  iface eth0 inet static
  pre-up /usr/sbin/ethtool -s $IFACE 10 duplex half
  address ...

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Constraining an via rhine to 10mb full duplex

2007-01-06 Thread Tzahi Fadida
Can someone enlighten me to how to constrain my on board ethernet
card to 10Mb full duplex (due to cable problems only this speed works).
I can use mii-tool to force it but i prefer some kind of options in 
modules.conf or conf.modules whichever works.
I am using kubuntu edgy and via-rhine.ko is a module in my kernel.

lspci output:
00:12.0 Ethernet controller: VIA Technologies, Inc. VT6102 [Rhine-II] (rev 70)

10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: IO stats per process

2006-12-20 Thread Tzahi Fadida
Check out the archives.
We discussed this in the past.
IIRC, you can use echo 1  /proc/sys/vm/block_dump 
(it might also involve /proc/sys/vm/laptop_mode but i can't remember why).
and then tail /proc/kmsg to see the dump of the processes i/o requests.

On Wednesday 20 December 2006 12:32, Danny Lieberman wrote:
 guys

 How do I get the real IO (block reads/writes per second, not cached) of
 each process on a running Linux system?
 vmstat and iostat dont provide process level detail

 I'm looking at a system right now which is CPU idle but very IO busy and
 I cant figure out who's moving the disks around.


 I thought this would be the most trivial questionmaybe it is

 10x

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: FireFox and Bank haPoalim

2006-12-08 Thread Tzahi Fadida
Just being a bit argumentative, you have 2 issues with this claim:
1) You can't tell by that logic if the zip has already been cracked.
The many eyes principle says that only you is probably not enough.
Now, if it was tab mix plus... :)
2) 6 months from now, when it will ask for an update, someone will
probably forget the original reasoning for installing an unsigned xpi
and will update the file.

On Friday 08 December 2006 08:07, Zvi Har'El wrote:
 Once you installed it, it really doesn't matter if somebody later breaks
 to Effie's site (to Effie it would matter :'( ). If you haven't
 installed it yet - you can download it, and examine the file contets (an
 XPI file is just a fancy extension to a ZIP archive) before installing
 it (from the local file, of course).  I would do it even if the file
 were signed, and even if it came by a stork directly from mozdev :-) .

 Amos Shapira wrote, On 07/12/06 23:28:
  On 08/12/06, *Oded Arbel* [EMAIL PROTECTED]
  mailto:[EMAIL PROTECTED] wrote:
 
  On Thu, 2006-12-07 at 18:38 +0200, Ilya Konstantinov wrote:
   While I agree this is risky, I must correct your assumption that
   the fact this extension claims to deal with a banking site makes it
   more or less likely to be spyware.
 
  I think the main issue is that it's unsigned - it means that a cracker
  who breaks into Effie's web site and installs a modified version won't
  be caught by the extension singing mechanism.
  I might be naive, but I pretty much trust Effie himself not to do any
  monkey business with his own extensions.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: FireFox and Bank haPoalim

2006-12-07 Thread Tzahi Fadida
Just a warning,
Installing an unsigned plugin that affects the connection with your bank is 
not very smart (i remembered late myself, so don't feel bad).

On Thursday 07 December 2006 13:00, Michael Sternberg wrote:
 Hello.

 For everybody who's tired of looking on reversed hebrew of Bank haPoalim
 personal account web interface I have found a wonderful extension for
 FireFox that solves this problem:

 http://www.effie.co.il/mozvuvu/home.html

 enjoy :)

 P.S. I do not understand why Bank haPoalim still did not fixed their
 problem with reverse Hebrew in FireFox, its there for at least three
 years..

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: FireFox and Bank haPoalim

2006-12-07 Thread Tzahi Fadida
Let me refine this. Very much like linux kernel is more or less trusted
to have no secret evil code because of the many eyes principle, so does,
for example, tab mix plus.
However, an obscure plugin like this bank hapoalim plugin does not have many 
eyes on its code and thus, more dangerous than other more famous plugins.

On Thursday 07 December 2006 18:38, Ilya Konstantinov wrote:
 There's nothing in this extension which limits it to work only with
 the bank's site, nor is there anything to limit any other extension
 you've installed from stealing your data from any site.

 While I agree this is risky, I must correct your assumption that the
 fact this extension claims to deal with a banking site makes it more
 or less likely to be spyware.

 On 12/7/06, Tzahi Fadida [EMAIL PROTECTED] wrote:
  Just a warning,
  Installing an unsigned plugin that affects the connection with your bank
  is not very smart (i remembered late myself, so don't feel bad).

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Orange SMS

2006-11-19 Thread Tzahi Fadida
If icq can send to orange (and i think it can) you can use the script i use
to send sms using icq. It is called vicq and it is a console script.

On Sunday 19 November 2006 14:00, Amichai Rotman wrote:
 Hi all,

 I need so GUI solution to send messages inside the Orange network.

 It dosn't have to be in Hebrew.

 I am at work every day from 08:00 to 17:00 and my wife is at home. As you
 know, we can't use Orange's SMS sending page - dosn't work with Firefox...

 Any other solution? She is tired of SMSing from the phone, so she calls me
 voice...

 Thanks!

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: tnef winmail.dat

2006-11-17 Thread Tzahi Fadida
Login to your google account on the web and download it from there? :)

On Friday 17 November 2006 10:12, Shlomo Solomon wrote:
 It's an unfortunate fact of life that we all sometimes get Microsoft only
 attachments. KMail solves this with a utility called ktnef, so the problem
 doesn't usually bother me.

 But I just got one of these attachments and ktnef only partially opened it.
 The attachment was over 500K and ktnef only extracted a 60K .gif file. So I
 asked the sender what he had sent me and was told that I should have
 received the .gif file and also a larger .jpeg file. I then tried two other
 command line utilities (tnef and opentnef). They both did the same thing as
 ktnef.

 I Googled and found cases of tnef not working, but not of partial
 extraction. Any ideas?

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Limiting the number of simultaneous HTTP connection per IP

2006-10-10 Thread Tzahi Fadida
You can do that on netfilter using iptables but i suggest caution.
You see, there are many organizations that uses NAT or MASQUERADING
so, to netfilter it will look like you are getting flooded.
What you actually need is to identify somehow that some specific client
opened the connection and limit it to let's say 5 connections. 

On Tuesday 10 October 2006 12:23, Sagi Bashari wrote:
 Hi List,

 We've recentely had trouble with some misbehaved web clients that opened
 dozens of HTTP connections to our web server, causing it to reach the total
 connection limit and just hang until they timeout or until the server is
 restarted.

 We're sure that this is not an intentional DoS attack and these clients
 will probably be fixed, but I would like to prevent the possibility of such
 attacks in the future, intentional or otherwise.

 I managed to replicate such attack against our server by running a trivial
 script on my workstation:
 for i in `seq 100`; do (nc HOST 80 ); done

 Our servers are running Apache/2.0.54 on Debian Sarge.

 There are many Apache modules that aim to solve such problems. I've tested
 a few, and they all seem to not prevent it completely. These modules wait
 until the client sends a complete request and only then check if it should
 be blocked, serving Apache error page. They don't take any action if the
 client just opens a TCP connection and leaves it hanging, for example.

 I'm looking for a way to prevent such attack in a higher level, before it
 even reaches Apache. I found a iptables module named connlimit/iplimit,
 that is supposed to do just that, but it seems the official kernels do not
 support it and there's a serious lack of information about it.

 I guess I'm not the only one who experienced such problems and there must
 be a better known solutions.

 Please advice,
 Sagi

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Contacts program

2006-09-29 Thread Tzahi Fadida
I guess my best bet is to build it myself. In access i can do this in an hour 
or so. Of course, it would not be so integrated just a container for the 
contacts and meeting logs and not very customizable.
The problem is that i am in linux, not windows.

Perhaps Base can do that too, but my previous experience with it was that
its normal operation is to crash and also that it is bloated in memory and
takes forever to start.

Another option is to use FreeMind to log the contacts but i don't particularly 
excited about it. I use FreeMind to store my knowledgebases, todos and ideas 
etc... Though i feel less comfortable to turn it into some kind of database.


On Thursday 28 September 2006 12:26, you wrote:
 On Thu, 2006-09-28 at 09:25 +0300, Tzahi Fadida wrote:
  In Kontact, the events are very loosely coupled to a contact and
  it is more on the context of sending the contact an email. More so, i
  can't find a search criteria for an event on an attendee.

 That's bad. I recommend opening a feature request (wishlist) ticket in
 the KDE bugzilla on that - my impression is that Kontact developers are
 very forthcoming towards feature requests that make sense, like the
 above one.

  For example, in ACT you can automatically setup a
  rule that will alert you to call a client after a few months so you won't
  loose touch with that person.

 That's a really nice feature. Sadly, probably due to never having
 thought of needing such feature, I'm not familiar with any calendaring
 software that does this, unless you specifically schedule an event for
 such. BTW - if you're interested in checking yet another calendaring
 software I would recommend looking at the Chandler alphas (I've had some
 success with .7a2, but not so much with .7a3) - it has some interesting
 features, and it might have your pet features :-)

 Also, there are many many many web based open-source calendaring
 applications (or even full CRM suits) which I would normally not
 recommend for someone looking for a desktop application, but might come
 in handy until a desktop calendar with all of the features you require
 comes along.

 --
 Oded

 ::..

 There's only one way to have a happy marriage and as soon as I learn
 what it is I'll get married again.
 -- Clint Eastwood

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Contacts program

2006-09-28 Thread Tzahi Fadida
In Kontact, the events are very loosely coupled to a contact and 
it is more on the context of sending the contact an email. More so, i can't 
find a search criteria for an event on an attendee.
What i need is something like ACT which is a mini CRM basically. Each contact 
can be directly assigned a meeting log with attachments etc...
It is not that Kontact is bad. It is very good and convenient to hold contacts 
but it fails to understand the concept of keeping in touch with 
clients/colleagues/friends. For example, in ACT you can automatically setup a 
rule that will alert you to call a client after a few months so you won't 
loose touch with that person.

On Thursday 28 September 2006 01:54, you wrote:
 On Wed, 2006-09-27 at 14:25 +0300, Tzahi Fadida wrote:
  On Wednesday 27 September 2006 13:01, you wrote:
   I thought that meetings in Kontact/Evolution already are linked with
   contacts - I add the contacts I want to be on the meeting, and it even
   mails them invitations and updates.
   Regarding summaries, I'm using Evolution and I can open the meeting
   information dialog and I can add attachments to the meeting, IIRC
   Kontact has a similar feature as well. Or am I missing something
   important ?
 
  The problem with evolution is that it crashes when i click on new
  meeting. Basically, it crashed from day 1 when i installed ubuntu (main
  reason why i upgraded to kubuntu).

 Evolution 2.8 has some crashes in the Calendar code - there is an open
 bug about it. For me it works as long as I don't try to add email
 attachments to meetings, or anything that fancy :-) normal attachments
 usually work.

  There is no meeting log in kontact that i can find.
  Perhaps some plugin? there is a place to write some notes in clear text,
  but certainly not to add attachments.

 Meetings in Kontact are called events, and you can add attachments as
 you wish to them (last tab on the dialog). You can also drag an email to
 the calendar icon on the shortcut bar to create a new event with the
 email attached (which is the main calendar oriented feature I'm missing
 from Evolution). While the description field in Kontact events or
 Evolution meetings can be used to type any amount of text, I wouldn't
 use that for a message log. Currently I open a new text file, type what
 I want, save it and then attaches it to the meeting. with Evolution I
 can then remove the original file, while with Kontact I have to keep it
 stored, because it only keeps links to attachments. It still a lengthy
 and annoying process though.

 --
 Oded

 ::..

 Scully: You know...Most people wanna live forever
 Felig: Most people are idiots
 -- The -X- Files

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Build processes

2006-09-28 Thread Tzahi Fadida
I am starting to understand now that for that particular job i need to be 
ultra proficient with RCS and RCS related tools and how it can be used with 
scripts like rsh based shell scripts, perl, etc...
(before anyone starts, yes i know RCS is also ultra limited, but that's what 
happens with legacy huge projects that no one wants to revise now).
Are there some good manuals, books, articles, anything out there that can give 
me the edge on this?
10x.

On Wednesday 27 September 2006 19:10, Shlomi Fish wrote:
 On Wednesday 27 September 2006 14:34, Alexander Indenbaum wrote:
  Tzahi,
 
  I could recommend classic paper Recursive Make Considered Harmful
  http://aegis.sourceforge.net/auug97.pdf

 Other links:

 * What is wrong with make? - http://freshmeat.net/articles/view/1702/

 * Make Alternatives - http://freshmeat.net/articles/view/1715

 * Stop the Autoconf Insanity - http://freshmeat.net/articles/view/889/

 * The Software Carpentry Project - http://www.software-carpentry.com/

 Regards,

   Shlomi Fish

  It describes a way to build Makefiles with automatic dependencies,
  meaning no separate make dep target. Non-recursive Makefile builds
  *exactly* needed files and only them. Saves tones of time and
  resulting binaries always consistent, meaning no need to do make
  clean just to be sure.
 
  The main idea behind it *not* using industry standard practice of main
  Makefile recursively running make in subdirectories. Instead single
  Makefile should be used which includes subdirectory's Makefiles. It
  makes a difference, because this way make program has a full picture
  of project's dependency graph.
 
  This model is clearly a beauty if you ask me. It is simple and it works.
 
  Alexander Indenbaum
 
  On 9/27/06, Tzahi Fadida [EMAIL PROTECTED] wrote:
   Hi,
   I wish to learn more about build process/es. Specifically in C.
   Background: i am potentially (/trying to) entering the field of
   CM - Configuration Management. I am told a main part of that
   is the build process, specifically in the area of dependencies.
   I am talking about such jobs in very big companies.
   I understand that in such companies, builds can run for days if
   not setupped/optimized correctly.
  
   Obviously i compiled my own projects, which had dependencies in the
   code. :) I am guessing that it is much more than just building small
   libraries.
  
   Where can i find information that will show me how this is done
   and what i need to know. Some tips...
   10x.
  
   --
   Regards,
   Tzahi.
   --
   Tzahi Fadida
   Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
   WARNING TO SPAMMERS: see at
   http://members.lycos.co.uk/my2nis/spamwarning.html
  
   To unsubscribe, send mail to [EMAIL PROTECTED] with
   the word unsubscribe in the message body, e.g., run the command
   echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Build processes

2006-09-27 Thread Tzahi Fadida
Hi,
I wish to learn more about build process/es. Specifically in C.
Background: i am potentially (/trying to) entering the field of
CM - Configuration Management. I am told a main part of that
is the build process, specifically in the area of dependencies.
I am talking about such jobs in very big companies.
I understand that in such companies, builds can run for days if
not setupped/optimized correctly.

Obviously i compiled my own projects, which had dependencies in the
code. :) I am guessing that it is much more than just building small
libraries.

Where can i find information that will show me how this is done
and what i need to know. Some tips...
10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Contacts program

2006-09-27 Thread Tzahi Fadida
On Wednesday 27 September 2006 13:01, you wrote:
 --=-WIkWkV5dFzHgEB7+1zmN
 Content-Type: text/plain
 Content-Transfer-Encoding: 7bit

 On Mon, 2006-09-25 at 23:12 +0300, Gil Freund wrote:
  On 9/25/06, Tzahi Fadida [EMAIL PROTECTED] wrote:
   I am looking for a contacts program (for linux) that could replace
   the one that i was working with - symantec ACT, or saga ACT.
   Basically, it also has the ability to also log meeting summaries, files
   and phone calls. It can also assign contacts under categories, such
   friends, business, etc...
 
  I use Kontact, and by the look of things, Evolution has similar
  functionality. The only thing I miss from my old Ecco days is the linking
  of meetings and contacts (now fondly called CRM). Kontact has a
  journal
  feature, but I don't think this quite meets the bill.

 I thought that meetings in Kontact/Evolution already are linked with
 contacts - I add the contacts I want to be on the meeting, and it even
 mails them invitations and updates.
 Regarding summaries, I'm using Evolution and I can open the meeting
 information dialog and I can add attachments to the meeting, IIRC
 Kontact has a similar feature as well. Or am I missing something
 important ?

The problem with evolution is that it crashes when i click on new meeting.
Basically, it crashed from day 1 when i installed ubuntu (main reason why i 
upgraded to kubuntu). There is no meeting log in kontact that i can find.
Perhaps some plugin? there is a place to write some notes in clear text, but 
certainly not to add attachments.



 Also, both Kontact and Evolution support assigning contacts and meetings
 into (multiple) categories.

 --
 Oded

 ::..

 This signature was intentionally left boring


 --=-WIkWkV5dFzHgEB7+1zmN
 Content-Type: text/html; charset=utf-8
 Content-Transfer-Encoding: 7bit

 !DOCTYPE HTML PUBLIC -//W3C//DTD HTML 4.0 TRANSITIONAL//EN
 HTML
 HEAD
   META HTTP-EQUIV=Content-Type CONTENT=text/html; CHARSET=UTF-8
   META NAME=GENERATOR CONTENT=GtkHTML/3.12.0
 /HEAD
 BODY
 On Mon, 2006-09-25 at 23:12 +0300, Gil Freund wrote:
 BLOCKQUOTE TYPE=CITE
 PRE
 FONT COLOR=#00On 9/25/06, Tzahi Fadida lt;A
 HREF=mailto:[EMAIL PROTECTED][EMAIL PROTECTED]/Agt; wrote:/FONT
 FONT COLOR=#00gt; I am looking for a contacts program (for linux)
 that could replace/FONT FONT COLOR=#00gt; the one that i was
 working with - symantec ACT, or saga ACT./FONT FONT COLOR=#00gt;
 Basically, it also has the ability to also log meeting summaries, files
 and/FONT FONT COLOR=#00gt; phone calls. It can also assign
 contacts under categories, such friends,/FONT FONT COLOR=#00gt;
 business, etc.../FONT

 FONT COLOR=#00I use Kontact, and by the look of things, Evolution
 has similar functionality./FONT FONT COLOR=#00The only thing I
 miss from my old Ecco days is the linking of meetings/FONT FONT
 COLOR=#00and contacts (now fondly called CRM). Kontact has a
 journal/FONT FONT COLOR=#00feature, but I don't think this quite
 meets the bill./FONT /PRE
 /BLOCKQUOTE
 BR
 I thought that meetings in Kontact/Evolution already are linked with
 contacts - I add the contacts I want to be on the meeting, and it even
 mails them invitations and updates.BR Regarding summaries, I'm using
 Evolution and I can open the meeting information dialog and I can add
 attachments to the meeting, IIRC Kontact has a similar feature as well. Or
 am I missing something important ?BR BR
 Also, both Kontact and Evolution support assigning contacts and meetings
 into (multiple) categories.BR BR
 TABLE CELLSPACING=0 CELLPADDING=0 WIDTH=100%
 TR
 TD
 --BR
 OdedBR

 ::..BR

 This signature was intentionally left boringBR
 BR
 /TD
 /TR
 /TABLE
 /BODY
 /HTML

 --=-WIkWkV5dFzHgEB7+1zmN--


 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Build processes

2006-09-27 Thread Tzahi Fadida
10x, i already started reading it, looks interesting.

I am also mainly interested in what is currently done, clear case, clear 
quest, cvs, svn, etc... and how all the configuration management components 
comes together. What kind of problems and ways to solutions a CM person have 
to deal with. Some case studies are appreciated, 
War stories are very welcome.
10x.

On Wednesday 27 September 2006 14:34, you wrote:
 Tzahi,

 I could recommend classic paper Recursive Make Considered Harmful
 http://aegis.sourceforge.net/auug97.pdf

 It describes a way to build Makefiles with automatic dependencies,
 meaning no separate make dep target. Non-recursive Makefile builds
 *exactly* needed files and only them. Saves tones of time and
 resulting binaries always consistent, meaning no need to do make
 clean just to be sure.

 The main idea behind it *not* using industry standard practice of main
 Makefile recursively running make in subdirectories. Instead single
 Makefile should be used which includes subdirectory's Makefiles. It
 makes a difference, because this way make program has a full picture
 of project's dependency graph.

 This model is clearly a beauty if you ask me. It is simple and it works.

 Alexander Indenbaum

 On 9/27/06, Tzahi Fadida [EMAIL PROTECTED] wrote:
  Hi,
  I wish to learn more about build process/es. Specifically in C.
  Background: i am potentially (/trying to) entering the field of
  CM - Configuration Management. I am told a main part of that
  is the build process, specifically in the area of dependencies.
  I am talking about such jobs in very big companies.
  I understand that in such companies, builds can run for days if
  not setupped/optimized correctly.
 
  Obviously i compiled my own projects, which had dependencies in the
  code. :) I am guessing that it is much more than just building small
  libraries.
 
  Where can i find information that will show me how this is done
  and what i need to know. Some tips...
  10x.
 
  --
  Regards,
  Tzahi.
  --
  Tzahi Fadida
  Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
  WARNING TO SPAMMERS: see at
  http://members.lycos.co.uk/my2nis/spamwarning.html
 
  To unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: setting up a cable modem under linux (barak)

2006-09-27 Thread Tzahi Fadida
perhaps what you need to do is to delete the old gateway using a script like 
so (it was customized for me and the bugs i used to get in my old dist, so it 
is not state of the art but...):
TODEL=`/sbin/route -n|/bin/grep eth1|/bin/grep UG|/usr/bin/cut -c17-32`
/sbin/route -n |/bin/grep $TODEL 21 /dev/null
if [ $? = 0 ]; then
/sbin/route del -net 0.0.0.0 gw $TODEL netmask 0.0.0.0 dev eth1

and then add the right gateway like so:
/sbin/route add default gw $NEWGW

On Thursday 28 September 2006 00:20, you wrote:
 I just switched from using adsl to cable (hot + barak) but I seem to be
 having
 problems setting up the connection.

 I couldn't find any l2tp client for debian (l2tpd seems to be a server and
 not
 a client).

 I did setup pptp. Eth0 comes up and gets an address using dhclient. At this
 point I can resolve ip address (ping find the ip address). I can also surf
 to
 the hot website. I then bring up ppp0 which comes up and gets an address.
 The
 default gw still points to the modem instead of the other side of the ppp
 connection and I can't surf. IPs are still resolved but very slowly.
 Changing
 the default gw doesn't help either.

 The modem is a terayon, the provider is barak, the windows connection uses
 l2tp and runs fine.



 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Contacts program

2006-09-25 Thread Tzahi Fadida
I am looking for a contacts program (for linux) that could replace
the one that i was working with - symantec ACT, or saga ACT.
Basically, it also has the ability to also log meeting summaries, files and 
phone calls. It can also assign contacts under categories, such friends, 
business, etc...

What is recommended, what is commonly used?
I plan to have several hundred contacts in it.

10x.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Contacts program

2006-09-25 Thread Tzahi Fadida
I know Ecco, i use it for my passwords :) it's the best hierarchical note 
taking. But, it's only for windows and its development is long long long 
dead. :)

On Monday 25 September 2006 23:12, Gil Freund wrote:
 On 9/25/06, Tzahi Fadida [EMAIL PROTECTED] wrote:
  I am looking for a contacts program (for linux) that could replace
  the one that i was working with - symantec ACT, or saga ACT.
  Basically, it also has the ability to also log meeting summaries, files
  and phone calls. It can also assign contacts under categories, such
  friends, business, etc...

 I use Kontact, and by the look of things, Evolution has similar
 functionality. The only thing I miss from my old Ecco days is the linking
 of meetings and contacts (now fondly called CRM). Kontact has a journal
 feature, but I don't think this quite meets the bill.

 There are on-line services which provide this, yahoo, plaxo and
 others, but you have to be online to use them...

 Gil

  What is recommended, what is commonly used?
  I plan to have several hundred contacts in it.
 
  10x.
 
  --
  Regards,
  Tzahi.
  --
  Tzahi Fadida
  Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
  WARNING TO SPAMMERS: see at
  http://members.lycos.co.uk/my2nis/spamwarning.html
 
  To unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Backup advice

2006-08-22 Thread Tzahi Fadida
On Monday 21 August 2006 11:54, you wrote:
 Hi all,
 I would like to get some advice for a backup utility with the following
 capabilities:

 1. Be able to snap-shot directories and databases (not the entire file
 system).

You need to specify what databases. Most databases need to be closed
before copied or you will have discrepancies in the data.
Not only that, if you will try to use the database files in the future
with even a slightly different version of the software it might not work.
You'd better run a script to dump all the database into a text file.
all databases can do that.

 2. Perform incremental backups (at least for directories)
 3. Good integrity checks
 4. Sane recovery process

 The main problem I have with rsync and friends is that I cannot get just
 the increment between two snapshots.

 Any recommendation will be much appreciated.

 Thanks,
 Ami

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: PPPoE on Linux - timeout waiting for PADO/PADS packets

2006-08-09 Thread Tzahi Fadida
Just a guess, but if inactivity or activity of the line has any affect on the 
line disconnecting then there are 2 possibilities i can see:
1) Line problems, i had this a few years back. The line disconnects every
once in a while for no apparent reason. The fact that you connect a mac
and it does not disconnect (or for the period of time you have tested it),
does not prove you do not have a line problem just that the mac tunnel program
is more resilient.
2) The tunnel program on your ubuntu just does not have an effective timeout
check. Somehow, perhaps because of your ISP etc..., the program sort-of pings
are not getting back. I would suggest to try to either increase the timeout or 
disable it. including a keepalive option (enable/disable) which surely be in 
the conf. for example, play with the following options:
CONNECT_POLL 
LCP_INTERVAL
LCP_FAILURE
and this one is very suspicious PPPOE_TIMEOUT
and i have this one (for some reason i can't recall) in my script: 
PPPD_EXTRA=lcp-echo-interval 60

On Tuesday 08 August 2006 21:07, Elizabeth Katznelson wrote:
 Please help! I have ADSL from Bezeq, using an Aztech ADSL2+ Ethernet
 Bridge Router DSL-600E (Texas Instrument AR7 chipset). I am running
 Ubuntu Dapper Drake with the 2.6.15-26-686 kernel. I set up the
 connection with pppoeconf. I not only have to restart the connection
 every time I boot, but it times out about every 5 minutes, whether I'm
 using it or not. Pon dsl-provider doesn't do anything. I restart with
 sudo /etc/init.d/networking restart. Skype works all the time, but the
 internet connection times out. Does anyone have any ideas? I am new to
 Ubuntu--my brother set up my machine. but he doesn't know why it's not
 working (he just set up ADSL on an Ubuntu laptop in Beijing and isn't
 having any problems. My husband just plugs the ethernet cable into his
 Macintosh and doesn't have any problem.



 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Writing I/O intercepting driver in Linux

2006-06-18 Thread Tzahi Fadida
I suggest you start by simply echo 1 /proc/sys/vm/block_dump
and turning off klogd

http://linux.inet.hr/proc_sys_vm_block_dump.html

If it is not enough, search the archive for a similar discussion
i had a year ago. There is a code in the kernel where the block_dump
is hooked thus, you can find out a lot just by looking in there.

On Sunday 18 June 2006 17:05, Michael Sternberg wrote:
 We have to write driver that intercept all I/O to disk and notifies
 user-mode application with following data: block length and device number.
 What is a proper way to implement it: to write a block driver above disk
 layer or to implement a file system filter ? If we'll implement a file
 system filter could we miss some direct I/O to the disk ? What kind of I/O
 could we miss ?

 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Xen and Windows XP

2006-06-14 Thread Tzahi Fadida
I am curious,
When you say windows xp emulation, you mean by XEN itself
because of the VT feature or in QEMU or VMWARE in the XEN
guest system. If it is directly with XEN, can the same thing be
done with AMD pacifica which is now shipping out?
http://www.ynet.co.il/articles/0,7340,L-3262066,00.html
I believe that the 5000 definately supports pacifica.

On Tuesday 13 June 2006 19:00, Shachar Shemesh wrote:
 Marc A. Volovic wrote:
  Quoth Shachar Shemesh:
  Not to rub it in for anyone (hi Ira), even as I'm writing this email
  we're running a Windows XP installation inside a XEN VM running on
  Debian.
 
  Rub, rub!
 
  Which windows? which Xen? Which CPU?

 Xen 3.0.2. I had to apply a patch so that the vmx support be included,
 but the maintainer promised me that the package with the patch is
 already waiting in NEW.

 The domain0 machine is running Debian Etch, as it does little and is the
 hardest to upgrade remotely (the machine goes to a client). The windows
 is Windows XP Pro. It is using a LVM over RAID-1 as a physical disk.

 We have had several vnc mishaps during the graphical part of the
 installation, so we switched to SDL. I'm hoping that once the actual
 system is installed I can switch back, as I don't think it's possible to
 start a machine headless and then start SDL at a later point.

 The CPU is a Pentium-D 930 dual core. It has VT and XD (aka NX). I have
 not tested whether the linux kernel actually picked up on the later and
 does non-execute regioning.

 I'll also mention that it took us over a month to get the machine. The
 price quote had to be revised a lot of times, and when we finally placed
 the order, it took over three weeks for the machine to actually arrive.

  M

Shachar

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Ruby on Rails for a high-profile site?

2006-06-09 Thread Tzahi Fadida
On Friday 09 June 2006 18:35, Ira Abramov wrote:

 well, I'll look into that. if it can take Hebrew and hunderds of
 thousands of users as a start without a flinch, it's certainly worth
 mentioning to them. it's also a good idea to find out if RoR is not.

 Thanks for the help guys, I'll keep researching...

For hebrew, like most framework, there is no internal support.
However, this is a web application, thus, mostly you all the hebrew
stuff is done in the browser. The server, lets say postgesql, have support
for UTF8 etc... 
There is an internationalization support for cases where you want to
have both english and hebrew there is a package called globalize.
here are a few links i bookmarked on the subject:
http://www.globalize-rails.org/wiki/
http://dev.rubyonrails.org/browser/plugins/localization/README
http://dev.rubyonrails.org/browser/plugins/localization
http://www.robbyonrails.com/articles/2006/05/14/rails-i18n-gets-a-mailing-list
http://railsconf.org/talks/selected/show/112
http://www.planetrubyonrails.org/show/search?q=internationalization
http://rubyforge.org/projects/gloc/

Also, you are probably going to overide the native english error 
messages anyway if you are even going to use the native error
flash messages handler.
there aren't many and RoR provides a one line method to overide them.
I think there are about 10-15 or something like that.

I personally believe it can be done though as to the resource signature, i 
don't know. It can scale well, but will you need one/two server more than
a straightforward PHP site? can't tell.

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Re: Ruby on Rails for a high-profile site?

2006-06-08 Thread Tzahi Fadida
I recently played with ruby and ruby on rails.
37signals, ruby rails creator, says it scales very well on their servers
It has a mod for apache and i understand that if you want even
greater scalability (resources wise), you can install lighthttpd
server which is considered faster than apache and secure.
Regarding the development scalability, from what i have seen
it scales well. You have infrastrucutre called migration which allows
you to migrate your schemas and data to other platforms and
there is advanced session storage where you can store the session
in database instead of files thus, user accessing one site and jumps to 
another will receive same session information.
Enjoy.

On Thursday 08 June 2006 16:35, Lior Kesos wrote:
 Het Ira,
 Well RoR is a sexy contendent but more in from the ajax/web2.0 point of
 view...
 If it's a community oriented site providing community services I'd go with
 drupal.
 There's even a Israeli community at http://www.drupal.org.il  which I'm
 involved in (as you know).
 I'm sorry about the abscence of the themes directory there but if you want
 a scalable flexible community oriented site drupal is my tip (drupal 4.7
 has sprinkles of ajax resources and supportas well if that's what rocks
 their world).
 I have no idea how far RoR scales but spreadfirefox, snowboard magazine,
 kerneltrap, debian planet have big communities and seem to be doing fine on
 drupal.
 If you find a scalability doc please do share.. I'm interested to see how
 these two measure up...
 best regards
 Lior

 On 6/7/06, Ira Abramov [EMAIL PROTECTED] wrote:
  Top o' the morning, gang!
 
  I haven't played around with RoR yet, but a client of mine wants to
  launch a full-fledged community site with oodles of content, and his web
  programmer is pressing in the direction of RoR. I'm starting to read
  about it but I can't find a lot to read about how scalable it is and
  about dimensioning (i.e. how to plan the CPU power needed for this
  venture and if Ruby+Rails can be realistic for a majorly busy portal,
  how efficiant it is treating transaction compared to say PHP).
 
  Anyone got some tips or bookmarks?
 
  TIA,
  Ira.
 
  --
  Nothin' but a hounddog
  Ira Abramov
  http://ira.abramov.org/email/
 
  =
  To unsubscribe, send mail to [EMAIL PROTECTED] with
  the word unsubscribe in the message body, e.g., run the command
  echo unsubscribe | mail [EMAIL PROTECTED]

-- 
Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at 
http://members.lycos.co.uk/my2nis/spamwarning.html

To unsubscribe, 
send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



RE: ICQ in sendsms 3.11 broken

2006-05-30 Thread Tzahi Fadida
I recommend to use vicq.
I use it in my pop2sms script.
works great.
Basically it's a terminal icq program that can also perform
command line tasks like sending sms. 

 -Original Message-
 From: [EMAIL PROTECTED] 
 [mailto:[EMAIL PROTECTED] On Behalf Of Nadav Har'El
 Sent: Tuesday, May 30, 2006 4:27 PM
 To: Arieh Skliarouk
 Cc: [EMAIL PROTECTED]
 Subject: Re: ICQ in sendsms 3.11 broken
 
 On Tue, May 30, 2006, Arieh Skliarouk wrote about ICQ in 
 sendsms 3.11 broken:
  Hello Nadav,
 
 Hi. Glad to know that linux-il is my personal mailing list :-)
 
  Seemingly ICQ decided to stop the SMS sending from their 
 web interface, and
  now allow only in-protocol SMS sending. Page
  http://www.icq.com/sms/login.php returns an error, and 
 alternative web page
  is nowhere to be found.
 
 Unfortunately, you are very correct. ICQ no longer appear to 
 have an SMS
 sending form on their web site. This situation is not new - 
 it happened
 about a year ago, and unfortunately since then I never 
 modified SendSMS
 to use an alternative technique.
 
 The alternative technique is to use some sort of program or 
 library that
 speaks the ICQ protocol, rather than HTTP. Any graphical or 
 command line
 ICQ implementation (e.g., micq) should work, and there are 
 also several
 Perl and other libraries that do that. Some other people on this list
 may be able to recommend a good command line program or library -
 unfortunately, my life got a little too busy for continuing to work on
 SendSMS...
 
 -- 
 Nadav Har'El|   Tuesday, May 30 
 2006, 3 Sivan 5766
 [EMAIL PROTECTED] 
 |-
 Phone +972-523-790466, ICQ 13349191 |Why do programmers mix 
 up Christmas and
 http://nadav.harel.org.il   |Halloween? Because DEC 
 25 = OCT 31
 
 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]
 



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



RE: Google SOC.

2006-05-30 Thread Tzahi Fadida
First, I appollogize in advance for posting to everyone
but due to some communication problems it can't be helped.

Hi Oded,
Please contact me off list as soon as you can about SOC.

10x.

 -Original Message-
 From: [EMAIL PROTECTED] 
 [mailto:[EMAIL PROTECTED] On Behalf Of Oded Shimon
 Sent: Saturday, May 27, 2006 5:43 PM
 To: 'Linux-IL'
 Subject: Re: Google SOC.
 
 On Sat, May 27, 2006 at 05:11:09PM +0200, Tzahi Fadida wrote:
  A bit of sharing.
  
  My application was accepted to Google SOC.
  http://code.google.com/soc/pgsql/about.html
  http://tzahi.webhop.info/socproposal.txt
  
  Are there more Israeli accepted soc applications?
 
 For a minute there I thought this mail was personal to me! 
 Was like, wow, 
 how'd you find out I'm in this. :P
 
 I'm a mentor, for 2 students. 
 http://code.google.com/soc/ffmpeg/about.html
 
 - ods15
 
 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]
 



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



RE: Google SOC.

2006-05-29 Thread Tzahi Fadida
Great,
Good luck with the students. 

 -Original Message-
 From: [EMAIL PROTECTED] 
 [mailto:[EMAIL PROTECTED] On Behalf Of Oded Shimon
 Sent: Saturday, May 27, 2006 5:43 PM
 To: 'Linux-IL'
 Subject: Re: Google SOC.
 
 On Sat, May 27, 2006 at 05:11:09PM +0200, Tzahi Fadida wrote:
  A bit of sharing.
  
  My application was accepted to Google SOC.
  http://code.google.com/soc/pgsql/about.html
  http://tzahi.webhop.info/socproposal.txt
  
  Are there more Israeli accepted soc applications?
 
 For a minute there I thought this mail was personal to me! 
 Was like, wow, 
 how'd you find out I'm in this. :P
 
 I'm a mentor, for 2 students. 
 http://code.google.com/soc/ffmpeg/about.html
 
 - ods15
 
 =
 To unsubscribe, send mail to [EMAIL PROTECTED] with
 the word unsubscribe in the message body, e.g., run the command
 echo unsubscribe | mail [EMAIL PROTECTED]
 



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



Google SOC.

2006-05-27 Thread Tzahi Fadida
A bit of sharing.

My application was accepted to Google SOC.
http://code.google.com/soc/pgsql/about.html
http://tzahi.webhop.info/socproposal.txt

Are there more Israeli accepted soc applications?

Regards,
Tzahi.
--
Tzahi Fadida
Blog: http://tzahi.blogsite.org | Home Site: http://tzahi.webhop.info
WARNING TO SPAMMERS:  see at http://members.lycos.co.uk/my2nis/spamwarning.html



=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]



  1   2   3   4   5   >