Re: OT: looking for a job

[Yedidyah Bar David]

Hi Maayan,

Not sure why you think you must choose between posting to this list
and searching LinkedIn.
Why not both?

I am by no means a LinikedIn fan, but it's definitely a real thing,
and I wouldn't be surprised if I
heard that tens of percents of all tech job assignments in Israel are
through/thanks to it. I have
absolutely no idea about real numbers and/or deeper implications (such
as job retention when
through LinkedIn vs otherwise, etc.).

I personally work for Red Hat, for 8+ years now. No idea if we have
jobs that can suit
you, but you are welcome to have a look. The Israeli office has ~ 500
people now and is
growing quite quickly.

Incidentally, I also got this job thanks to linkedin - I noticed there
that some old acquaintance
moved to Red Hat and thought it might be a good idea for me as well...

To the rest of the list: I hope this doesn't sound like an
advertisement for Red Hat. I
considered replying in private and decided it's ok as-is. I hope that's ok...

Good luck and best regards,
--
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: recover ssh-agent socket

[Yedidyah Bar David]

On Sun, Jan 9, 2022 at 8:11 AM Daniel Shahaf  wrote:
>
> Tzafrir Cohen wrote on Sat, 08 Jan 2022 09:06 +00:00:
> > I accidentally deleted my ssh-agent's socket from /tmp. The agent is
> > still running and I have $SSH_AGENT_PID and $SSH_AUTH_SOCK set in
> > various processes, so I know where it should have been.
> >
> > Is there any way to recover the socket? Short of restarting the X
> > session, of course.
>
> Could you just spawn a new ssh-agent instance and have it use
> $SSH_AUTO_SOCK as the name of the socket, and re-add private keys to it?
> $SSH_AGENT_PID would still point at the old instance, sure, but what
> would that break?

In the past, I did something similar, which partially worked - I started a new
ssh-agent with its defaults, and linked the newly-created socket to the old
location.

Actually what I did was in a slightly different flow - I had my X session killed
for some reason, but had screen/tmux sessions in the background. I logged
in again to a new X session, attached the screen/tmux's, and wanted to be
able to start X clients/windows from shells there.

I don't remember anymore what didn't work perfectly, but I do remember
that I eventually closed the old screen/tmux's and started new ones, due
to  some annoyances I didn't bother debugging further.

Good luck,
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: diff

[Yedidyah Bar David]

On Tue, Dec 14, 2021 at 5:45 PM Mark E. Fuller  wrote:
>
>
> It's such a small thing, but I really enjoyed being privy to this
> question and discussion today.
>
> On 14/12/2021 17:21, Daniel Shahaf wrote:
> > אורי wrote on Tue, 14 Dec 2021 07:44 +00:00:
> >> Actually I prefer the "<(..)" method, because sometimes I want to compare 2
> >> commands:
> >>
> >> diff <(pip freeze | sort) <(cat requirements.txt | sort)
> >>
> >
> > Consider s/diff/comm -12/.
> >
> >> Or even use cat and echo to add a specific line to one of the outputs. If I
> >> want the result to be completely null. For example to run in crontab and
> >> send me mail whether it's not null.
> >
> > Careful here.  By default, cron jobs run not under your login shell but
> > under /bin/sh, and it's possible the latter doesn't support «<(…)» even
> > if the former does.
>
> Out of curiosity I asked the internet:
> https://unix.stackexchange.com/questions/159513/what-are-the-shells-control-and-redirection-operators
>
> Looks like < as a redirect is POSIX-compliant, but as soon as it's
> <(...) it's not a redirect anymore but instead a process substitution
> which is implemented in BASH, ZSH, etc.:
> https://www.gnu.org/software/bash/manual/bash.html#Process-Substitution
> https://zsh.sourceforge.io/Intro/intro_7.html
>
> Nice to learn a new trick - I probably would have just redirected to a
> temp file instead of the compact solution discussed today.

If we are heading in this direction, may I remind of another, likely
POSIX-compliant
(didn't check), more "traditional" way, that still does not require
(much) I/O, at least on
any reasonably-modern *nix (perhaps e.g. not including cygwin etc. -
again, didn't check):

$ mkfifo pipe1
$ mkfifo pipe2
$ command1 > pipe1 &
$ command2 > pipe2 &
- These are now blocked, waiting in the background until someone reads
from the pipes
$ diff pipe1 pipe2
- After 'diff' finishes reading, command1 and command2 will finish too.
- On a job-controlling shell, you'll get some messages about this.
$ rm pipe1 pipe2

Best regards,
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: multiple python versions

[Yedidyah Bar David]

On Thu, Jul 22, 2021 at 8:31 PM Shlomo Solomon  wrote:
>
> OK - to sum up:
> 1 - thanks to Omer and Dan
> 2 - it seems that I did understand the concept of virtual environments
> - as I wrote in my original post:
> > I know I can use venv to set up virtual environments for different
> > versions, but although I haven't tried it, I don't see how that would
> > solve the problem.
> 3 - based on the answers you both wrote, there's no solution to the
> "problem" of having to re-install modules.
> 4 - and it's probably not a good idea to try :-)

That's the main point, not that there is no solution. If you want to
try to enforce use of modules from the other version, you can use
PYTHONPATH and/or sys.path, or even play with importlib. But it
remains a bad idea, unless you have very specific needs - not a
mere "trying to save time".

Best regards,

> 5 - based on all of the above, I'll probably pass on the idea of using
> 2 versions and wait for Kubuntu to move to 3.9, which will probably not
> be too far in the future.
>
> Again - thanks
>
>
>
> On Thu, 22 Jul 2021 20:07:20 +0300
> Omer Zak  wrote:
>
> > FALSE ECONOMICS ALERT!
> > FALSE ECONOMICS ALERT!
> > FALSE ECONOMICS ALERT!
> >
> > You say that "most modules DO work when moving to a newer version of
> > Python".
> > However when they do not work, it is a lot of work diagnosing the
> > problem and finding which module needs to have both versions installed
> > in parallel.
> >
> > It is better to spend the (relatively short and predictable) time
> > maintaining a full virtualenv for each project. And if you are short
> > on disk space, then today's disks are big and inexpensive - much less
> > expensive than the time you spend trying to save few megabytes by not
> > installing parallel versions.
> >
> >
> >
> >
> > On Thu, 2021-07-22 at 20:02 +0300, Shlomo Solomon wrote:
> > > On Thu, 22 Jul 2021 19:50:46 +0300
> > > Omer Zak  wrote:
> > >
> > > > Why do you want to avoid having to re-install modules for each
> > > > version/environment?
> > > >
> > > The short answer: too much work
> > >
> > > The slightly longer answer: Although there are certainly changes
> > > between versions of Python and/or modules, most modules DO work when
> > > moving to a newer version of Python. I don't remember re-installing
> > > everything when upgrading to a newer version of Python.
> > >
> > > But here, my problem is not upgrading, but keeping both versions.
> > >
> > >
> > >
> > >
> > > > In the general case, a module version is compatible only with a
> > > > subset
> > > > of Python versions, due to API changes from Python version to
> > > > Python
> > > > version.
> > > >
> > > > You also want to let each project decide with which module version
> > > > it
> > > > wants to work, due to potential incompatibilities between module
> > > > versions (it is no accident that pip freeze preserves installed
> > > > module
> > > > versions).
> > > >
> > > >
> > > > On Thu, 2021-07-22 at 19:37 +0300, Shlomo Solomon wrote:
> > > > > Omer Zak  wrote:
> > > > > > The answer to your prayers is pyenv.
> > > > > > It allows you to install multiple Python versions in parallel,
> > > > > > and
> > > > > > for
> > > > > > each version you can maintain several virtualenvs.
> > > > >
> > > > > Dan Yasny  wrote:
> > > > > > How about using virtualenv for alternative versions?
> > > > >
> > > > > Yes, I know about pyenv and virtualenv, but wouldn't I have to
> > > > > re-install modules for each version/environment? That's what I'm
> > > > > trying
> > > > > to avoid.
> > >
> > >
>
>
>
> --
> Shlomo Solomon
> http://the-solomons.net
> Claws Mail 3.17.5 - KDE Plasma 5.18.5 - Kubuntu 20.04
>
> ___
> Linux-il mailing list
> Linux-il@cs.huji.ac.il
> http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Where to buy a Linux based desktop in Israel?

[Yedidyah Bar David]

On Mon, May 11, 2020 at 10:49 PM Dimid Duchovny  wrote:
>
> Hi,
>
> I'm looking to buy a desktop computer with Linux pre-installed for a couple 
> of senior citizens. I'd usually buy a generic PC and install myself, but now 
> due to the coronavirus that's not an option.
> The main uses are web browsing and skype/zoom. Looking for a stable distro 
> and long term support (e.g. the latest Ubuntu LTS). The hardware needs to 
> work out-of-the-box.
> Specification wish list:
> 512GB SSD
> 16GB RAM
> A high-quality case and PSU (preferably 80+ gold)
> Webcam
>
> The budget is around 2k shekels.
> Are there any Israeli retailers that sell such configurations?

FWIW: Around two months ago, when it was clear that with the corona
situation I'll need another laptop, for my kids' home-schooling, I
went to the nearby KSP store, asked what they can offer (without even
searching their site), and bought something that I can't find on their
site anymore, but similar ones are:

https://ksp.co.il/?select=.268..271..132..1784.==1=2=0=0_search===0=0==_real=

This link to a search result gives you all their Dell laptops with
Linux on them, cheapest first, which is currently ~ 2500 NIS. Mine was
~ 2300 NIS and has 256GB SSD (vs 512 for cheapest now).

It arrived with Ubuntu 18.04, everything worked out of the box. My
kids generally like it, and I too (but hardly ever use it, other than
for helping them), definitely for this price.

Cheapest there with 16GB RAM (which is what you wanted) is ~ 4000,
quite a lot higher, but perhaps you can add RAM to one of the others,
no idea.

 Of course, if you buy a brand name, you can get it in any other
store, and are quite likely to get exactly the same thing, as opposed
to someone building it for you from parts.

Best regards,

>
> TIA
> ___
> Linux-il mailing list
> Linux-il@cs.huji.ac.il
> http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Gmail and Claws

[Yedidyah Bar David]

Not directly related, but I'd like to also point out, as no-one
mentioned this yet:

https://gsuiteupdates.googleblog.com/2019/12/less-secure-apps-oauth-google-username-password-incorrect.html

TL;DR: they are stopping Less-Secure-Apps support in June. Blog post
talks about "G Suite", so perhaps does not apply to gmail.com, not
sure.

So: Be prepared to start using OAUTH2, and if your favourite app does
not support that, plan ahead - find an alternative, spend time adding
that yourself, lobby...

My guess is that even if this does not apply to gmail.com, nor to
app-specific passwords, it will likely eventually apply to these.
Didn't do a thorough search on the subject.

Best regards,

On Tue, Apr 28, 2020 at 5:37 AM Steve Litt  wrote:
>
> On Sun, 26 Apr 2020 14:43:51 +0300
> shlomo solomon  wrote:
>
> > I see what you're doing, but don't see how this would solve my
> > problem. Moving from pop3 to imap would not change the fact that
> > Google is not allowing me to login.
>
> It takes about 10 minutes to make the switch, and if necessary 10
> minutes to switch it back. If it changes the symptom, then your mental
> model of how things really are (as opposed to their design or how a
> sane person would design it) was inaccurate. In other words, the bug
> could be bizarre, and if that's true, 10 minutes is the cheapest
> possible way to find out.
>
> I wish I had a dime for every time somebody said "it can't be that,
> that won't change it" when I recommended a diagnostic test, and then the
> diagnostic test changes it in an unexpected way.
>
>
> > And most probably, Ori is correct
> > in his opinion that the problem is the login to Google services - not
> > to pop3.
>
> >
> > Or do you think that procmail would be allowed to login where claws is
> > not allowed?
>
> That would be fetchmail grabbing your mail from the server. Fetchamail
> grabs it, then hands it off, on your local machine, to procmail to be
> sorted or /dev/nulled. In answer to your question, if Claws is buggy
> when it pulls pop, then swapping in fetchmail would fix the problem.
> Or, if the problem is with Google's pop server but not their imap
> server, then switching to imap would fix the problem. And if the
> problem doesn't get fixed, you ruled out some things.
>
> >
> > Just to remind you of what I wrote in my original post - this has
> > worked fine for many years and since nothing has changed on my end, I
> > have no reason to think that pop3 is the problem.
>
> That's my point. The past few weeks you've randomly gotten this error
> when pulling email. It's an intermittent problem, logic is the usual
> victim of intermittent problem, and it's often best to use an alternate
> set of debugging tactics when dealing with intermittents.
>
> What I'd do in your situation is:
>
> 1) Make a Claws folder just to contain downloaded stuff.
>
> 2) Temporarily configure fetchmail to retrieve from the offending gmail
> account and send the messages to procmail.
>
> 3) Configure procmail to place all messages in your new folder
>
> 4) Look for evidence of login failures
>
>
> In case you're using MH folders with your Claws, here are some docs on
> how to get procmail to drop messages into MH folders:
>
> https://www.gnu.org/software/emacs/manual/html_node/mh-e/Procmail.html
>
> https://unix.stackexchange.com/questions/336420/how-do-you-use-procmail-mda-to-deliver-files-to-be-read-by-mh
>
> https://linux.die.net/man/5/procmailex
>
> http://www.cs.utah.edu/~clake/soc_mh.html
>
> SteveT
>
> Steve Litt
> March 2020 featured book: Troubleshooting: Why Bother?
> http://www.troubleshooters.com/twb
>
> ___
> Linux-il mailing list
> Linux-il@cs.huji.ac.il
> http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: OT: Non-Israeli domain name registration.

[Yedidyah Bar David]

Last time I had to renew my only (.org) domain I decided to transfer it to
box.co.il, which is also a .il registrar (which I why I knew them). It was
a bit more expensive than the cheapest options abroad, but the ability to
pay in NIS with an Israeli (-only) credit card, and be able to talk with
someone over a phone, in Hebrew, made it, for me. The transfer and renew
was just fine (although did take some time and work, also because I had to
ask the previous registrar to allow transfer but also because they had some
mistakes etc.). Didn't hear from them since then, and do not expect to hear
- the yearly "update your details at icann" emails include links to "
name-services.com", and the last one was also sent From: there. They are a
sub-registrar of enom.com, which IIUC is a rather large registrar that only
has sub- ones, does not work with end customers directly.

Best regards,

On Thu, Aug 8, 2019 at 1:19 PM Yedidyah Bar David  wrote:

> Last time I had to renew my only (.org) domain I decided to transfer it to
> box.co.il, which is also a .il registrar (which I why I knew them). It
> was a bit more expensive than the cheapest options abroad, but the ability
> to pay in NIS with an Israeli (-only) credit card, and be able to talk with
> someone over a phone, in Hebrew, made it, for me. The transfer and renew
> was just fine (although did take some time and work, also because I had to
> ask the previous registrar to allow transfer but also because they had some
> mistakes etc.). Didn't hear from them since then, and do not expect to hear
> - the yearly "update your details at icann" emails include links to "
> name-services.com", and the last one was also sent From: there. They are
> a sub-registrar of enom.com, which IIUC is a rather large registrar that
> only has sub- ones, does not work with end customers directly.
>
> Best regards,
>
> On Wed, Jul 31, 2019 at 2:02 PM Rabin Yasharzadehe  wrote:
>
>> I'm using Gandi <https://www.gandi.net/en> and Hover
>> <https://www.hover.com/> and I'm very satisfied by them both.
>>
>> --
>> Rabin
>>
>>
>> On Wed, 31 Jul 2019 at 13:10, vordoo  wrote:
>>
>>> Regarding .com, .org, domains. What are the best registrars? Which are
>>> to be avoided (beside Godaddy).
>>>
>>>
>>> Thanks!
>>>
>>>
>>> ___
>>> Linux-il mailing list
>>> Linux-il@cs.huji.ac.il
>>> http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
>>>
>> ___
>> Linux-il mailing list
>> Linux-il@cs.huji.ac.il
>> http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
>>
>
___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: oVirt over CentOS or ovirt-node

[Yedidyah Bar David]

Hi Rabin :-),

On Mon, Jan 8, 2018 at 10:48 PM, Rabin Yasharzadehe  wrote:

> Just wondering which OS used for the host, I was wondering if one version
> is more supported then the other.
>

Why not start a thread on us...@ovirt.org, then?

You might find this useful/interesting:

http://lists.ovirt.org/pipermail/users/2017-April/081221.html

Partial quote:

- oVirt Node jumped from 7.1% on 3.6 survey to 19.9% in 4.1!

So most people use CentOS, but some do use ovirt-node,
and now more than in the past.

Both are supported. I'd say node is more for people that prefer
pre-made appliances, while CentOS is for people that want the
most control.

If you consider node, you might want to read e.g.:

https://www.ovirt.org/node/faq/
https://www.ovirt.org/develop/release-management/features/
node/node-next-persistence/

Disclaimer: I am an oVirt developer.

Best regards,

>
> --
> Rabin
>
> On 8 January 2018 at 09:50, Gilboa Davara  wrote:
>
>> On Tue, Jan 2, 2018 at 5:41 PM, Rabin Yasharzadehe 
>> wrote:
>> >
>> > ovirt-node with version 4.1 gave a alot of problems with nfs shares
>> > but 4.2 did work much better "out of the box" after clean install (I
>> was not
>> > able to upgrade from 4.1 to 4.2 with ovirt-nodes)
>> >
>> > so my question is basicly , if I'm looking to setup something for the
>> long
>> > run, which I can extend and upgrade in the near future,
>> > which host platform should I choose ? ovirt-node or a simple CentOS7 ?
>> >
>> > Tx
>>
>> We're using 3-node oVirt 4.0.6 over GlusterFS cluster (w/ self hosted
>> engine) in production for long while now.
>> The system was installed by my IT manager from barebone (was
>> originally installed in the 3.6 days).
>> If you have specific questions I can forward them...
>>
>> - Gilboa
>>
>
>
> ___
> Linux-il mailing list
> Linux-il@cs.huji.ac.il
> http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
>
>
___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Systemd error messages for custom service

[Yedidyah Bar David]

On Sun, Sep 3, 2017 at 6:44 PM, David Suna  wrote:
> I am getting started with systemd on an Ubuntu 16.04 system. I have a custom
> service implemented in PHP which is outputting messages

How? Writing to stdout?

> which I can see
> using journalctl. I would like to be able to filter on the message priority
> to distinguish between error messages and debug messages. However, all of my
> messages seem to be showing as info messages. How can I format my messages
> so that they will fit into the different log levels?

I think you should use syslog, and/or systemd-/journald-specific means, e.g.:

https://github.com/systemd/php-systemd

Best,
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How to search Linux Kernel changelogs? (USB disconnect problem)

[Yedidyah Bar David]

On Mon, Dec 21, 2015 at 12:53 PM, Omer Zak  wrote:
> During the last several months, I was having a problem of USB mouse
> disconnecting and reconnecting very often in my Linux system (Debian
> Wheezy, kernels 3.16.0-4-amd64 and 4.2.0-0.bpo.1-amd64). Recently a
> similar problem started to affect also my printer.
>
> Hardware problems were ruled out yesterday in the computer repair lab.
>
> I found some information which suggested that it has to do with USB
> power control, and that there are several problems with USB handling in
> general.
>
> I also saw that other people complain about similar problems, often
> without a solution to their problems.
>
> So I wanted to search the Linux Kernel changelogs for any recent changes
> in the USB subsystem.
>
> Unfortunately I found no such search function.
> https://www.kernel.org/pub/linux/kernel/ has all changelogs in its
> subdirectories but no search function.
> The search function in http://kernelnewbies.org/LinuxChanges is broken -
> I get Web searches even when I choose "This site".
>
> What do you do when you need to search Linux kernel changesets for
> relevant changes?

If no better solution, you can clone the kernel git repo and use git to
search there.
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How to install several kernels on Debian

[Yedidyah Bar David]

On Sun, Dec 13, 2015 at 3:06 PM, Oleg Goldshmidt  wrote:
>
> Hi,
>
> I am confused and I wonder if some Debian expert can calm me down. I
> want to install several different kernel versions on a Debian
> server. Off-topic: this is something that Red Hat do without being asked
> (they keep several versions, usually 3), so it is something that seems
> natural to me.

Generally, should work similarly in Debian.

>
> However, I cannot find any documentation anywhere on the 'net that tells
> me how to get two or more kernel versions on the same Debian machine without
> compiling. I just want to install 2 or more linux-image packages and
> whatever else needs to be installed - and have them in grub, etc. The
> only things I see are "don't install new kernels" and "you can upgrade"
> and "download and compile".
>
> # apt-cache search linux-image
>
> shows 3.2.0, 4.2.0, and 2.6. I will be happy with the first two (though
> I'd like something in between as well).
>
> However,
>
> # apt-get install linux-image-4.2.0-1-amd64
> Reading package lists... Done
> Building dependency tree
> Reading state information... Done
> Some packages could not be installed. This may mean that you have
> requested an impossible situation or if you are using the unstable
> distribution that some required packages have not yet been created
> or been moved out of Incoming.
> The following information may help to resolve the situation:
>
> The following packages have unmet dependencies:
>  linux-image-4.2.0-1-amd64 : Breaks: initramfs-tools (< 0.110~) but
>  0.109.1 is to be installed
>Breaks: udev (< 208-8~) but 175-7.2 is to be installed
>E: Unable to correct problems, you have held broken packages.
>
> does not add to my confidence. I want 2 kernels, 2 initial ramdisks,
> etc., on the same machine and in the same grub without investing time to
> compile.

This break due a specific different issue, not because Debian does not
support this in general.

No idea about your specific issue. Did you try to also upgrade udev and
initramfs-tools?
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How to install several kernels on Debian

[Yedidyah Bar David]

On Tue, Dec 15, 2015 at 2:03 PM, Oleg Goldshmidt <p...@goldshmidt.org> wrote:
> Yedidyah Bar David <linux...@didi.bardavid.org> writes:
>
>> On Sun, Dec 13, 2015 at 3:06 PM, Oleg Goldshmidt <p...@goldshmidt.org> wrote:
>>>
>>> this is something that Red Hat do without being asked
>>> (they keep several versions, usually 3), so it is something that seems
>>> natural to me.
>>
>> Generally, should work similarly in Debian.
>
> So what's the apt-get equivalent of yum install (as apt-get install is
> similar to yum update)?

In a sense, you are right, but not in the way you intended: yum
install will also
update an installed package if an update is available.

>
>> This break due a specific different issue, not because Debian does not
>> support this in general.
>
> I think that the root cause is that "apt-get install" updates rather
> than installing together.
>
>> No idea about your specific issue. Did you try to also upgrade udev and
>> initramfs-tools?
>
> Will my 3.2 keep working? I have no confidence in that. Again, I do not
> want to "upgrade" anything - I want to switch between several kernels at
> will. I also do not want to compile - I want stock Debian kernels.

AFAICS wheezy had 3.2, jessie has 3.16.

My laptop was upgraded from wheezy to jessie and has both kernels installed.
IIRC I successfully booted the 3.2 one after the upgrade, didn't try
that recently,
as I have no need for that.

Didn't check about udev and initramfs-tools.

The kernel packages normally arrive with a kernel image, plus many modules.
initramfs-tools is used to build an initrd image, which mainly includes the
relevant modules.

In principle, building such an initrd of the 3.2 kernel with
jessie-updated tools
is indeed risky.

I am pretty certain that the above is almost identical in rhel/centos/fedora,
except that upgrading there between releases is significantly different.
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: xargs guide

[Yedidyah Bar David]

2015-07-30 4:37 GMT+03:00 Steve Litt sl...@troubleshooters.com:
 Hi all,

 xargs is a gender-changer type program that adapts one program's stdout
 to the next program's command line arguments. It's extremely handy for
 shell scripting, but it can be tricky. I've written a short guide for
 xargs that shows how to get around the usual xargs landmines:

 http://www.troubleshooters.com/linux/xargs.htm

 Hope you like it.


I find missing the mentioning of '-0' (for both xargs and find). I use
them almost always when doing find | xargs, unless I have specific
needs.

Also a comparison with the shell's command substitution (`` or $()( is
probably in place.

Not sure why you need to scare your readers so much...

Nice writeup!
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

[JOB OFFER] Community Impact and Action Lead for Fedora

[Yedidyah Bar David]

Hi all,

I was asked to forward this:

Red Hat is looking for a Fedora Community Action and Impact Lead to join
the Fedora Council and lead initiatives to grow the Fedora user and
developer communities, as well as make Red Hat and Fedora interactions
even more transparent and positive. The Council is responsible for
stewardship of the Fedora Project as a whole, and supports the health
and growth of the Fedora community. In this role, you'll facilitate
decision making on how to best focus the Fedora community budget to meet
our collective objectives, work with other council members to identify
the short, medium, and long-term goals of the Fedora community, and
organize and enable the project. You will also participate in decision
making about use of trademarks, project structure, community disputes or
complaints, and other issues. In this position, you'll hold a full
council membership, and not an auxiliary or advisory role.

http://jobs.redhat.com/jobs/descriptions/fedora-community-action-and-impact-lead-westford-massachusetts-job-1-4988121

Best,
-- 
Didi
___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: terminal emulator

[Yedidyah Bar David]

2014-08-28 15:13 GMT+03:00 Efraim Flashner efraim.flash...@gmail.com:

 What terminal emulators are out there that correctly do bidirectional
 text, for displaying hebrew in a readable manner.  finding one that
 displays hebrew, but left to right, isn't hard, but either i'm not
 configuring it correctly or I'm not finding one that does it right.


I personally use mlterm, and it's good enough for my simple needs.
Generally speaking, there is no correct way, as in according to standard
X,
and AFAIK it's considered these days out of scope for terminals. In that
sense
mlterm is not really following any standard, apart from the unicode bidi
algorithm.
What doens't/can't it do? E.g. correctly handle different parts of a line
that are
unrelated (e.g. if you open mutt and you have Hebrew both in the sender's
name
and in the Subject, it might not display correctly), it has no means to set
the
paragraph direction, etc.
 --
Didi
___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: terminal emulator

[Yedidyah Bar David]

2014-08-28 18:06 GMT+03:00 Efraim Flashner efraim.flash...@gmail.com:

 tilda shows up left-to-right with hebrew letters, mlterm shows up
 right-to-left with boxes.  All on debian sid.
 https://dl.dropboxusercontent.com/u/7048321/tilda.jpg
 https://dl.dropboxusercontent.com/u/7048321/mlterm.jpg


I now verified that konsole also shows hebrew right-to-left.
You should probably configure mlterm (ctrl-rightclick) to use some other
font.

I personally use both xterm and mlterm with a very old raster (pcf) font I
found somewhere a very long time ago, don't remember anymore where, and
tweaked a bit since. I don't mind sharing it, but any modern vector font
will probably look better.
-- 
Didi
___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: reverse ssh

[Yedidyah Bar David]

If you just want an ssh connection you can simply redirect connection
attempts to some port on the
Internet-accessible machine to port 22 on the private-ip one - using
whatever tool that fits you best -
iptables, xinetd, redir, probably many others.
-- 
Didi


2014-07-20 13:31 GMT+03:00 Erez D erez0...@gmail.com:

 looks a little complicated - extra ssh server, firewall with port knocking
 all this for a ssh connection ...

 On Sun, Jul 20, 2014 at 11:38 AM, Rabin Yasharzadehe ra...@rabin.io
 wrote:
  you can add a port-knocking tool like fwknop to add a dynamic rule to
  forward your connection into the privet machine.
 
  --
  Rabin
 
 
  On Sun, Jul 20, 2014 at 12:16 PM, Erez D erez0...@gmail.com wrote:
 
  On Sun, Jul 20, 2014 at 11:06 AM, Lior Kaplan kaplanl...@gmail.com
  wrote:
   Didn't check it, but login in with a user who has /bin/true might do
 the
   trick.
  you are correct, it works.
  however it is still a security risk, as this means the client may
  listen on unused port ...
 
  
   Kaplan
  
  
   On Sun, Jul 20, 2014 at 12:03 PM, Erez D erez0...@gmail.com wrote:
  
   On Sun, Jul 20, 2014 at 10:39 AM, Lior Kaplan kaplanl...@gmail.com
   wrote:
ssh itself ?
   
http://www.thegeekstuff.com/2013/11/reverse-ssh-tunnel/
   nice, however this requires me to give access to my server, which i
 do
   not want ...
   (or, can i give people permission to ssh to my server only for
 reverse
   tunnels and no shell ?)
  
   
Kaplan
   
   
On Sun, Jul 20, 2014 at 11:36 AM, Erez D erez0...@gmail.com
 wrote:
   
hello
   
i have a linux machine with a private ip connected to the internet
i have a public ip and need to ssh to the linux box
   
any tools for that ?
   
___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
   
   
  
  
 
  ___
  Linux-il mailing list
  Linux-il@cs.huji.ac.il
  http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
 
 

 ___
 Linux-il mailing list
 Linux-il@cs.huji.ac.il
 http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Retooling my mail server

[Yedidyah Bar David]

2013/7/18 Ira Abramov lists-linux...@ira.abramov.org:
 Howdie,

 I've been delaying this for years, but for the last 15 of them I've been
 working with qmail and ezmlm and they are getting dated. no DKIM, no
 smtp auth, lots of modern features and community missing. Then the other
 day someone tweeted me this:
 http://sealedabstract.com/code/nsa-proof-your-e-mail-in-2-hours/ and I

Seems nice.
Not sure it really matters (as you can find them on google), but I'll
mention two others I looked at (but did not try):

http://workaround.org/ispmail/squeeze
Similar in attitude

http://www.iredmail.org/
A script that you run and it does everything (?) for you
-- 
Didi

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: c/unix q

[Yedidyah Bar David]

ְAlso, you might cause other software that inherits the fds to
fail/complain/whatever.
I only mention this because just yesterday I noticed that when running
'lvs' on my
Debian wheeze laptop, I get:
File descriptor 3 (/usr/share/bash-completion/completions) leaked on
lvs invocation. Parent PID 11833: -su
File descriptor 4 (/usr/share/bash-completion/completions) leaked on
lvs invocation. Parent PID 11833: -su
File descriptor 5 (/usr/share/bash-completion/completions) leaked on
lvs invocation. Parent PID 11833: -su
and when I searched I found:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=432986
FYI
-- 
Didi

2013/6/4 Geoffrey S. Mendelson geoffreymendel...@gmail.com

 On 06/04/2013 02:43 PM, ronys wrote:

 Nothing. You're just wasting resources (file descriptors) and making
 your code a bit harder to understand and maintain.


 It kind of says to anyone reading the code that you put the minimum into 
 creating it you could, and implies there are details that were not addressed.

 Of course I'm an old assembly language programmer, where everything is 
 declared, nothing is left to default, and anything allocated  or opened is 
 explicitly freed or closed when you are done with it.

 Geoff.


 --
 Geoffrey S. Mendelson,  N3OWJ/4X1GM/KBUH7245/KBUW5379
 It's Spring here in Jerusalem!!!


 ___
 Linux-il mailing list
 Linux-il@cs.huji.ac.il
 http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: The New Coolest Hack: emscripten - an LLVM bitcode to JavaScript Compiler

[Yedidyah Bar-David]

On Sun, Jan 06, 2013 at 06:55:22PM +0200, Shlomi Fish wrote:
  Emscripten has many more, much cooler, demos here:
  https://github.com/kripken/emscripten/wiki including ports of Python, Ruby 
  and
  Lua and some graphical games.

On a semi-related note, see this, if you haven't already:
http://bellard.org/jslinux/
It's a Javascript emulator of a PC, booting linux right inside your
browser.
In case you do not recognize the name, this is the guy that wrote
and maintained for a long time qemu, ffmpeg, and other, lesser known
but not less amazing stuff. Have a look at his homepage.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Is forbidding concurrent ssh sessions a good idea?

[Yedidyah Bar-David]

On Mon, Nov 12, 2012 at 12:51:46PM +0200, Nadav Har'El wrote:
 On Mon, Nov 12, 2012, Elazar Leibovich wrote about Re: Is forbidding 
 concurrent ssh sessions a good idea?:
  While I can certainly see what's broken with it for using a regular
  computer, whose stability I do not value much, and while there are
  difficulties this may cause, do you see anything specific that will break
  in the use case of a production server?
 
 Let me offer another completely different idea, without any kills and
 similar tricks: End your ~/.profile with screen -R -D
 
 What will this do?
 
 The login shell will start screen(1), and let the admin work in it.
 If another admin logs in, he doesn't just kill the existing session - he
 also takes over the existing instance of screen, and can see what the
 other admin was in the middle of doing.
 
 This screen will also allow the admin to have multiple screens - which
 you prevent him from doing with several separate sshs, so he'll
 appreciate screen anyway.
 
 If you don't know screen(1), I suggest you learn it - it is an
 absolutely wonderful tool.

...and also look at its '-x' option which will allow sharing a session
from two (or more) connections. This way your two admins will be able
to talk over the phone while solving a problem together and not having
to tell each other what they did and what happened.

And while at it, also have a look at tmux, which is a screen replacement.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Is forbidding concurrent ssh sessions a good idea?

[Yedidyah Bar-David]

On Mon, Nov 12, 2012 at 06:32:25PM +0200, E.S. Rosenberg wrote:
 +1 for tmux, this does however imply that all the admins are using the
 same account to login (bad scenario) with or instead the tmux/screen
 line should be added at the end of /root/.profile and not ~/.profile

IIRC screen, perhaps tmux too, has some rather complex set of permissions
management options, which I would not be surprized if it turns out that
actually do allow sharing a session from two different users. I did not
try this. I su to the person's user if I want to share a session. Note
that (by default?) you need to chown your tty to the other user or
screen won't work.

 
 Just note that tmux inside of tmux or screen inside of screen tend to
 behave bad/not work, screen inside of tmux works, I never tried the
 other way around.

Never had problems with screen inside screen. Actually, my everyday
work environment comprizes of 3-level screens, some of which also have
tmux inside them. I use different key combinations to travel around and
it works very well. IIRC I did have to change a few configuration options
when moving it from RedHat/CentOS to Debian, but other than that, it
worked more-or-less error-free for something like 5 years now.

Tell me if you want my configuration. It's not very elaborate, only
does what it's meant to do - if you google for things like screenrc
you'll find many examples of what people do with it.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Looking for recommendations: sending SMSs from linux

[Yedidyah Bar-David]

Hi all,

I am looking for recommendations for hardware to send SMSs from a linux
server and connected through its USB port. It's important that it'll be
stable and robust. I'd also like it to:
1. be easy to buy in Israel and get warranty/support
2. have good support in both the kernel and relevant user-mode tools
3. be able to manage from remote, especially reboot if it's stuck
4. be fast
5. not be expensive

Relevant options I thought about:
* a cellular modem plus gammu
* a normal cellular phone and gammu
* an android cellular phone and adb and/or some relevant application

I currently have a very old Nokia 3100 and a USB-serial converter.
Some notes as to how does it pass my criteria above, and related stuff:
1. The phone was then easy to buy, the converter a bit less so - the
official Nokia CA-42 did not work, but this nameless clone did.
2. The converter did not have very good support in old kernels but is
ok with recent ones. gammu works well.
3. The nokia is not easily managable from remote. In certain occasions I
had to go there and reboot and/or disconnect/reconnect it. If I connect
it without a battery and to a remote power device, I can reboot it, but
it does not come up without pressing its on button. I did not check how
do other phones do in this regard. Rebooting the entire server is an
option, but I'd rather not if possible.
4. gammu is slow - sending an SMS takes around 10 seconds. I also tried
gammu with the modem built into a recent Thinkpad (which gammu identifies
as 'F5521gw') and it also took around 9 seconds. I have a feeling that
most of this time is gammu sleeping to be on the safe side, but did
not look at the source or try to make it sleep less.
5. prices are not very relevant to current hardware, but it would be
nice if it didn't cost more than what they did then. E.g. buying the
newest/strongest android phone is certainly not preferable, but a
cheap ~500 NIS one might do better than the ceapest ~200 NIS modem if
people have better experience with it.

Opinions, ideas, and especially recommendations based on real experience
are very welcome.

Thanks,
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Looking for recommendations: sending SMSs from linux

[Yedidyah Bar-David]

Thanks for the info.

On Mon, Oct 29, 2012 at 01:38:55PM +0200, Geoffrey S. Mendelson wrote:
 I use a Hauwei (or is it Huawei?) dongle. The asterisk chan_dongle
 driver allows you to make and receive calls and send and receive
 SMSs.
 
 I was able to find one for sale in Haifa using ZAP, after someone
 who is not on this list pointed me to,a I bought one from an eBay
 vendor selling them. He sells them for under $50 including
 registered mail from Hungary, and they take  less than 2 weeks to
 get here.
 
 I paid about 300 NIS including postage for the one I bought here. It
 was the only one that store had at that time.
 
 Since the dongle is managed by asterisk, you can send an SMS using
 all of the asterisk functions, or by the shell command asterisk -rx
 dongle sms dongle name number message.
 
 You can also query its status, the status of the connection, and
 load and unload the module with asterisk commands.

How long does it take to send 1 message? 3? 10?
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Fw: Establishing a Mageia Mirror in Israel

[Yedidyah Bar-David]

On Sat, Oct 27, 2012 at 12:09:20AM +0200, Shlomi Fish wrote:
 
 
 Begin forwarded message:
 
 Date: Fri, 26 Oct 2012 23:58:52 +0200
 From: Shlomi Fish shlo...@shlomifish.org
 To: mageia-disc...@mageia.org
 Cc: Mageia development mailing-list mageia-...@mageia.org
 Subject: Establishing a Mageia Mirror in Israel
 
 
 Hi all,
 
 if you are a resident of Israel and would like to have an Israeli mirror of
 Mageia, please let http://mirror.isoc.org.il/ and me know. So far, the
 administrator of this mirror ( mirrormas...@isoc.org.il ) believes there is 
 too
 little demand for it, and I would like to show him that this is not the case.

BTW, the relevant criterion isn't being a resident of Israel but being
connected to the IIX, which is almost the same, but probably not exactly.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How do I disable NetworkManager in favor of dhcp setup?

[Yedidyah Bar-David]

On Thu, Mar 08, 2012 at 06:09:33PM +0200, ronys wrote:
  I've had luck with 'sudo telinit 6' in cases where the reboot command
 failed. There's also a /proc variable that will force a reboot when written
 to, but its name escapes me for the moment.

/proc/sysrq-trigger

Read 'man proc'

Also note that writing 'b' to it is similar to unplugging the power cord,
so use with caution. If you suspect a normal reboot won't work you might
want to write there 's' (sync), 'u' (umount/mount readonly), then 'b'
(boot). I know there are people that are used to even more complex
sequences...
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: vim mappings for Hebrew

[Yedidyah Bar-David]

On Wed, Mar 07, 2012 at 10:19:34PM +0200, Nadav Har'El wrote:
 On Wed, Feb 15, 2012, Avraham Rosenberg wrote about Re: vim mappings for 
 Hebrew:
The above from Nadav, using printable characters
   map! F12 esc:set invhk invrlcra
   map  F12  :set invhk invrlcr
 
 I'm now using a slightly improved version:
 
   map F12 :set invrl invhkcr
   map! F12 C-O:set invrl invhkcr
   cmap F12 C-_
 
 For UTF-8, you also need
   set aleph=1488
 
 And I also like this font
   set gfn=Luxi\ Mono\ 15
 
 The only problem is that now that I've gotten used to the Lyx keyboard
 mapping, with shift-minus being the maqqaf, and other shifts giving me
 niqqud, I'm missing that also in vim... Anybody set up such a keymap?

Not directly related to your question, I personally gave up on Hebrew in
vim, and in any specific console application which is not developed for
this specifically, and in normal day work use a combination of 'xterm'
and 'mlterm' running the same screen (with '-x') session. So when I need
Hebrew I move to the mlterm window and when I miss the better feel of
xterm (selection is slightly better and some other things), or when I
need for some reason to see the non-bidi raw text, I go back to xterm.
Ugly but works. Also through ssh etc.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How do I disable NetworkManager in favor of dhcp setup?

[Yedidyah Bar-David]

On Thu, Mar 08, 2012 at 12:21:02AM +0200, Micha wrote:
 I have a redhat 6 desktop based system that exhibits a behavior where
 sshd fails to start on startup despite being enabled and requires manual
 startup after user login. I want to enable remote access before there is
 actual user login on the system
 
 I believe I pinpointed the problem tp NetworkManager being installed and
 enabled, which means that no network connection is actually configured
 before a user is logged in.
 
 I want to know how to remove NetworkManager and enable automatic network
 connection startup before sshd is started.
 
 /etc/network/interfaces doesn't exist, so I'm not sure how this is
 supposed to be achieved these days (under debian it's still there, just
 not active by default). It used to be possible to define the interface
 as auto dhcp in there to get the behavior I want. Will just adding that
 file do the trick is it supposed to be done differently?

On RedHat it's in /etc/sysconfig/network-scripts/ifcfg-interface-name,
see 'man ifup' there. A working (for me, currently) example:

/etc/sysconfig/network-scripts/ifcfg-eth0

DEVICE=eth0
NM_CONTROLLED=yes
ONBOOT=yes
HWADDR=xx:xx:xx:xx:xx:xx
TYPE=Ethernet
BOOTPROTO=dhcp
DEFROUTE=yes
PEERDNS=yes
PEERROUTES=yes
IPV4_FAILURE_FATAL=yes
IPV6INIT=no
NAME=System eth0
UUID=----

Note that although the above says 'NM_CONTROLLED=yes', I do not have
NetworkManager installed so I do not know if it affects anything.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: libreoffice+hebrew nikud

[Yedidyah Bar-David]

On Sun, Feb 26, 2012 at 11:39:19PM +0200, Nadav Har'El wrote:
 On Sun, Feb 26, 2012, Dan Kenigsberg wrote about Re: libreoffice+hebrew 
 nikud:
  ditto. I've been using the lyx layout ever since Tzafrir had incorporated 
  it
  to xkb (but not very often). They are almost sensible (you cannot ask for 
  more
  when it comes to Hebrew niqqud)... Dagesh is AltGr-Daleth, Segol is
  AltGr-Samech. No mnemonics for the Hatafim, I'm afraid. So
  /usr/share/X11/xkb/symbols/il comes handy when I need to type them.
 
 Very nice!
 I was able to enable this with the following command:
 
 setxkbmap -option grp:switch,grp:shift_toggle,grp_led:scroll us,il ,lyx
 
 (what an ugly command line... Even the weird , isn't redundant...
 Users of KDE or GNOME probably have a more graphical way of enabling these
 options).
 
 For me (with the above command), the niqqud is actually chosen with the shift
 (either left or right, it doesn't matter), NOT with the right ALT key.
 But it is indeed working well.

BTW, there is a similar setup for Windows, done by Sivan Toledo:
http://www.tau.ac.il/~stoledo/Tools/Keyboards/
The page also has a nice explanation and a map.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Migrating a Linux (Debian Squeeze) system from one HD to another HD

[Yedidyah Bar-David]

On Mon, Feb 20, 2012 at 06:07:37PM +1100, Amos Shapira wrote:
 You'll HAVE to, especially /boot, before chroot'ing do:
 
 # for i in proc dev sys boot; do mount -o bind /$i /chroot/$i; done

The OP specifically said he mounted a NEW boot partition under the
chroot. I agree with the others. I usually remember only /proc but
can see why all might be needed.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Preparing to convince to shift to non-propriety documents formats

[Yedidyah Bar-David]

On Mon, Feb 20, 2012 at 10:40:58AM +0200, Nadav Har'El wrote:
 On Sun, Feb 19, 2012, Dotan Cohen wrote about Re: Preparing to convince to 
 shift to non-propriety documents formats:
  Undocumented? Which file format is that? All the .doc and .docx
  formats are documented, even the older binary formats.
 
 Where is the .doc format documented?
 
 I once wrote a tool to extract the text in MS Office files (for a search
 engine). It was a really annoying reverse-engineering-like
 trial-and-error process, and I could hardly find any documentation.
 The PowerPoint format (.ppt) was particularly odd.
 
 What documentation do you refer to?

According to Wikipedia, it's partially documented. I did not follow the
links inside:
http://en.wikipedia.org/wiki/DOC_(computing)#Specification
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Migrating a Linux (Debian Squeeze) system from one HD to another HD

[Yedidyah Bar-David]

On Sun, Feb 19, 2012 at 04:03:05AM +0200, Omer Zak wrote:
 My PC has a 500GB hard disk, and I want to migrate it to a 2TB hard
 disk.
 The new hard disk has been formatted to have two physical partitions,
 one serves as /boot and the other is an encrypted LVM volume, which has
 its own division into logical partitions.
 
 I rsync'ed the old hard disk's contents into the new one (excluding
 - /lib/init/rw/***
 - /dev/shm/***
 - /proc
 - /sys
 - /tmp/***
 - /var/lock/***
 - /var/tmp/***
 - /cdrom/***
 - /media/***
 - /etc/fstab
 ).
 
 I edited /etc/fstab in the new hard disk to specify the correct
 partitions.
 
 Then I ran grub-install on the new hard disk using:
   chroot /tmp/new_2T grub-install /dev/sdb
 where the new hard disk's /boot and logical partitions are mounted on
 subdirectories of /tmp/new_2T, and the new hard disk itself is /dev/sdb.
 
 However the logical volume paths in the old hard disk and the new one
 differ.
 It turns out that the kernel and initrd images in the old hard disk have
 the old logical volume paths hardwired in them.

Are you certain about that? I do not have cryptfs (and the cryptroot
script below) installed, but I think very little is hardwired into
Debian's mkinitramfs-generated initrd images.

 So when trying to reboot the PC from the new hard disk, it doesn't find
 the root filesystem.  The failing script is /scripts/local-top/cryptroot
 (/scripts/local-top template can be found in /etc/initramfs).

Can you try passing (adding or changing) root=THENEWROOT from your
bootloader?

If that works, you'll need to change your boot loader's configuration.
In a default squeeze install, that's in /etc/default/grub and
/etc/grub.d . Then run 'update-grub'.

 
 So it seems that I need somehow to recreate the initrd.img's contents.
 
 I may have missed also other things.
 
 Questions:
 1. How to recreate initrd.img's appropriate for the new hard disk?

Not sure how to do this from the old disk, but once you manage to boot
to the new disk (e.g. using some rescue CD or whatever), you should be
able to use mkinitramfs without special options - hopefully it will
detect the changes etc by itself.

Note that I have no experience with cryptfs and related stuff.

 2. Is there any guide about migrating a Debian Squeeze system inside an
 encrypted LVM partition from one hard disk to another hard disk?

No idea.

 
 In conclusion: I am dismayed at the fact that migrating Linux systems,
 which used to be a simple affair (like MS-DOS), is now complicated (like
 versions of MS-Windows after Windows 95).

That's not a fair comparison - you decided to do some non-trivial changes
to your disk layout, and that's not mere migrating. That said, I agree
that in the process of making the boot process more robust, distributions
(Debian included) made it also more complex. One thing I noticed, which
might or might not be relevant here, is that in certain cases the boot
loader identifies filesystems with their UUID, which is not copied by
most people when copying a disk. So now if something changes your
/dev/sda2 to /dev/hda2 or vice versa, you'll still boot, unlike in the
past, but copying root elsewhere will also require updating the boot
loader unless you made sure you also copied the UUID.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Migrating a Linux (Debian Squeeze) system from one HD to another HD

[Yedidyah Bar-David]

On Sun, Feb 19, 2012 at 03:47:32PM +0200, Omer Zak wrote:
 Thanks, Daniel and Amos, for your suggestions.
 However they address the wrong part of my problem.
 All my regular files were already copied to the new hard disk and it was
 properly set up.
 The only problem is to boot from it.
 
 I successfully set up grub2 to boot from the new disk's boot partition
 (a physical partition, not under LVM control).  For this purpose, I
 modified grub.cfg.
 I need to rebuild initrd.img files for all kernels that I have.

need or needed?

For testing, building for one is enough, BTW.

 
 Yedidiah, you questioned my finding that logical volume paths are
 hardwired in the initrd images.  They are.
 The relevant files are:
   /scripts/local-top/cryptroot and /conf/param.conf
 in the initrd image.
 
 According to man mkinitramfs, the file which configures the above
 is /etc/initramfs-tools/initramfs.conf and in practice I see files also
 in /etc/initramfs-tools/conf.d
 
 I created files which set
 RESUME=/path/to/swap/volume
 ROOT=/path/to/root/volume
 
 However even after doing so, the boot loader doesn't find them.

You mean after doing so and building another initrd image?

 So I must have been doing something wrong.
 Any other suggestions?

Did you try passing 'root=DEV'?

Where does the boot process get stuck? What are the last few lines?

Can you try 'break=premount' and debug a bit from there - e.g. see
if you can manually mount the new root, etc.?

You might also try dracut, BTW, and even yaird (which isn't in squeeze) -
I have no experience with either of them.

You can also try booting using some kind of rescue system (Debian's
bootable media or something else) and then mkinitrd again.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: iconv_open fails when suid bit is on

[Yedidyah Bar-David]

On Mon, Feb 13, 2012 at 11:45:26AM +0200, guy keren wrote:
 
 running strace on an suid binary - ignores the 'suid' bit. so the
 test with strace is not relevant.

Indeed, and to strace programs that do this, I do something like that:
Open a root shell:
# echo $$
let's say it printed 1234, so its pid is 1234


Open another root shell:
# strace -f -o /tmp/trace1 -p 1234

Now, inside 1234 run what you need to debug.
'strace -f -p pid' does follow correctly the uid changes caused by
suid binaries and setuid etc.

Some years ago I played a bit with the more obvious
# strace -f su - user -c program
which then did not work well. Not sure if this was changed. '-f -p pid'
always worked for me.

Sorry I can't comment about the actual OP quesion, but just a guess -
perhaps it's on nfs?
-- 
Didi

 On 02/13/2012 10:56 AM, Elazar Leibovich wrote:
 In RHEL 5 system, libc-6, I'm seeing the following strange phenomena
 
 $ cat iconv_test.c
 #include stdio.h
 #include errno.h
 #include fcntl.h
 #include iconv.h
 
 void iconv_test() {
static int nr = 0;
iconv_t iconv = iconv_open(MSCP949,UTF-8);
//iconv_t iconv = iconv_open(UTF-16,UTF-8);
if (iconv == (iconv_t)-1) {
  puts( can't initialize iconv);
} else {
  puts( iconv open success! );
}
nr++;
 }
 
 int main(int argc,char **argv) {
iconv_test();
return 0;
 }
 
 $ gcc iconv_test.c
 $ ./a.out
 iconv open success!
 $ sudo su -
 # chown root:foo a.out
 # chmod 4555 a.out
 # su foo -
 $ ./a.out
 can't initialize iconv
 $ strace ./a.out 2/dev/null
 iconv open success!
 
 
 iconv_open on UTF-16 to UTF-8 succeeds!
 This phenomena doesn't happen in recent Ubuntu.
 
 I'm not familiar with the inner workings of iconv, but stracing a good
 iconv run reveals it dlopen so files according to the chosen encodings,
 maybe it's related.
 
 1) I'll be glad for any thoughts or ideas how to debug this issue, other
 than downloading the libc source rpm, compiling it, LD_PRELOAD, and hope
 the problem will be recreated.
 
 2) If someone can test this on a RHEL-5 machine, and report if it
 happens to him too, it could be helpful.
 
 Thanks,
 
 
 ___
 Linux-il mailing list
 Linux-il@cs.huji.ac.il
 http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
 
 
 ___
 Linux-il mailing list
 Linux-il@cs.huji.ac.il
 http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Modern Linux memory management

[Yedidyah Bar-David]

On Thu, Jan 26, 2012 at 10:54:56AM -0500, Ori Berger wrote:
 On 01/26/2012 10:16 AM, Baruch Siach wrote:
 
 Only by using valgrind, that I could find the exact location and figure
 out, that it was another function that had the problem.
 
 How does the modern memory management system is working then, that it takes
 so much time for the problem to surface ?
 
 Now, if you corrupt the internal glibc data structure, glibc won't notice
 until you try to call one of malloc(), free(), etc.
 
 And in addition to what Baruch said:
 
 Valgrind will always catch these errors, but will result in
 significant slowdown (x10-x20). There are tools like DUMA (and its
 earlier incarnation, Electric Fence) incur almost no CPU overhead
 and can detect many kinds of corruptions as soon as they happen, by
 using the memory management units.
 
 (Because of the MMU granularity, you need to run your program twice
 - one in which allocations are aligned to the lower address, and one
 when they are aligned to the top address)
 
 There is also a middle ground; gcc's mudflap
 http://www.stlinux.com/devel/debug/mudflap and -- if your program
 is pure C and can be compiled by tcc,
 http://bellard.org/tcc/tcc-doc.html#SEC21; These are comparable to
 valgrind in functionality (for code you compile with them; standard
 library code runs at full speed/unchecked), but usually only
 introduce a small slowdown (10% or so).

BTW, in case you do not know Fabric Bellard, he does lots of very cool
stuff - he is the author of qemu, and every now and then I give a look
at his site. I now did because of this post and found out an amazing
project of a PC Emulator in Javascript:
http://bellard.org/jslinux/
The guy is simply amazing.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Israeli website on Linux: beer-sheva.muni.il

[Yedidyah Bar-David]

On Wed, Jan 25, 2012 at 08:45:51AM +0200, shimi wrote:
 On Wed, Jan 25, 2012 at 8:38 AM, Dotan Cohen dotanco...@gmail.com wrote:
 
  Can anyone reach beer-sheva.muni.il on Linux? I tried in Chrome and in
  Firefox, I get a timeout as if the server is not running. When I try
  in Firefox on the wife's Windows7 laptop, the site comes right up. I
  have not tried to spoof a Windows UI string, but can anyone confirm
  that the site isn't responding to Linux browsers?
 
 
 Works for me. Gentoo, FF 9.0.1.
 
 I had to prefix with www though. Appears they don't have an A record on the
 zone itself. Someone didn't read http://no-www.org 

Still doesn't work for me. www.beer-sheva.muni.il resolves for me to
147.236.237.102 and connect to port 80 failes. Tried from several
different machines and ISPs, both Israeli and abroad.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Israeli website on Linux: beer-sheva.muni.il

[Yedidyah Bar-David]

On Wed, Jan 25, 2012 at 08:52:55AM +0200, Noam Meltzer wrote:
 On Wed, Jan 25, 2012 at 8:38 AM, Dotan Cohen dotanco...@gmail.com wrote:
 
  Can anyone reach beer-sheva.muni.il on Linux? I tried in Chrome and in
  Firefox, I get a timeout as if the server is not running. When I try
  in Firefox on the wife's Windows7 laptop, the site comes right up. I
  have not tried to spoof a Windows UI string, but can anyone confirm
  that the site isn't responding to Linux browsers?
 
 
 Hi,
 I have no problem accessing the site from firefox9.0.1 on Fedora15.
 Only thing is that the URL must include the prefixing www.

Those that can access it - it might help debug this if you also post
other relevant information, such as the IP address you connect to and
ISP you connect through.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Israeli website on Linux: beer-sheva.muni.il

[Yedidyah Bar-David]

On Wed, Jan 25, 2012 at 08:38:22AM +0200, Dotan Cohen wrote:
 Can anyone reach beer-sheva.muni.il on Linux? I tried in Chrome and in
 Firefox, I get a timeout as if the server is not running. When I try
 in Firefox on the wife's Windows7 laptop, the site comes right up. I
 have not tried to spoof a Windows UI string, but can anyone confirm
 that the site isn't responding to Linux browsers?

Tried both chrome on linux and IE on XP, from two different connections,
neither worked. dns lookups show that there is no A record for it, there
is one A record for www.beer-sheva.muni.il, and telnet to port 80 on it
does not manage to connect. Are you certain that it worked for you on
Windows? On the specific configuration I tried, IE searched google for
it after failing. Perhaps that's what you saw?
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Wikis (Was: Re: Video Tape encoding)

[Yedidyah Bar-David]

On Sun, Jan 15, 2012 at 10:36:57PM +0200, Dov Grobgeld wrote:
 As promised, here is my write-up of the steps I did to capture a video tape
 under Linux:
 
 http://www.wikihow.com/index.php?title=Encode-a-Video-Tape-to-Mpeg4-%28Divx%29-Under-Linuxaction=history
 
 Feel free to edit and discuss on the wikipage.
 
 This is the first time I'm using wikihow, and I'm not really sure it is an
 appropriate place for hosting such explanations. If someone has a better
 suggestion, please let me know.

Never stumbled upon it. It might be interesting, I don't know.
I saw several significant FOSS projects, including vim and requesttracker,
that moved their own self-maintained wikis to wikia.com, founded by the
founders of wikipedia. Both of them are (obviously?) powered by mediawiki,
so for simple uses there might not be a big difference.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: HTML Mail (was: Re: Which technology should I learn to do this?)

[Yedidyah Bar-David]

On Sun, Jan 08, 2012 at 09:38:21AM +0200, Dotan Cohen wrote:
 On Sun, Jan 8, 2012 at 09:15, Yedidyah Bar-David
 linux...@didi.bardavid.org wrote:
  Should I officially give up and declare text email as a thing of the
  past? Should I move to some graphical mail client that displays and
  composes html mail? I tried Thunderbird a few times, for a few days
  each, over the last 5 or so years, and always went back to mutt.
 
  I see on this list some people that still use mutt. I also have a
  feeling that some of the past users of mutt on this list moved to html
  mail clients.
 
  I do not intend to start a flame war. Feel free to ignore, give advice,
  share your struggles, etc...
 
 
 
 For my personal email I use Thunderbird with text email. Why do you
 want to give up on text email? HTML email is the problem, not text.

Well, at least you and Shachar, which happen to also be the participants
in this discussion - but I think it's just a coincidence, happen to send
html-only to the list. mailman added a signature as text-only, and mutt
shows me the text by default. So I have to 'v' and choose the html part
to see the actual message. But do not take this persoanally - I think
that's a symptom, while the real issue is that no significant part of the
population uses text email, and so those who do see it becoming less and
less comfortable.

BTW, if you were under the impression that you use TB BUT send text only,
you have just discoverred that at least in some cases it sends html even
when you ask it to send text only.

My most recent real issue was with the way some mail clients format the
replies, and the way lynx (also tried others) renders them to text,
which in certain cases makes it impossible to understand who wrote what.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: HTML Mail

[Yedidyah Bar-David]

On Sun, Jan 08, 2012 at 10:03:27AM +0200, Dotan Shavit wrote:
 On 01/08/2012 09:15 AM, Yedidyah Bar-David wrote:
 On Sun, Jan 08, 2012 at 08:50:45AM +0200, Dotan Shavit wrote:
 ___
 Linux-il mailing list
 Linux-il@cs.huji.ac.il
 http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
 Dear Linux-IL subscribers,
 
 Should I officially give up and declare text email as a thing of the
 past?
 Sorry.
 As a Thunderbird user I'll have to remember to select Plain text
 only for each message to this list...
 Probably one of my plugins' fault.

Sorry, I mixed the two Dotan's...

 
 Any way, wouldn't lynx display HTML only messages?

No - as explained in a reply to the other Dotan, mailman adds a text
part so mutt displays it only.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

HTML Mail (was: Re: Which technology should I learn to do this?)

[Yedidyah Bar-David]

On Sun, Jan 08, 2012 at 08:50:45AM +0200, Dotan Shavit wrote:
 ___
 Linux-il mailing list
 Linux-il@cs.huji.ac.il
 http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Dear Linux-IL subscribers,

Should I officially give up and declare text email as a thing of the
past? Should I move to some graphical mail client that displays and
composes html mail? I tried Thunderbird a few times, for a few days
each, over the last 5 or so years, and always went back to mutt.

I see on this list some people that still use mutt. I also have a
feeling that some of the past users of mutt on this list moved to html
mail clients.

I do not intend to start a flame war. Feel free to ignore, give advice,
share your struggles, etc...
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: What is residual memory? Can anyone explain?

[Yedidyah Bar-David]

On Mon, Dec 19, 2011 at 08:31:55AM +1100, Amos Shapira wrote:
 On 18 December 2011 18:07, Yedidyah Bar-David
 linux...@didi.bardavid.org wrote:
  In addition, I also check free memory before running the app, after
  starting, several times, etc.
 
 That's a good practice as well.
 
 Do you do it manually or do you have some background tool which
 collects this data constantly for review
 (collectd/ganglia/zabbix/something else?)?

I personally use none, but can see the value of using some.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: What is residual memory? Can anyone explain?

[Yedidyah Bar-David]

On Fri, Dec 16, 2011 at 09:22:43PM +0200, Raz wrote:
 i tend to calculate a process' memory consumption in deltas, meaning,
 how memory it consumes in addition to the shared sections. So i
 usually take a look at the write-able
 and private consumption as reported by pmap. ( pmap pid ).
 Please note, there are numerous discussions of how to estimate
 system/process memory consumption. Consider for example an in-kernel
 inode, it is not counted in the process address space even though it
 serves the process, what about in-flight bio, skbuff ?
 
 
 These problem are very common in the embedded world, we  fear of being
 oomed :)

In addition, I also check free memory before running the app, after
starting, several times, etc.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Unix History: Why does hexdump default to word alignment?

[Yedidyah Bar-David]

On Thu, Dec 01, 2011 at 01:55:24PM +0200, Nadav Har'El wrote:
 On Thu, Dec 01, 2011, guy keren wrote about Re: Unix History: Why does 
 hexdump default to word alignment?:
  apparently, you did not use binary data serialization in the past
  two decades. when you serialize data and store it into a file (also
  on the network), it is very useful to be able to see the data as
  2-byte or 4-byte or whatever-byte numbers, when debugging.
 
 Well, for debugging you typically use tools like a debugger (gdb, ddd,
 etc.) or network sniffer or something - and those have their own methods
 of displaying data, and do not use od. So using the actual od command
 in a shell or shell-script is not something I ended up doing in recent years.
 I don't think I even noticed the new hexdump sibling of od cropped up
 in Linux ;-)

Regarding new siblings of od, and just in case someone expects a useful
piece of information in this thread - I happened to use several times
xxd, which can also do the reverse - convert its output back to binary -
so you can use it together with your $EDITOR as a poor man's binary
editor. I guess people used uuencode/uudecode for this in the past,
perhaps I did too, but xxd is much more comfortable.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: NFS + NIS madness

[Yedidyah Bar-David]

On Mon, Nov 28, 2011 at 11:35:33PM +0200, Hetz Ben Hamo wrote:
 Hi,
 As you can see, all of them (from root) works perfectly:
 
 [root@client ~]# ypcat passwd.byuid
 vic:$6$FcNMjbbl$8wGzWhtEK9P0.WdoqE78xI9VDzmaH1wTF.2vax9VERW3uiqhytNjXXzVccCjnWRxV7ApHL.JibC0Ar4spM6In1:500:500:Vicky
 the kitten!:/home/vic:/bin/bash
 [root@client ~]# ypmatch 500 passwd.byuid
 vic:$6$FcNMjbbl$8wGzWhtEK9P0.WdoqE78xI9VDzmaH1wTF.2vax9VERW3uiqhytNjXXzVccCjnWRxV7ApHL.JibC0Ar4spM6In1:500:500:Vicky
 the kitten!:/home/vic:/bin/bash
 [root@client ~]# getent passwd 500
 vic:$6$FcNMjbbl$8wGzWhtEK9P0.WdoqE78xI9VDzmaH1wTF.2vax9VERW3uiqhytNjXXzVccCjnWRxV7ApHL.JibC0Ar4spM6In1:500:500:Vicky
 the kitten!:/home/vic:/bin/bash
 
 only when I do su - vic or login as vic, then it happens..

Can you try these as vic?

It's been some time since I last debugged NIS, but IIRC there are
mechanisms that try to prevent normal users from seeing some stuff,
e.g. shadow passwords. Obviously you did not use shadow passwords,
judging from the above. Perhaps the server replies to these only from
low ports or something like that, don't remember.

You can try to run the server with debugging/verbose/whatever and see
if there is anything in its logs.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: remote directory/partition

[Yedidyah Bar-David]

On Sat, Oct 22, 2011 at 11:03:49PM +0200, Hetz Ben Hamo wrote:
 Hi,
 
 Here is a theoretical question:
 
 Lets say I have a Linux server in Israel, and I have a block of storage
 (lets say iSCSI partition for this example) in USA, and I want to mount it
 on my server in Israel.
 iSCSI over such a long distance and with big latency (thanks to our ISP's)

Not sure it's mainly the ISPs, BTW. You do also depend on the physics of
speed of light.

 is a big no no, it's too slow. NFS is also not a good idea (here's
 whyhttp://goo.gl/vn4GM
 ).
 
 I can take this storage, format it and export it from my server in USA, but
 which protocol would give me:
 
1. All (or almost all) functionality of a local mounted device

Do you need it read/write on both sides? If so, you are going to have
big problems if the link is cut.

2. Can work with long distance latencies
3. won't kill the machine if the remote directory is disconnected /
disappeared
4. If possible - supported (either directly or using 3rd party driver) on
Windows 2008 (Linux is the main concern, Windows is optional)

I used drbd on a LAN, and know that it can theoretically work rather well
on larger distance when used as read-write on one side only. They also
have a pay-for tool to do this asyncronously called drbd proxy. This
implies using a local copy and have drbd sync it. You can choose between
three what they call Protocols to affect the perceived local latency.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: changes-only filesystem

[Yedidyah Bar-David]

On Sun, Sep 25, 2011 at 02:03:06PM +0300, Oleg Goldshmidt wrote:
 2011/9/25 Erez D erez0...@gmail.com:
 
  There used to be a filesystem which one would mount over a read-only
  filesystem, and that filesystem would only hold the changes.
 
  anyone remembers which filesystem it was ?
 
 UnionFS?
 
 There is also MVFS from Rational (the basis of ClearCase), but it's
 proprietary and I doubt you meant it.

A few years ago lwn published a two-part review on a few such things:
http://lwn.net/Articles/324291/
http://lwn.net/Articles/327738/
You might read them through if interested, or just use as a list of
pointers to projects.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: HW/SW for license server

[Yedidyah Bar-David]

On Fri, Sep 23, 2011 at 01:19:51AM -0700, Valery Reznic wrote:
 Hi, All.
 
 I need to setup license server and looking for advice what hardware/software 
 to use
 
 
 General schema is following:
 
 There are about 20K clients all of them periodically (let say once a week) 
 query license server for license via https
 
 I'll try my best to spread all those requests over the week, but in worse 
 case scenario all those 20K request can came during 1hour,
 spread relative evenly i.e about 6 request/sec
 
 When license server get request it query database ~20K records with 2-3 small 
 fields, generate license and return it to client.
 
 License generation is light - md5sum of the very small input ( few hundreds 
 bytes)
 
 I think use following software for it
 
 
 1. CentOS 6
 
 2. Apache with mod_python
 3. Mysql

Just a short note - make sure you configure mysql to do query caching.
google for 'mysql query cache'. It wasn't on by default in centos5, I
did not check 6 yet. It has, just as any cache, the potential to improve
performance dramatically.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Thinkpad T520

[Yedidyah Bar-David]

On Mon, Sep 12, 2011 at 04:11:49PM +0400, Michael Vasiliev wrote:
 On 09/06/2011 11:49 AM, Yedidyah Bar-David wrote:
  Hi all,
 
  I am considering ordering a Lenovo Thinkpad T520, and have the following
  questions. I intend to use it with Debian, can go with Ubuntu if it makes
  a difference.
 
  1. It has Integrated Mobile Broadband and an optional Gobi 3000.
  Will they work in Israel? Am I limited to specific suppliers? Is there
  a reason for the extra $125 for a Gobi 3000, assuming I'll use it only
  in Israel?
 It is important you check with the cellphone providers about
 compatibility. And then you make sure that the hardware you receive is
 not locked to any US provider.

OK, I'll remember mentioning this in the purchase order, although I
can't believe a decent Israeli supplier will sell a US-only modem without
mentioning that. A Decent one...

  A salesman now told me that it's not integrated but I have
  to buy some Erricsson modem (part number 0a36186).
 ׂI think you're being screwed. Gobi 3000 is manufactured by Quallcom,
 it's an integrated add-on card that goes inside your laptop into the
 PCIE slot (not into PCCard/ExpressCard slot on the side). AFAIK, it in
 impossible to install one without full disassembly of the laptop. It is
 a complete communication solution, and no additional hardware is needed.
 While it can be possible to install both cards, depending on your
 configuration, there is only two antennas and both cards would like to
 use both.

I am not sure you understood what I asked about, which I also might have
understood incorrectly. Noone said I should buy both the Erricsson and
the Gobi 3000.

Lenovo's site talks about just two options: Either use the Built In
Integrated Mobile Broadband, for no additinal money, or buy an extra
Gobi 3000. The salesman I talked to said there is no Build In
Integrated Mobile Broadband - that if I want mobile internet, I have
to buy the Erricsson modem.

What I currently understand (perhaps partially, not sure) is that the
site I looked at is for US, and US versions always have the Integrated
Mobile Broadband, which do work in the US but not in some other coutries,
and the Gobi 3000 should work Almost Everywhere (perhaps not in the
mathematical meaning of the notion), and the Israeli guy I talked to
talks about Israel models, which do not have anything intergrated, and
for which one can buy the Erricsson modem which works in Israel (and
perhaps other countries, perhaps not in the US).

All of these options are PCIe, built inside the laptop, and require
(partial?) disassembly to install, and so would better be done by the
supplier.

 
  2. It has a Thinkpad b/g/n Integrated WiFi and few optional extras.
  Is there a reason to pay for for one of the options, assuming I do
  not have extreme needs of either bad reception or high bandwidth?
  Will any of them work better in linux? FWIW, I currently have a T42
  with its integrated 2200BG which basically works well, but recently
  I did not manage to connect to few networks - IIRC I saw somewhere it's
  a (fixed) bug in the dhcp client but I did not bother debugging.
 I have a Thinkpad W700, which is a somewhat similar model based on a
 different chipset and a previous generation of the wireless core. I
 haven't got to setting it up on Gentoo yet, but it worked reasonably
 well on Ubuntu and yet had no failures in connecting to any properly set
 up network.

Not sure they were properly set up, they were not set up by me, and
many people used them successfully, in one case an Ubuntu user that sat
next to me (I had Debian Squeeze with some manual configuration).

 
 Judging from a recent thread on this very list, Micha Feigin
 mi...@post.tau.ac.il has the same base model (W520) as you intend to
 buy. I suggest you try to mail him for all the gory details.

Look, as far as I know, all thinkpads that ever had built in wifi, worked
well in Linux - if not from the beginning, after some time. My main
concern is why are there several other options in Lenovo's site - are they
only for specific rare needs (e.g. long range/low signal) or I should
spend time understanding them because they might be relevant for me, as
an Israeli user whose only special need is that it works in Linux.

Anyway, ik (Ido Kanner?) already emailed a few days ago to this list
saying that the builtin wifi and the Erricsson modem in a similar model
both work well under Linux, so I think I can safely go on with my order.
I'll of course appreciate more relevant input, especially some that
contradicts my current understanding.

Thanks everyone,
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Jewish Nanny-Cam

[Yedidyah Bar-David]

On Mon, Sep 12, 2011 at 02:11:41PM +0300, geoffrey mendelson wrote:
 
 On Sep 12, 2011, at 1:59 PM, Steve G. wrote:
 
 Could someone recommend a security system that would allow me to
 keep an eye on the place from the US or Central America? Is there
 a system that I can connect wirelessly to the DSL modem, which
 will broadcast its name or ip address for access (without a
 membership in a for-fee service)? If not what hardware/software do
 you recommend for ubuntu (preferably) or windows xp that can
 accomplish the same through the computer?
 
 For that I use DYNDNS.org. They allow you up to 5 names (per free
 membership) using their domains. It's not very good for a commercial
 operation, but for someone who wants what is a random domain name
 it's fine. Every router I have seen in the last 5 years supports
 them out of the box.
 
 If you wanted to keep your own domain name, you can use them for the
 DNS servers. It's about $30 a year, but they only take credit cards,
 not paypal.

And for that matter, there are other providers that give you dns hosting
for free too, for a few domains - there was a thread about them here
some time ago, and a few I can recall - zoneedit, afraid.org, dnsexit.

About monitoring software - I never used any, but there are a few for
linux, probably built into ubuntu. apt-cache search on my Debian Squeeze
finds camstream and zoneminder.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Thinkpad T520

[Yedidyah Bar-David]

On Fri, Sep 09, 2011 at 04:38:12PM +0300, ik wrote:
 On Thu, Sep 8, 2011 at 19:20, ik ido...@gmail.com wrote:
  1. It has Integrated Mobile Broadband and an optional Gobi 3000.
  Will they work in Israel? Am I limited to specific suppliers? Is there
  a reason for the extra $125 for a Gobi 3000, assuming I'll use it only
  in Israel? A salesman now told me that it's not integrated but I have
  to buy some Erricsson modem (part number 0a36186).
 
 
  The cellular modem works well for me. I entered the SIM and connected it
  and it worked.
  I still haven't install Linux on it, but with the Windows version it does
  work.
 
 I'm answering you using the internal cellular modem.

Two more questions:

Can you please tell the exact model of the cellular
modem, both Lenovo part numbers etc (from the order or whatever) and
pci IDs?

Also - what cellular provider do you use it with? Do you know if it
matters? Does it also support Pelephone frequencies?

Thanks a lot for the encouraging report!
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Thinkpad T520

[Yedidyah Bar-David]

Hi all,

I am considering ordering a Lenovo Thinkpad T520, and have the following
questions. I intend to use it with Debian, can go with Ubuntu if it makes
a difference.

1. It has Integrated Mobile Broadband and an optional Gobi 3000.
Will they work in Israel? Am I limited to specific suppliers? Is there
a reason for the extra $125 for a Gobi 3000, assuming I'll use it only
in Israel? A salesman now told me that it's not integrated but I have
to buy some Erricsson modem (part number 0a36186).

2. It has a Thinkpad b/g/n Integrated WiFi and few optional extras.
Is there a reason to pay for for one of the options, assuming I do
not have extreme needs of either bad reception or high bandwidth?
Will any of them work better in linux? FWIW, I currently have a T42
with its integrated 2200BG which basically works well, but recently
I did not manage to connect to few networks - IIRC I saw somewhere it's
a (fixed) bug in the dhcp client but I did not bother debugging.

I already searched a bit the net, so I'd like mainly advices from people
who actually have one (or similar) and real experience.

Thanks,
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: checking wireless cable

[Yedidyah Bar-David]

On Thu, Aug 18, 2011 at 10:02:14PM +0300, Oleg Goldshmidt wrote:
 Has anyone encountered anything similar? Any suggestions for things I
 should try?

What is the output of 'iwconfig' and of 'iwlist scanning', both as root?
I have a feeling it won't say much, but just in case.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Thunderbird mailer

[Yedidyah Bar-David]

On Thu, Aug 04, 2011 at 03:17:36PM +0300, Stan Goodman wrote:
 So far, I have not found a way to write a Hebrew message with lines
 beginning at the right side of the window. With Kmail it was obvious
 --- there were buttons for right, left, and center positioning.
 Where is something equivalent in Thunderbird?

Here:
https://addons.mozilla.org/en-us/thunderbird/addon/bidi-mail-ui/
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Thunderbird mailer

[Yedidyah Bar-David]

On Thu, Aug 04, 2011 at 03:23:22PM +0300, shimi wrote:
 On Thu, Aug 4, 2011 at 3:17 PM, Stan Goodman 
 stan.good...@hashkedim.comwrote:
 
  So far, I have not found a way to write a Hebrew message with lines
  beginning at the right side of the window. With Kmail it was obvious ---
  there were buttons for right, left, and center positioning. Where is
  something equivalent in Thunderbird?
 
 
 https://addons.mozilla.org/en-us/thunderbird/addon/bidi-mail-ui/

Do you think there is some deep hidden relation between the fact that
both of us replied almost the same answer, almost at the same time,
and our email addresses for this list?

:-)
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Security patches for Apache 1.3.x?

[Yedidyah Bar-David]

On Thu, Jul 14, 2011 at 04:29:00PM +0300, Ira Abramov wrote:
 howdie!
 
 I have an embeded system (roughly based on CentOS 3) with a few legacy
 components, one of which is Apache 1.3.42, which has served us well this
 far, but now we bumped into these:
 
 http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2011-1928
 http://web.nvd.nist.gov/view/vuln/detail?vulnId=CVE-2011-0419
 
 Since the Apache 1.x line is EOL and I don't see this package has been
 maintained with sec patches by Debian or even RHEL (correct me if I
 missed anything)
 
 Before I'm forced to rock the boat with a move to Apache2, lighty or
 nginx, is there a source for patches for this that I missed?

You might consider RedHat's Extended Lifecycle Support. I do not see
freely distributable SRPMs for it - not sure why, whether that's legal
etc.

I used to compile and use apache 2.x on RHEL/CentOS 3 with no problem.
It will obviously require reviewing your config/modules/etc which might
be a significant task...
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: USB I/O draining my userspace (Ubuntu Natty 64b)

[Yedidyah Bar-David]

On Wed, Jun 15, 2011 at 06:42:48PM +0300, Ira Abramov wrote:
 Quoting Yedidyah Bar-David, from the post of Wed, 15 Jun:
  Did you try cp (or cp --sparse=always if you really want to) and
  see if it helps? I don't know of a similar option for dd.
 
 I have here an image with partitions in it, which is why I use DD. cp
 can't help me, sadly.

Why not cp?
cp image /dev/sdX
should work with no problem. Indeed you won't have control on the block
size, but I can't think of other important options of dd that you'll
miss.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: USB I/O draining my userspace (Ubuntu Natty 64b)

[Yedidyah Bar-David]

On Wed, Jun 15, 2011 at 04:19:08PM +0300, Ira Abramov wrote:
 First, an appology. I added another message on the RMS fiasco only to
 find out later the threads all died out and I was committing a faux pas.
 my appologies! please ignore I posted it if at all possible.
 
 Second, here's my problem: I have here a workstation running an Athlon
 3700+, and part of my job is to occasionally write out an image file to
 USB universal card reader, testing the product of my builds. The writing
 takes forever (since I haven't discovered how to get dd to write out the
 sparse image to the CF card sparsely). Also, untill I moved the card
 reader to one of the backpanel ports, the write would drag my entire
 environment to a halt at the same time - even the mouse pointer gets
 stuck at some point, until dd would finish. switching from front to back
 panel and adding the oflag=dsync option solved the freַ¯ing of the
 userspace but not the horrible writing speeds.

Perhaps it uses USB1 and not 2?
Can you rmmod all *hci_ucd modules except for ehci_hcd and see what
happens?

Did you try cp (or cp --sparse=always if you really want to) and
see if it helps? I don't know of a similar option for dd.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Palm to Android via Linux

[Yedidyah Bar-David]

On Wed, Jun 01, 2011 at 10:22:32AM +0300, shimi wrote:
 2011/6/1 Mordecha Behar mordecha.be...@mail.huji.ac.il
 
  I've encountered the opposite problem.
  Files I create in a Linux text editor (Kwrite) don't open properly in
  Windows Notepad. I get the entire file on one line.
  Maybe there's something strange going on behind the scenes in KDE?
 
 
 
 That's something else and relates to the way Unix and Windows define
 newline.
 
 You could run your file through the unix2dos utility prior to moving it to
 Windows, and it will most probably solve your issue.
 
 Alternatively, if you'll be using Kate (which I think is better than
 KWrite), you have Tools - End of Line - UNIX (and you can change there to
 Windows/DOS)

You can also open it in windows with Write (sometimes called Wordpad)
by right-click - Open with - Write (or wordpad). It knows Unix newlines.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Problems with ubuntu and default file manager

[Yedidyah Bar-David]

On Mon, May 16, 2011 at 10:10:48AM +0300, Maayan Eshed wrote:
 Just a thought.. have you tried to uninstall and reinstall chrome?

Assuming the configuration is per-user, I'd first try to create another
user, login with it, and see if it works. If it does, you might try to
compare the configuration between the two accounts.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Problems with ubuntu and default file manager

[Yedidyah Bar-David]

On Mon, May 16, 2011 at 11:05:43AM +0300, Nitzan Brumer wrote:
 Maayan - As I've said - its not only in chrome and happens when ever I try
 to open something with file manager.
 Its looks like I've override the file manager application with something
 else.
 
 Didi - Sounds like an option, though which config file ?

I never used xfce - no idea. If you do know it's under ~/.local, do
something like (probably as root)
diff -ur ~user1/.local ~user2/.local | less
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Are Android questions on topic here?

[Yedidyah Bar-David]

On Wed, May 11, 2011 at 12:56:28PM +0300, Oleg Goldshmidt wrote:
 Question to the (non-existent) LINUX-IL cabal: this is a list for
 technical questions on Linux. Some Android issues have been discussed
 here. Are technical Android discussions officially on topic or off
 topic? Can we decide once and for all?

Isn't Android Linux? Why is there a question? Did anyone ever try to
imply that Android is off-topic?

BTW, once upon a time it was defined as Linux AND Israel.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How to check initrd's contents? (was: Re: Disk I/O as a bottleneck?)

[Yedidyah Bar-David]

On Mon, May 09, 2011 at 03:18:08PM +0300, Omer Zak wrote:
 My kernel is configured to have AHCI as a module:
 CONFIG_SATA_AHCI=m
 
 However I understand that it means that this module is needed also in
 the initrd image.  How can I check which modules made it to the initrd
 image last time it was rebuilt?

something like

mkdir /some/temp/dir
cd /some/temp/dir
gzip -dc  /boot/my-initrd | cpio -idm

That is, assuming it's a ramfs-based initrd and not an old-style
filesystem image.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Disk I/O as a bottleneck?

[Yedidyah Bar-David]

On Sun, May 08, 2011 at 07:28:49AM +0300, Nadav Har'El wrote:
 On Sat, May 07, 2011, guy keren wrote about Re: Disk I/O as a bottleneck?:
  and if you have a lot of money to spend - you could consider buying an
  enterprise-grade SSD (e.g. from fusion I/O or from OCZ - although for
  your use-case, some of the cheaper SSDs will do) and use it instead of
  the hard disks. they only cost thousands of dollars for a 600GB SSD ;)
 
 Instead of buying a huge SSD for thousands of dollars another option you
 might consider is to buy a relatively small SSD with just enough space to
 hold your / partition and swap space. Even 20 G may be enough.
 The rest of your disk - holding your source code, photos, songs, movies,
 or whatever you typically fill a terabyte with, will be a normal, cheap,
 hard disk.
 
 Several of my friends have gone with such a setup on their latest computer,
 and they are very pleased.

I am considering, for my next laptop, and taking into account the fact
that most laptops do not have space for two disks but do have some kind
of flash memory slot (card reader) - usually SD-something, to have the
OS on a (e.g.) SD card of 16 or 32 GB. I have no other experience with
such cards, so I do not know if they are considered durable enough, fast
enough - both random and sequential IO, both compared to SATA mechanical
disks and to SATA flash ones, etc. Comments are welcome :-)
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: A little GPL riddle (was: GPL as an evaluation license)

[Yedidyah Bar-David]

On Mon, Apr 11, 2011 at 12:33:05AM +0300, Aviad Mandel wrote:
 For the sake of fairness, I'll say that the questions I had about my own
 little venture are answered, thank you all.
 
 But for pure curiosity, I'm left wondering how effective GPL really is.
 
 Let's leave the microwave for a second, and think about a proprietary web
 software browser, for a desktop, using a lot of GPLed code, and should
 remain closed source.
 
 The trick is like this: The installation program generates the executables
 by compiling the GPLed sources (Gentoo style) and linking them with the
 proprietary object code. What you get is a binary nobody is allowed to copy,

How is this proprietary object code generated? If for compiling it to
object code you use header files which are under the GPL, your object
files are derived work. IANAL.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: OT: Einstein Writer under QEMU

[Yedidyah Bar-David]

On Tue, Mar 29, 2011 at 09:37:06AM +0200, geoffrey mendelson wrote:

 On Mar 29, 2011, at 8:45 AM, geoffrey mendelson wrote:

 I'm trying to help someone run Einstein Writer under QEMU. I set up a 
 virtual machine and installed FreeDos on it.

 I found a demo of Einstein which is supposed to be a running version  
 of it.

 When I start it up, I get a few English words and a lot of junk on the 
 screen. I assume I am missing installing Hebrew support in the virtual 
 video card.

 I did a web search and found a wikipedia entry for codepage 862. I  
 tried to load it with the FreeDos command display con=(ega,862,1). It 
 loads, but I still get garbage.

 Any ideas?


 To answer my own question, I found a dos boot disk complete with the  
 necessary files and Einstein all on it. It was designed for a different 
 virtualization system, but it works with QEMU:

 http://masa.googlepages.com/eini.zip


 Alternatively is there a way to convert EinsteinWriter files to  
 something useable without Einstein itself?


 Still looking for an answer.

IIRC it was mostly text with rather simple controls (escape sequences)
for markup. Did you look at its files?
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Getting mouse buttons to work

[Yedidyah Bar-David]

On Tue, Feb 15, 2011 at 10:44:08AM +0200, Dotan Cohen wrote:
 Hi all, first off I must confess that this is a crosspost. I posted a
 similar question to the Debian list, then to the Fedora list, but it's
 not getting very far and I know that there are some smart folks here
 who could probably help.
 
 I have a nice new Teac OX-1100 mouse with two extra multimedia buttons
 that supposedly perform the Zoom functions in Windows. When I try to
 get the scancodes with xev, xbindkeys -k, showkey, or showkey
 -s then I see no output. There are some other functional keys on this
 mouse which also show no output with those tools, such as the side
 scroll feature, so I suspect that there must be _some_ way to get the
 codes.
 
 Other than Xev, Xbindkeys, and Showkey, what tools can I use? I
 specifically bought this mouse because of the location of the two
 extra keys, as I have a manual disability and cannot use the special
 keys of regular mice.
 
 This is the rodent:
 ✈ganymede:~$ grep -i mouse /var/log/Xorg.0.log
 [15.197] (==) RADEON(0): Silken mouse enabled
 [15.289] (II) Microsoft Natural® Ergonomic Keyboard 4000: Found 1
 mouse buttons
 [15.289] (II) Microsoft Natural® Ergonomic Keyboard 4000:
 Configuring as mouse
 [15.292] (II) config/udev: Adding input device MLK OX-1100
 wireless Laser Mouse (/dev/input/event2)
 [15.292] (**) MLK OX-1100 wireless Laser Mouse: Applying
 InputClass evdev pointer catchall
 [15.292] (**) MLK OX-1100 wireless Laser Mouse: Applying
 InputClass evdev keyboard catchall
 [15.292] (**) MLK OX-1100 wireless Laser Mouse: always reports core events
 [15.292] (**) MLK OX-1100 wireless Laser Mouse: Device: 
 /dev/input/event2
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Found 9 mouse buttons
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Found scroll wheel(s)
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Found relative axes
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Found x and y relative 
 axes
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Found absolute axes
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Found keys
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Configuring as mouse
 [15.300] (II) MLK OX-1100 wireless Laser Mouse: Configuring as keyboard
 [15.300] (**) MLK OX-1100 wireless Laser Mouse: YAxisMapping:
 buttons 4 and 5
 [15.300] (**) MLK OX-1100 wireless Laser Mouse:
 EmulateWheelButton: 4, EmulateWheelInertia: 10, EmulateWheelTimeout:
 200
 [15.300] (II) XINPUT: Adding extended input device MLK OX-1100
 wireless Laser Mouse (type: KEYBOARD)
 [15.301] (II) MLK OX-1100 wireless Laser Mouse: initialized for
 relative axes.
 [15.301] (WW) MLK OX-1100 wireless Laser Mouse: ignoring absolute axes.
 [15.301] (II) config/udev: Adding input device MLK OX-1100
 wireless Laser Mouse (/dev/input/mouse0)
 
 
 However, 10 of the 12 buttons work, not just the 9 that it found. I've
 tried to google a picture of the mouse, I see no info on Teac mice
 even on the Teac website. The buttons are zoom buttons that I
 suppose are activated by a Windows driver on the OS that the package
 states that it supports.
 
 Thanks in advance for any advice on how to continue!

I have no idea about the specific mouse or issue, but other places you
can check are:

1. Outside of X, do
od -tx1 /dev/input/mice
then press various buttons and see what happens.

2. Try playing with acpi/acpid. E.g., from the examples of acpid
- look at /usr/share/doc/acpid/examples/default{,.sh}
(or at least that's where they are on my laptop - Debian Lenny).
I personally managed to make Fn F7 move between internal/external
monitor by playing with it and an example I once found on google -
I think it was this one:
http://www.thinkwiki.org/wiki/Sample_Fn-F7_script

I have no idea if you can get acpi events from normal keys (not Fn)
and did not try this (yet?).
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Networking: How to add another router

[Yedidyah Bar-David]

On Sun, Feb 13, 2011 at 10:14:27PM +0200, Geoff Shang wrote:
 On Sun, 13 Feb 2011, Michael Tewner wrote:

 Shimi's solution will work - use a cross-over cable, though, in order to
 connect the switches together.

 hmm.  Is there an easy way to tell a crossover cable from a straight one? 
 I've never really had to worry about it before except in one specific  
 instance and I had many fewer ethernet cables in those days.

By the colors of the wires:
http://en.wikipedia.org/wiki/Ethernet_crossover_cable
http://en.wikipedia.org/wiki/File:Vergleich_2von2_Crossoverkabel.jpg
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: IPV6?

[Yedidyah Bar-David]

On Wed, Feb 02, 2011 at 10:35:14AM +0200, Hetz Ben Hamo wrote:
 Hi,
 
 I'm planning to roll out at my hosting business a special package of virtual
 machine with IPV6 for people and companies who would like to deploy their
 web site for people who want to use IPV6.
 (so far the adoption of IPV6 in Israel sucks, even my upstream provider
 [Netvision] don't offer it).

Not sure about the status for Israeli ISPs, but at least one is registered
at ISOC-IL's IIX as peering also IPv6:
http://isoc.org.il/iix/2x_list.html

 
 I would like to ask: does anyone knows which ISP here has dial-up plans with
 IPV6?
 (if I'll find that there isn't such a thing, perhaps I'll set up something,
 although this take some considerable investment)

You might consider using some tunneling service. There are a few that
provide some service for free. I personally did a small test with
Hurricane Electric, see http://tunnelbroker.net/
which worked well enough for the test - allowed me to run a client and
a server and connect to it. I did not measure bandwidth/latency.
Do you mean phone line dial-up when you refer to dial-up plans?
A 56Kbit line? I guess most if not all tunnel brokers will give you
more than that.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: total uptime between shutdowns

[Yedidyah Bar-David]

On Tue, Feb 01, 2011 at 10:00:52AM +0200, Tom Rosenfeld wrote:
 Hi Guys,
 
 Is there a tool that will show me the total uptime (availability) of a
 machine between reboots?
 That is, if the machine was up for 24 hours and then shutdown for an hour
 and then up for 23 hours I want an answer of 47.

I do not know about anything that does exactly this, but you might want
to look at 'ud - Uptime Daemon'. But...

 
 I need this for charge-back of departments using cloud computer which they
 sometimes turn off.

I wouldn't use something from inside the guest for this. How do they
turn it off? Using some interface you created for them? I'd stuff into
this interface the counting. You should also consider what happens when
they hibernate (e.g. Xen's 'xm save'/'xm restore'), in case you let
them do this. This does not cost you CPU time or RAM - except for the
memory image which takes space (and needs to be backed up) it's exactly
like shutting down. But the guest will not notice - you'll not see
inside it anything that says that it was shutdown and restarted.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: small linux hardware

[Yedidyah Bar-David]

On Mon, Jan 10, 2011 at 10:08:52AM +0200, Erez D wrote:
 I am looking for a linux hardware which would be small, low power and cheep
 
 I found the telit GE863-PRO (pdf:
 http://www.telit.com/module/infopool/download.php?id=725 )
 which is actually a gprs module + linux on arm9. it is 41.4x31.4x3.6 mm in
 size and i can get it in 50$ (in quantities)
 
 however as i do not need the gprs module, i thought i could find something
 smaller and cheaper...
 
 
 anyone knows of such a hardware ?

I recalled from a few years ago an ad/article talking about a linux
system the size of an RJ45 jack. A quick search finds this:
http://www.linuxfordevices.com/c/a/News/Linux-system-squishes-into-Ethernet-connector/
which links to netsilicon.com, which redirects to:
http://www.digi.com/products/embeddedsolutions/
which has some nice toys. No prices listed.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Try reproducing Firefox Bug No. 611511

[Yedidyah Bar-David]

Hi all,

I am a sysadmin in ISOC-IL for some years, so far did not identify myself
as such...

On Wed, Dec 01, 2010 at 09:09:20AM +0200, Shlomi Fish wrote:
 Hi all,
 
 Please try to reproduce this Firefox bug (that also happens in Opera and 
 Konqueror):
 
 https://bugzilla.mozilla.org/show_bug.cgi?id=611511
 
 [quote]
 When I access the URL (
 ftp://mirror.isoc.org.il/pub/mandriva/devel/cooker/i586/media/main/release/ ),
 firefox becomes unresponsive for many seconds, and the menus, buttons, tabs,
 etc. won't respond. This also happens with other pages and also HTTP pages.
 
 I tried it in safe mode and was able to reproduce it there. I'm on Mandriva
 Linux Cooker on a Pentium 4 2.4 machine. I recall this started only a few days
 ago. I'm using the firefox-latest-nightly.
 
 Reproducible: Always
 
 Steps to Reproduce:
 1. Type firefox -no-remote -safe-mode.
 2. Go to the URL.
 Actual Results:  
 Firefox become unresponsive.
 
 Expected Results:  
 Firefox should load the page while still allowing to interact with it.
 [/quote]
 
 I'm on bezeqint.net.il.
 
 Please comment on the bug report or here.

Well, a few notes:

1. It also happened now to me, using google-chrome 7.0.517.44-r64615.
It does not happen with lftp.

2. I looked at the server logs, and it seems to me to happen to all the
clients that use EPSV (Extended Passive mode, see rfc 2428 if interested).
At least most if not all of them seem to give up after up to around half
a minute and revert to normal Passive mode, which works.

3. It seems to be due to our filtering, which currently does not support
EPSV mode. I have no idea if iptables supports EPSV and how well - if
this is a serious issue, I can try looking at this. You are welcome to
vote by emailing our official address mirrormas...@isoc.org.il which
currently reaches Lior Kaplan and me.

4. The fact that a client, especially a sophisticated graphical one, gets
stuck for an extended time, is obviously a bug, so you are welcome to
continue with this venture of opening bugs and make them all behave more
nicely.

5. Since I do not know how, if at all, you can make firefox, and most
other clients, use normal Passive mode and not Extended, the simplest
workaround seems to me, for now, to be to use http. Any specific reason
for using ftp in this case?

6. I have no idea what caused it to start happening a few days ago. We
did not change anything on our side, as far as I know. Perhaps firefox
changed its behaviour on a version you recently updated to. I definitely
see in the logs such behaviour many months ago.

Regards,
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Try reproducing Firefox Bug No. 611511

[Yedidyah Bar-David]

On Wed, Dec 01, 2010 at 11:44:00AM +0200, Shlomi Fish wrote:
 Hi Yedidyah,
 
 On Wednesday 01 December 2010 10:08:16 Yedidyah Bar-David wrote:
  5. Since I do not know how, if at all, you can make firefox, and most
  other clients, use normal Passive mode and not Extended, the simplest
  workaround seems to me, for now, to be to use http. Any specific reason
  for using ftp in this case?
 
 Well, for a time, I had horrible download speeds when downloading using 
 http:// from http://mirror.isoc.org.il/ and it was better at using ftp:// so 
 that's what I used.

It was probably due to changes we did to our filtering and related stuff.
Please try http and tell us if you have a problem.

 
  
  6. I have no idea what caused it to start happening a few days ago. We
  did not change anything on our side, as far as I know. Perhaps firefox
  changed its behaviour on a version you recently updated to. I definitely
  see in the logs such behaviour many months ago.
 
 Well, I said a few days ago back at the bug report, which was reported at 
 11-Nov-2010 (about three weeks ago).

many months means more than one. I checked again and see that it seems
to start being used around 2009-08. ftp does not report a User Agent.
Based on the files downloaded, it seems it was incorporated to opensuse's
update mechanism (yum? forgive my ignorance), with others following, and
firefox probably did this too now. Well, I guess we'll somehow have to make
this work, one way or another. For the time being, please use http if you
have problems with your ftp client, and tell us if you have problems.
If you have or know about an important client that supports only ftp,
and which moved or is expected to move soon to EPSV, please tell us, so
we'll consider this more urgent.

Regards,
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Pop toaster recommendations seeked

[Yedidyah Bar-David]

On Sun, Oct 31, 2010 at 10:23:34PM +0200, Ira Abramov wrote:
 Howdie folks.
 
 for years I would install qmail+vpopmail+qmailadmin to give a client a
 nice robust virtual-domains web manager with mailing lists and all they
 needed, but today I discovered, that after 2-3 years of code freeze and
 maybe drop of users, courier imap seems has broken the option of
 authenticating with vpopmail, which means I have a wonderful virtual
 mail management and delivery system, but no pop3 and imapd to use it
 with. maybe it's time to ditch this kit and find another.
 
 option 1: courier MTA and SQWebmail - not so happy about it. I don't
 know the MTA, the message store is proprietary and limiting, no central
 management via web of the users (other than PHPMyAdmin, not friendly
 enough for my client.
 
 options 2: Zimbra. seams like a bit of an overkill but I'm told it
 works well, has postfix for an engine, and rumors say it supports
 virtual domains well (though I could not be sure from the confusing
 admin manual).
 
 Option 3: ? Donno... can you recommend?

I worked for several years with postfix+dovecot+postfixadmin. IIRC it
was mostly based on this howto:
http://bliki.rimuhosting.com/space/knowledgebase/linux/mail/postfixadmin+on+debian+sarge
which is pretty dated, but postfixadmin itself (and the underlying
tools, no doubt) is still maintained. It's pretty basic but working.

You might consider trying one of virtual hosting packages out there.
I then did and decided they were either too big or not mature enough
(or both) and went with postfixadmin which did a rather small part of
their common denominator but seemed mature. A partial list of the ones
I then looked at: ispconfig web-cp dtc ispman vhcs gnuhh ravencore.
Googling for most subsets of them will probably find others/reviews/etc.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: convert old DOS Hebrew encoding

[Yedidyah Bar-David]

On Thu, Oct 28, 2010 at 08:48:46AM -0400, Tzadik Vanderhoof wrote:
 I have a binary data file, in a format used by a relatively ancient program,
 which I am trying to convert into something sane. With the help of a Hex
 editor I have basically worked out the file format except that it contains
 Hebrew characters with an odd encoding.
 
 All characters are 8 bits. The standard 27 consonants (including final
 consonants) go from hex *80* to *9A*. Then there are *vowels* that seem to
 start around hex *9B* or so (I'm guessing right after the standard
 consonants end). Then there are *dotted consonants* that seem to start at
 hex *E0*.
 
 If I remember correctly, I think this is some sort of DOS encoding (perhaps
 connected to the old Hebrew chip). Does anyone have a table of this
 character mapping or a tool that will translate this mapping into a more
 normal Hebrew encoding like Unicode?

It's called CP862 and is supported by iconv. Try:
iconv -f cp862 -t utf-8  inp  out
Not sure about the vowels/dotted consonants/etc.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: cheap linux box ?

[Yedidyah Bar-David]

On Mon, Oct 11, 2010 at 10:30:39AM +0200, Erez D wrote:
 hey,
 
 
 my NSLU2 just died ( http://en.wikipedia.org/wiki/NSLU2 )
 I can replace it with a new one, however i was thinking - it has been long
 time since nslu2 showed up, and there must be a more powerful device
 available for a similiar price...
 
 I am actually looking for a cheap linux box, with usb2 and lan, which I can
 install a general purpose linux distribution on (e.g. debian and such. not
 openwrt which is good for routers, but not for other things like hosting a
 mail server etc ...)
 
 anyone knows of such a linux box ?

A few days ago someone wrote here about PandaBoard, and in a comment
in the mentioned article someone recommended a Seagate Dockstar:
http://www.semiaccurate.com/2010/10/04/after-beagleboard-comes-pandaboard/#comments
I did not verify the specs he mentioned, but if it's true, it's
much stronger than the average wireless router, and perhaps cheaper.
It's just not looking like a server - more like a part of a case of
one...
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How can I grab the filesystem from corrupted lvm (help recover data from lvm with one pv missing)

[Yedidyah Bar-David]

On Mon, Oct 11, 2010 at 10:36:33AM +0200, Boris shtrasman wrote:
 Hi,
 Yesterday my harddrive had an inconvenient event: when I was using a d-i
 from testing I accidentally overwritten on of the pv's connected to the lvm.
 I  started to panic  and found myself making more damage by running vgreduce
 --*remove*-*missing* since in one of the rescue files it noted this is the
 way to fix it.
[snip]

Did you try gpart?
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: How can I grab the filesystem from corrupted lvm (help recover data from lvm with one pv missing)

[Yedidyah Bar-David]

On Mon, Oct 11, 2010 at 11:41:38AM +0200, Boris shtrasman wrote:
 On Mon, Oct 11, 2010 at 11:33 AM, Yedidyah Bar-David 
 linux...@didi.bardavid.org wrote:
 
  On Mon, Oct 11, 2010 at 10:36:33AM +0200, Boris shtrasman wrote:
   Hi,
   Yesterday my harddrive had an inconvenient event: when I was using a d-i
   from testing I accidentally overwritten on of the pv's connected to the
  lvm.
   I  started to panic  and found myself making more damage by running
  vgreduce
   --*remove*-*missing* since in one of the rescue files it noted this is
  the
   way to fix it.
  [snip]
 
  Did you try gpart?
  --
  Didi
 
 
 I hope I did not miss something
 
 gparted see the parttion /dev/sda7 and recognize it as lvm but nothing from
 there.
 except that I got LVM isn't yet supported.
 
 it looks like the metedata is corrupted but the actual information is there
 (as when you loose the the parrtion table from the mbr (446-510))

Not gparted, gpart:
http://www.brzitwa.de/mb/gpart/index.html
Gpart is a tool which tries to guess the primary partition table of a
PC-type hard disk in case the primary partition table in sector 0 is
damaged, incorrect or deleted.

If you fail, and still want to resurrect specific files, you can also try
MagicRescue:
http://www.itu.dk/people/jobr/magicrescue/
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Open source English-Hebrew dictionary

[Yedidyah Bar-David]

On Wed, Oct 06, 2010 at 08:07:04PM +0200, Elazar Leibovich wrote:
 Is there an open source English Hebrew Dictionary? I'm interested with
 usable program that once you'll enter an English word, it would display the
 Hebrew translation. Specifically for the Android platform, but I believe
 that I'll be able to adapt any desktop application to a mobile device quite
 easily.
 
 I'm aware of the Limon project (http://www.jlc.org.il/limon/), but I'm not
 sure how to download the word list or the translation software. Except, the
 word list is not too long (6k).

If you'll search the list archives, you'll find some posts ten years ago
about wordtrans/qwordtrans/kwordtrans which can read babylon old-format
dictionaries and which I then managed to make read also the Hebrew
encoding. Not sure about the status of the project, but you might manage
to compile and use it.

Babylon's old dictionaries are not easy to find on the web, but you might
be able to find some.

 
 Is there a free dictionary in book form (ie, with no copyright holders)? I
 don't mind typing it.

As others said, I personally use only google translate these days. It's
good enough, of course not as fast as a local tool if you are reading in
some other software, but I did not bother looking for local tools. For
reading stuff on the net, I think it's very good. You can make it show
the original text and when you point at a sentence it pops up a
translation. The translation is usually very bad, which is why I never
use it to read the translation itself, but good enough if you only need
one word.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Open source English-Hebrew dictionary

[Yedidyah Bar-David]

On Thu, Oct 07, 2010 at 08:58:01AM +0200, Elazar Leibovich wrote:
 Thanks for the input.
 
 Two inputs:
 
 1) Babylon's dictionary files are not free or opensource (as far as I know).
 So distributing them might not be legal.

Indeed. But back then they were freely-downloadable from their site,
so if you manage to find one, I do not think it's a big difference
if you download it now or have downloaded it then from them. They moved
to a new format a long time ago and deleted all the old files.

Note IANAL.

 
 2) Google translate does not replace a dictionary, which defines the word
 exactly, when it's a noun, verb or otherwise, include usage examples, etc.

Indeed. For this I use google dictionary. And sometimes Reuven Alkalay's
paper dictionary... If it would have been sold on CDs, I am pretty certain
I'd buy one.

Wiktionary is also very useful. It has very little English-Hebrew material,
but more stuff in other languages, and I also think that if you decide to
donate some time (as you implied you might do), this is probably the place
to do this.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Two instances of Apache on the same server

[Yedidyah Bar-David]

On Tue, Oct 05, 2010 at 03:42:10PM +0200, Gabor Szabo wrote:
 I need to run two instances of Apache on the same SuSE based server but there
 is something very basic I don't manage. In order to run to servers
 I'll need two pid files
 to hold the two process ids. So the first thing I tried was to change
 the pidfile for the existing server.
 
 I shut down Apache and the changed  /etc/init.d/apache2
 the following line:
 
 : ${pidfile:=/var/run/httpd2.pid}
 
 to:
 
 : ${pidfile:=/var/run/httpd2b.pid}
[snip]

No recent experience with SuSE, but you probably have to also edit
apache's configuration file, perhaps somewhere in /etc/apache or
/etc/httpd or something like that. Search for 'PidFile'.

This is/was true for most distros. Debian (now, don't know since when)
keeps it in /etc/apache2/envvars, and both the init.d script and apache's
conf file are somehow set from it.

Personally, when I had to, I kept it only in httpd.conf, and wrote a
trivial init.d script that calls 'apachectl -f $CFG -k stop' to stop it,
instead of messing with pidfiles by itself. So far, it worked well.
Naturally, you might have problems e.g. if you change the conf while
apache is up, or stuff like that.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Two instances of Apache on the same server

[Yedidyah Bar-David]

On Tue, Oct 05, 2010 at 04:16:58PM +0200, Gabor Szabo wrote:
 On Tue, Oct 5, 2010 at 3:58 PM, Yedidyah Bar-David
 linux...@didi.bardavid.org wrote:
 
  No recent experience with SuSE, but you probably have to also edit
  apache's configuration file, perhaps somewhere in /etc/apache or
  /etc/httpd or something like that. Search for 'PidFile'.
 
 I grep-ped the whole /etc for -i pid  and the only hit was in
 /etc/init.d/apache2

Perhaps they keep the conf outside or rely on defaults or something like
that. You can try 'apachectl -V' for the defaults, or look at the init.d
script and see if it passes '-f someconffile' to apachectl and look at it
or at its includes, if any.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: asterisk client dummy question

[Yedidyah Bar-David]

On Mon, Sep 13, 2010 at 12:05:07PM +, Tzafrir Cohen wrote:
 On Mon, Sep 13, 2010 at 01:27:56PM +0200, David Ronkin wrote:
  i use ekiga 3.2.6 and i don't have such a tab.
 
 In the main window?
 
  what version should i download?
 
 Works fine here (3.2.7). IIRC it has always been in that tab.

Indeed, but depending on your options. In my case (2.0.12, Debian Lenny)
you should choose:
View - Control Panel - Dialpad
and
View - View Mode - Softphone

Some other variations work too - play with its buttons/menu.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

fic ce268 laptop

[Yedidyah Bar-David]

Hi all,

I am considering buying a FIC CE268 laptop, and would like to get reviews
about it from current users of it.

It's a low-profile netbook, costs 699 NIS in officedepot:
http://www.fic.com.tw/product/ce268.aspx
http://officedepot.co.il/catalog/catalogSku.do?id=4236592pr=QJ5N=2%2032937sgID=10

Does it work well? Is all the hardware working well? Sound? USB? Webcam?
Wireless? How long does the battery hold?

Does it seem to have reasonable quality? Can I expect it to work for more
than the 1 year warranty they provide?

What distribution of linux does it arrive with? Does it support Hebrew
well (i.e. as well as other current common distros)?

Is the webcam giving good pictures in linux? I recently used two different
USB webcams that worked well in Windows and in linux did work rather well
but gave a too dark image, with no apparent solution.

Does it work well in skype? That is, like a faster normal linux PC?

Is the keyboard comfortable enough for touch-typing? I do not expect to
type a lot on it, thogh.

Does it run hot after several hours of use?

Suggestions from current users of other similar small/cheap
laptops/netbooks that are easily bought in a store in Israel with local
support/warranty are also welcome.

TIA,
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Modern development environment on dated RHEL

[Yedidyah Bar-David]

On Tue, Apr 27, 2010 at 12:33:33PM +0300, Evgeny Budilovsky wrote:
 Why not compile all the latest software on some shared directory and then
 run it from there ...
 My company have policy not to give root access on development stations so I
 just compile all my development software in home directory and use it from
 there ...

If you go this route, you might consider using some package manager
for this, such as stow (or one of its clones) or gobolinux.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Redirecting mail

[Yedidyah Bar-David]

On Fri, Apr 09, 2010 at 04:15:58AM +0300, Geoff Shang wrote:
 Hello,

 I've got an interesting problem.

 I have been forced to stop hosting services for a few people.  One of 
 them transferred their DNS and hosting away from me, but stopped checking 
 for new mail on my host long before it stopped arriving.  So I now have 
 203 messages in Maildir format that I wish to resend to the intended  
 recipient, but at their new host.

 Anyone got any ideas how I can requeue this mail?  We're running postfix.

You can do this with mutt's bounce command - tag them all (e.g. 'T' '.')
then bounce all (';b').
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Switch between virtual consoles CtrlAltFx problem.

[Yedidyah Bar-David]

On Thu, Mar 25, 2010 at 10:57:17AM +0200, David Harel wrote:


 Etzion Bar-Noy wrote:

 Use Alt+F1-7 when in console mode. Don't use Ctrl+Alt+F1-7.

 Not working as well.
 Can a different key combination get assigned for console switching?

Can you try using chvt? If it works, it means that virtual console
switching works, and you only have a problem with key mappings.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: OT: recommendation for a small switch?

[Yedidyah Bar-David]

On Mon, Jun 08, 2009 at 02:45:36AM +0300, Hetz Ben Hamo wrote:
 Hi,
 
 I'm looking to buy a *cheap* 8 port gigabit switch to my house to
 connect all my computers.
 
 Can someone recommend a good switch (not hub) and a price or URL/phone
 where to buy?

Never used gigabit at home, or with small switches.

I had several edimax 100mbit 8port ones, and so far most if not all of
them died after a few years of use. I now have a Level1 switch, but only
for a few months, so I can't tell how well it survives over time.

At work we use Cisco switches, but that's not cheap at all.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Israeli spam! Who do I contact?

[Yedidyah Bar-David]

On Thu, May 07, 2009 at 12:29:02PM +0300, Dotan Cohen wrote:
 2009/5/7 Shachar Shemesh shac...@shemesh.biz:
  Dotan Cohen wrote:
 
  I just got a nice bit of spam, for a paid service, in Hebrew, to an
  obviously-harvested address! That sounds like a cool 1000 NIS to me.
  Does anyone know to whom to complain to collect?
 
  Naturally, half of it will be donated to hamakor.
 
 
 
  There is no-one to complain to.
 
  Contact the spammer and offer them to pay you directly. If they refuse, sue
  them in small claims. ISOC has court paper templates you can use.
  http://isoc.org.il/spam/
 
  There is, however, no one to do the work for you, I'm afraid.
 
  Shachar
 
 
 I was not expecting a one-stop solution. I see that small claims court
 is the answer that I was looking for.
 
 Is there a link to the actual spam law itself? Something that I could
 quote when contacting the spammer?

ISOC-IL's site, mentioned above, has a link to it, as well as other
useful material.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: the cost of switching to IPv6?

[Yedidyah Bar-David]

On Fri, May 01, 2009 at 08:07:57AM +0300, Geoffrey Mendelson wrote:
 For the first year or two, you probably won't get to any sites outside
 of your ISP until IIX is upgraded and the other ISPs upgrade their
 networks and interconnections.

IIX already supports v6:
http://isoc.org.il/iix/2x_list.html
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: ssh from 012 cable to server in US fail

[Yedidyah Bar-David]

On Tue, Apr 28, 2009 at 11:10:30AM +0300, Rami Addady wrote:
 Hi,


 I have weird problem , staring this morning I can't ssh to a server in  
 US,  from some computers that connect to the Internet using 012 cabels.

 But if I'm ssh to server in 012 farm and then from it to the US server  
 is work fine!

So you can look at the server logs and see what went wrong. You can also
try and run it with '-v' to add verbosity there too. If you do, first
try it on a local machine to make sure you do not kill it accidentally.

I have no idea re the actual problem.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: suid root - bash script

[Yedidyah Bar-David]

On Thu, Apr 23, 2009 at 08:56:45AM +0300, Erez D wrote:
 hi
 
 i have a bush script i want to be run with root permisions, no matter which
 user executes it.
 
 if it was a binary, i would only need set it suid root.
 
 but as it is a bash script, suid-ing it doesn't do anything, and suid-ing
 /bin/bash itself will make all scripts run suid root, which is surly not
 what i want.
 
 
 there must be a solution for that.

'sudo' is what you want.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: suid root - bash script

[Yedidyah Bar-David]

On Thu, Apr 23, 2009 at 11:31:38AM +0300, Shachar Shemesh wrote:

 Oron Peled wrote:

 There's a reason why the kernel does not respect suid/sgid bit on shell
 scripts -- It's because there are gazillions of ways a user can use
 this script to gain total root access.
   
 Name two?

The main famous one, inherent in the way scripts work, is that the
kernel has to look at the first line of the script, run the interpreter
mentioned there with the args provided, and this interpreter then runs,
looks at the script, and decides what to do. Running the interpreter
takes time, and so an attacker can make a symlink to it, run the
symlink, and replace it immediately, and have a chance to make the
interpreter run the attacker's version instead of the original. This is
different from running a binary directly, where the kernel knows where
it was and won't have to look again if you tried replacing a symlink to
it.


 Maybe writing a wrapper suid program that totally sanitize
 both the environment and command line arguments before
 exec'ing the script would make it. Although I wouldn't bet
 on it since it only covers the obvious attack vectors against
 shell scripts.
   
 Fine. Make the two cover these obvious vectors, one each.

 I have to say that I first heard about this restriction, I thought it  
 made a lot of sense. Since then, I have searched for these famed attack  
 vectors, and have come up short.

Well, I now googled for 'setuid scripts security' and found this FAQ:
http://www.faqs.org/faqs/unix-faq/faq/part4/section-7.html
It also mentions other, more-specific issues.

 Sure, if the script itself has security  
 holes, then a suid script will be vulnerable. As I'm sure you know well,  
 this is also true of C written code, however.

Indeed, but there are some differences - usually, finding bugs in
scripts is easier (especially if you do not have the sources for the
C-coded binary), and in the past there used to be bugs in various
interpreters of various OSes. The last point is hopefully less relevant
today, but so are setuid-scripts (I think no modern unix respects
these).


 So my question is: are there attack vectors against the following script?

 #!/bin/sh -e

 echo Hello, cruel world
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: suid root - bash script

[Yedidyah Bar-David]

Hi Noam,

On Thu, Apr 23, 2009 at 12:08:21PM +0300, Noam Rathaus wrote:
 Hi Yedidyah,
 
 This stupid - in my opinion - restriction also applies to perl script.

This is a free country, you know. You are entitled have your own
opinion. As I exaplained below, the main problem with setuid scripts is
irrespective of interpreter, be it /bin/sh, perl, or your own binary for
whatever language you invented.

 
 And there they also recommend using a C program that will be setuid
 that will run the  perl script.

Indeed. Or use sudo (which is the same, only general).

 
 This is of course an over-complicated manner of doing things,

I guess there are other ways to do this, but that's how it is in unix.
As far as I know, Windows does not have something similar at all - if
you want there to run some program as another user, you have to do much
more than setuid it.

 not to
 mention the fact that if this perl script or c program wrapper is then
 called from Apache the restriction still applies and I haven't been
 able to get around it.

I did not understand what exact restriction you talk about, what you
tried to achieve and what was the problem. If you want anyone to try and
help you, please provide some more details.

Best,
-- 
Didi

 On Thu, Apr 23, 2009 at 11:54 AM, Yedidyah Bar-David
 linux...@didi.bardavid.org wrote:
  On Thu, Apr 23, 2009 at 11:31:38AM +0300, Shachar Shemesh wrote:
 
  Oron Peled wrote:
 
  There's a reason why the kernel does not respect suid/sgid bit on shell
  scripts -- It's because there are gazillions of ways a user can use
  this script to gain total root access.
 
  Name two?
 
  The main famous one, inherent in the way scripts work, is that the
  kernel has to look at the first line of the script, run the interpreter
  mentioned there with the args provided, and this interpreter then runs,
  looks at the script, and decides what to do. Running the interpreter
  takes time, and so an attacker can make a symlink to it, run the
  symlink, and replace it immediately, and have a chance to make the
  interpreter run the attacker's version instead of the original. This is
  different from running a binary directly, where the kernel knows where
  it was and won't have to look again if you tried replacing a symlink to
  it.
 
 
  Maybe writing a wrapper suid program that totally sanitize
  both the environment and command line arguments before
  exec'ing the script would make it. Although I wouldn't bet
  on it since it only covers the obvious attack vectors against
  shell scripts.
 
  Fine. Make the two cover these obvious vectors, one each.
 
  I have to say that I first heard about this restriction, I thought it
  made a lot of sense. Since then, I have searched for these famed attack
  vectors, and have come up short.
 
  Well, I now googled for 'setuid scripts security' and found this FAQ:
  http://www.faqs.org/faqs/unix-faq/faq/part4/section-7.html
  It also mentions other, more-specific issues.
 
  Sure, if the script itself has security
  holes, then a suid script will be vulnerable. As I'm sure you know well,
  this is also true of C written code, however.
 
  Indeed, but there are some differences - usually, finding bugs in
  scripts is easier (especially if you do not have the sources for the
  C-coded binary), and in the past there used to be bugs in various
  interpreters of various OSes. The last point is hopefully less relevant
  today, but so are setuid-scripts (I think no modern unix respects
  these).
 
 
  So my question is: are there attack vectors against the following script?
 
  #!/bin/sh -e
 
  echo Hello, cruel world
  --
  Didi
 
 
  ___
  Linux-il mailing list
  Linux-il@cs.huji.ac.il
  http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
 
 

___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: suid root - bash script

[Yedidyah Bar-David]

On Thu, Apr 23, 2009 at 01:22:43PM +0300, Noam Rathaus wrote:
 Hi Yedidyah,
  I guess there are other ways to do this, but that's how it is in unix.
  As far as I know, Windows does not have something similar at all - if
  you want there to run some program as another user, you have to do much
  more than setuid it.
 
 
 Windows has the Run As service which does something similar, its a bit
 more.. complex, but it allows you to do what you mentioned. I am
 skipping on Windows' ability to run Service as other users :)

What's the difference (in use, not in implementation) between Run As and
'su'? Does it allow (by some configuration) running some program as
another user without providing its password? But this is getting
off-topic...

 My scenario is this:
 1) Apache runs a perl (which needs to be setuid = it changes IP addresses, 
 etc)
 2) This perl needs to call another perl responsible for updating the
 sytem = and managing that it works correctly
 3) This perl runs several other Perl scripts that are also setuid as
 they replace files
 
 When apache tries to execute the perl's line which says:
 system(/usr/local/bin/update.pl)
 
 I get the setuid warning that I need to put a wrapper
 
 I then did:
 system(/usr/local/bin/update)
 
 Where update
 
 Just executes update.pl
 
 Both update (written in C) and update.pl (written in perl) are setuid root.
 
 I still get the warning

But does it or does it not run as root? Perhaps it emits this warning
anyway when the script is setuid. I did not check. If it does not: What
happens when you simply run it yourself from the command line? Does it
work?

You can still try sudo anyway. Read 'man sudoers' and look at the
EXAMPLES section.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Kernel panic/exception

[Yedidyah Bar-David]

On Thu, Apr 23, 2009 at 01:59:19PM +0300, Noam Rathaus wrote:
 Hi,
 
 I am seeing these in the logs and I can't find a documentation to what
 might have been causing it:
 Apr 23 13:57:47 sp kernel:  1Unable to handle kernel paging request
 at virtual address 0804c3ac
 Apr 23 13:57:47 sp kernel:  printing eip:
 Apr 23 13:57:47 sp kernel: c0152dc0
 Apr 23 13:57:47 sp kernel: *pde = 089f6067
 Apr 23 13:57:47 sp kernel: *pte = 
 Apr 23 13:57:47 sp kernel: Oops: 
 Apr 23 13:57:47 sp kernel: CPU:0
 Apr 23 13:57:47 sp kernel: EIP:0010:[c0152dc0]Not tainted
 Apr 23 13:57:47 sp kernel: EFLAGS: 00010202
 Apr 23 13:57:47 sp kernel: eax: d201fadc   ebx: d201fac0   ecx:
 d7f86000   edx: 0804c3a4
 Apr 23 13:57:47 sp kernel: esi: d7f86000   edi:    ebp:
    esp: c83adf04
 Apr 23 13:57:47 sp kernel: ds: 0018   es: 0018   ss: 0018
 Apr 23 13:57:47 sp kernel: Process pidof (pid: 7080, stackpage=c83ad000)
 Apr 23 13:57:47 sp kernel: Stack: 522ba678   000c
   c02ba678 0073
 Apr 23 13:57:47 sp kernel:c4b75580 c4bb7c80 c02ba678 c02ba828
  01f0 c8443b00 
 Apr 23 13:57:47 sp kernel: 0c00 c0150e04 d7f86000
 dae26000   d7f86000
 Apr 23 13:57:47 sp kernel: Call Trace:[c0150e04] [c013389c] 
 [c0106cd3]
 Apr 23 13:57:47 sp kernel:
 Apr 23 13:57:47 sp kernel: Code: 8b 42 08 8b 4a 04 8b 52 0c 29 c8 01
 c5 85 d2 75 ef 8b 86 ec

To get any idea, you need to translate these addresses to function
names. You do this with ksymoops. Read the manpage or google - most
basic use (and which should work if you have the necessary files
installed) is to put the message in a file (e.g. file1) and run
ksymoops file1
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: suid root - bash script

[Yedidyah Bar-David]

On Thu, Apr 23, 2009 at 02:01:29PM +0300, Noam Rathaus wrote:
 Hi Yedidyah,
 
 1) It doesn't run = shows error = stops
 2) Under root it works = no error = works
 3) Should I test it under another user? :)

Yes, that's what I meant. Sorry.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: DavkaWriter or Dagesh With Wine or Crossover

[Yedidyah Bar-David]

Hello Moshe,

On Fri, Apr 17, 2009 at 11:38:00AM +, Moshe Brace using Yahoo wrote:
 
 Anybody got this to work with Crossover or Wine? I am looking for an 
 alternative if one exits other than Open Office for Hebrew Texting. Any 
 suggestions?
 
 Recently I tried Office 2003 and found that in Tables Hebrew came up as Left 
 to Right I presume this is the fault of Crossover. Does Abbyword behave any 
 better? 

You might want to try lyx. It's very different from what you are used to
in any other word processor, and does take time to get used to, but for
some kinds of uses, such as writing long, structured documents (e.g. a
book), many people find it much better than normal word processors.

Regards,
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il

Re: Kernel cares where it should reside?

[Yedidyah Bar-David]

On Thu, Apr 02, 2009 at 12:32:53PM +0300, David Harel wrote:
 Hi there,

 Using Gentoo (and have to configure kernel).
 When I do the command - file - on kernel file I just built it says,  
 among other things, root_dev 0x801.
 The working kernel says - root_dev 0x10.
 Where is this parameter determined

Not sure exactly, somewhere in the build process. You can change it with
'rdev', but you do not need to:

 and since when does the kernel care  
 where it resides?

It's not 'since', but 'until when'. It was so since the very beginning.
The kernel had a small boot loader in its start. You could 'cat zImage 
/dev/fd0' and get a bootable floppy. No standard boot loader (lilo,
grub) care{s,ed} about this. IIRC today it does not work anymore,
perhaps never worked with bzImage.
The reason it cared is obvious - you had to tell it which FS to mount as
root. And had no config file to write 'root=something'.
-- 
Didi


___
Linux-il mailing list
Linux-il@cs.huji.ac.il
http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il