Re: Why does Linux crash?

[Alex Mestiashvili]

On 04/20/2011 07:38 AM, Borden Rhodes wrote:

Thank you for your reply and your consolation that other people are
equally miffed with Eclipse.  My question, though, is about Linux
control systems.  Is one of the kernel's design goals to manage system
resources to prevent a buggy program from crippling the system and
forcing a hard restart?  If so, the control failed and it needs to be
reported... right?


   

OOM-killer , Grsecurity patches , ulimit , virtualization 

Run eclipse in a VM .

just a simple forkbomb will make your OS unusable ...

IMHO If you expect that your application can go crazy you should think 
in advance how to prevent it from eating up resources .


Regards ,
Alex


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4dae819c.8020...@biotec.tu-dresden.de

Re: best practice to use newer cpan modules on squeeze

[adris]

Hi,

Am Dienstag, den 19.04.2011, 23:55 -0400 schrieb Jim Green:
> what should be the best practice here right now? I use cpan command to
> install some modules that are not available in debian. but how about
> those not up-to-date ones?

I'm using CPAN at work and at home whenever a module is required with a
higher version number and if I do not have the time to package it.

They all get installed in /usr/local/lib/perl/5.x.x/.  Thus if you still
have a deb version installed, then Perl will use the newer version from
CPAN.  The good thing is, that it's the way you'd probably expect it to
be, but the bad thing is, that some Debian Perl programs which expect
version x.y.z might break.  I never had such a case yet though.

For example the JSON module once changed some function names.

The best way to go is to create a deb package for the module though.  It
takes a little more work, but a lot of people will benefit from it.

For development and testing I tend to have a local CPAN installation
directory ~/perl5.  CPAN can be tweaked to install all it's modules in
that users home directory, if run with that user.

Then I do not need to have root privileges to install and remove CPAN
modules and these modules are installed only for that user.

Have a look at local::lib
.
It's alos available as a deb (liblocal-lib-perl).


Regards,
Adris


signature.asc
Description: This is a digitally signed message part

Re: best practice to use newer cpan modules on squeeze

[Alex Mestiashvili]

On 04/20/2011 05:55 AM, Jim Green wrote:

Hello!
in squeeze most perl modules are not up-to-date. for example libmoose-perl.

In this case I include the sid repo and attempt to upgrade to sid
version of libmoose but a hell of dependency begins. I am afraid if
insist on upgrading this module, lots of perl core module/perl would be
upgraded in sid.

what should be the best practice here right now? I use cpan command to
install some modules that are not available in debian. but how about
those not up-to-date ones?

Thanks!
Jim


   

I would rebuild source packages from sid in squeeze .

Regards ,
Alex


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4dae8bad.3050...@biotec.tu-dresden.de

Re: [OT] Purchasing a wired switch; advice needed

[Jochen Schulz]

Klistvud:
> And here's what I need advice for:
> 
> I'm planning to purchase a wired (consumer grade) switch since I've
> heard they're inherently more robust than (consumer grade) routers,
> and I'm planning to connect it *directly* to our cable broadband
> modem. Then, the two families would connect their respective routers
> (we have some spare wireless routers) to this switch. The various
> computers and network printers would then be connected, in turn, to
> these routers.

You are looking for a router (OSI layer 3), not a switch (OSI layer 2).
It needs to have three distinct interfaces (1xWAN, 2xLAN). If it runs
something like OpenWrt you are completely free about its configuration,
e.g. you can put each interface in separate networks, enable or disable
routing between them, filter traffic to your liking etc. It should even
be possible to use some QoS features in order to share the bandwith
between the families (subnets).

I think even a simple Linksys WRT54GL would suffice. AFAIK you can
disable the bridge between the switchports and thus configure each
switchport as a separate interface. If you want GBit-Ethernet you can
build something yourself using a Routerstation Pro, for example. Take a
look at OpenWrt's Table of Hardware.

Alix and Soekris boards come to my mind as well, but IMHO they are too
expensive.

J.
-- 
If I could travel in time I would show my minidisc to the Romans and
become Caesar until the batteries ran out.
[Agree]   [Disagree]
 


signature.asc
Description: Digital signature

Re: file systems

[Stan Hoeppner]

Heddle Weaver put forth on 4/19/2011 6:58 PM:

> XFS is excellent for large file sizes - graphics, music, videos, etc, but
> ext3/4 are better for a range of file sizes and therefore better for a

This is simply not true.  Modern XFS is just as performant with small
files as EXT3/4, especially with multiuser or highly parallel workloads.
 EXTx traditionally has had two advantages over XFS:

1.  Workloads with zero or low parallelism
2.  Metadata write heavy operations

The first typically holds true for many single threaded workloads.
That's fine.  XFS was designed for single threaded workloads, but high
bandwidth multithreaded workloads.

The second evaporated when Dave Chinner introduced delayed logging last
year.  Today XFS metadata operations are on par with all Linux
filesystems, and surpass all others with many workloads.

Apparently you've not used XFS for maildir storage.  It's throughput is
quite a bit better than EXT3/4.  Based on the file types you mention
above, it would appear you are strictly a desktop Linux user.  This
would explain your lack of knowledge of XFS and your penchant for
repeating misinformation.  It would also explain and your preference for
EXTx.

> smaller operation, which is what the O.P. seems to be describing. 

XFS is just as applicable to a small operation as a large one.  For
instance, it is the premier filesystem used in building MythTV servers.
 A DVR is a pretty small operation.  XFS is the only Linux filesystem
with a defrag utility, and an online one at that.  This is beneficial to
all operations, regardless of size.  XFS has a far richer set of
management tools than any other Linux filesystem.

You simply can't go wrong with XFS on any size server, assuming you
first read the basic documentation and the XFS FAQ.

> In case of
> mishap, they fall back to ext2.

I'm not sure exactly what you mean by this.  I doubt you are either.
What kind of "mishap" would require converting the EXT3/4 filesystem
back to EXT2?

> Performance is trivial, as any file system can be tuned.

This statement clearly demonstrates your lack of filesystem architecture
knowledge.  Just as you can't tune a Ford Pinto to outrun a Ferrari, you
can't tune EXTx to outperform XFS in highly parallel workloads.  You may
want to do some reading and review some performance reports.

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.114.1918&rep=rep1&type=pdf

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_creates_num_threads=1.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_creates_num_threads=16.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_creates_num_threads=128.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_reads._num_threads=1.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_reads._num_threads=16.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_reads._num_threads=128.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_writes._num_threads=1.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_writes._num_threads=16.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_writes._num_threads=128.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_writes_odirect._num_threads=1.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_writes_odirect._num_threads=16.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_random_writes_odirect._num_threads=128.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_sequential_reads._num_threads=1.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_sequential_reads._num_threads=16.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Large_file_sequential_reads._num_threads=128.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Mail_server_simulation._num_threads=1.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Mail_server_simulation._num_threads=16.html

http://btrfs.boxacle.net/repository/raid/2.6.35-rc5/2.6.35-rc5/2.6.35-rc5_Mail_server_simulation._num_threads=128.html

-- 
Stan


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4daeadf2.30...@hardwarefreak.com

Re: Why does Linux crash?

[Axel Freyn]

Hi,
On Tue, Apr 19, 2011 at 10:57:19PM -0400, Borden Rhodes wrote:
> Good evening,
> 
> I do mean this earnestly and, despite my frustration, I am not trying
> to flame the good people of Debian, GNU or Linux.  Nevertheless, I
> have to ask: why is it that in 2011, the world's greatest operating
> system lets Eclipse seize control of my computer, eat up 2 GB of RAM,
> monopolise a 2.2 GHz, dual-core processor and flood my hard drive with
> I/O?  I thought that a computer capable of processing over 4 billion
> operations a second could sort itself out in 20 minutes but, alas, I
> had to yank the power.
> 
> I thought the Linux kernel was supposed to have controls in place to
> prevent programs from getting away with this.  Of course, the problems
> inherent in Java, and by extension, Eclipse, are a whole other topic.
> However, is there a kernel task force working to prevent this from
> happening and, if so, what's the best way of giving them feedback when
> my system locks up so they can plug up the hole?  I hate to think what
> a malicious program could do to a web server if Eclipse can do this to
> my computer.
The central point is: what rights/limits do you give to your programs?
on Web servers, such limits are normally activated -- while they are
deactivated on private systems.
Just a few examples:
 - you can add disk quota for each user/directory, limiting its maximal
   size and thus preventing the hard disk of overflowing
 - you can use "ulimit" to pass runtime-limits to eclipse (how much ram
   it may use, how much calculation time, )
 - you can limit the priority of eclipse compare to other jobs (that is:
   if the processor is occupied, will he give calculation time to
   eclipse or to something else?)
 - you can adjust which processes are killed in which order if you run
   out of memory (by adjusting /proc/PID/oom_adj)
 - ...

But the principal problem is: each of those limits/protections reduces
the usability (e.g. if you have 2GB Ram, and you limit eclipse to 2GB,
it will be killed by the Kernel as soon as it tries to use 2GB and 1
byte from the SWAP -- which would not cause a problem at all). and only
YOU can decide what you want ;-) May be there are people how give a
difficult task to eclipse and accept then to wait 2h until it
finisheѕ...

That's the reason (I believe) that those features are not activated by
default on "private" computers. And e.g. for Debian: It is almost
impossible to define "sensible" rules which would apply for ALL
computers and ALL users where Debian will be installed -- so I think the
solution "don't apply restrictions by default" is the best approach...

Axel


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110420100409.GQ7351@axel

Re: [OT] Purchasing a wired switch; advice needed

[Klistvud]

Dne, 20. 04. 2011 11:25:27 je Jochen Schulz napisal(a):

Klistvud:
> And here's what I need advice for:
>
> I'm planning to purchase a wired (consumer grade) switch since I've
> heard they're inherently more robust than (consumer grade) routers,
> and I'm planning to connect it *directly* to our cable broadband
> modem. Then, the two families would connect their respective routers
> (we have some spare wireless routers) to this switch. The various
> computers and network printers would then be connected, in turn, to
> these routers.

You are looking for a router (OSI layer 3), not a switch (OSI layer  
2).

It needs to have three distinct interfaces (1xWAN, 2xLAN). If it runs
something like OpenWrt you are completely free about its  
configuration,
e.g. you can put each interface in separate networks, enable or  
disable
routing between them, filter traffic to your liking etc. It should  
even

be possible to use some QoS features in order to share the bandwith
between the families (subnets).


Using a router is precisely what I'm trying to avoid here. There are  
multiple reasons for that, let me state just a couple:


1. Having had a router for the last 5 years or so, I've come to the  
conclusion that a single router, with a single configuration interface,  
can not accomodate our differing  needs (the other family uses strictly  
Windows and needs UPnP, we strictly use Debian and hate UPnP; resetting  
the shared router by one person has broken a download or some other  
internet-related task for the other person many a time; we both need so  
many ports forwarded that the 20 available ports of an average router  
simply aren't enough; and many other "social" issues). That's why I am  
looking for a "zero-configuration" solution in which there is no  
configuration interface and consequently no potential conflicts can  
arise.


2. Consumer grade routers are flaky at best; cramming a firewall, port  
forwarding, NATting, dhcp, routing and what not into a 32 MB device  
apparently wasn't such a great idea, technologically, and was, yet  
again, dictated by marketing concerns only. I need something robust. I  
need something you can fire-and-forget. That's why I'm trying to avoid  
(consumer grade) routers like the plague. I have four already, and they  
are nothing to get excited about. I'd rather have a good switch or even  
hub than a consumer-grade router doing my broadband sharing.


3. Segment independence. I want to be able to use my network segment no  
matter what is happening on the other segment. With one central router  
serving both segments -- no matter how separated they may be internally  
to the device -- this can't be achieved. It's still just one flaky  
device serving both, and when it goes down, it takes everything with  
it. With a sturdy entry point, I'm hoping this could be completely  
different?


4. Connecting the routers *downstream* of the switch (that's the whole  
idea) would hopefully halve the traffic passing through them, and so  
should ease their work (making them lock up less frequently).


What I had in mind is something like this:  
http://www.ehow.com/how_6823201_use-switch-hub-instead-router.html .  
Unfortunately, there are many posts on the Internet affirming that such  
a configuration can't and won't work, because a switch can't give out  
two IP's if your ISP just gives you one. So, in doubt, I turned to this  
list for advice. I don't want to buy a switch (or the wrong switch)  
only to discover that it won't work in that configuration.





I think even a simple Linksys WRT54GL would suffice. AFAIK you can
disable the bridge between the switchports and thus configure each
switchport as a separate interface. If you want GBit-Ethernet you can
build something yourself using a Routerstation Pro, for example. Take  
a

look at OpenWrt's Table of Hardware.

Alix and Soekris boards come to my mind as well, but IMHO they are too
expensive.

J.
--
If I could travel in time I would show my minidisc to the Romans and
become Caesar until the batteries ran out.
[Agree]   [Disagree]
  







--
Cheerio,

Klistvud  
http://bufferoverflow.tiddlyspot.com
Certifiable Loonix User #481801  Please reply to the list, not to  
me.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1303297702.18252.0@compax

Re: best practice to use newer cpan modules on squeeze

[Leonardo Ruoso]

2011/4/20 Jim Green 

> Hello!
> what should be the best practice here right now? I use cpan command to
> install some modules that are not available in debian. but how about
> those not up-to-date ones?
>

What about using locallib+cpan or a chrooted+(cpan|sid) enviromment for your
perl applications?
-- 
Leonardo Ruoso - Jornalista/Desenvolvedor
Assessoria de Imprensa. Consultoria de Marketing. Desenvolvimento e
Integração de Software.
Comunicação Social/Jornalismo - UFC/2006. Telecomunicações - ETFCE/1998.
Foos, Perl, Debian Gnu/Linux, Agile, UML, DBA e OOP. Coaching/NLP. Inglês e
Francês.
http://leonardo.ruoso.com - http://www.linkedin.com/in/lruoso

Re: [OT] Purchasing a wired switch; advice needed

[Russell L. Harris]

* Jochen Schulz  [110420 05:48]:
> Klistvud:
> > And here's what I need advice for:
> You are looking for a router (OSI layer 3), not a switch (OSI layer 2).
> It needs to have three distinct interfaces (1xWAN, 2xLAN). If it runs
...
> I think even a simple Linksys WRT54GL would suffice. AFAIK you can

If you have spare computer (old and slow is acceptable) and three
ethernet cards, you could download and install SmoothWall
(www.smoothwall.org), and be running within a half-hour, and at no
expense.

The point of using SmoothWall is that it is extremely simple to install
and configure -- even a Window$ user can do it. 

Then, once experimentation shows you what works, you can look for a
firmware router to reduce space and electricity requirements.

RLH


 


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110420110436.GA3344@cromwell.tmiaf

Re: [OT] Purchasing a wired switch; advice needed

[Jochen Schulz]

Klistvud:
> Dne, 20. 04. 2011 11:25:27 je Jochen Schulz napisal(a):
>> Klistvud:
>>> 
>>> I'm planning to purchase a wired (consumer grade) switch since I've
>>> heard they're inherently more robust than (consumer grade) routers,
>>> and I'm planning to connect it *directly* to our cable broadband
>>> modem. Then, the two families would connect their respective routers
>>> (we have some spare wireless routers) to this switch. The various
>>> computers and network printers would then be connected, in turn, to
>>> these routers.
>> 
>> You are looking for a router (OSI layer 3), not a switch (OSI layer 2).
>> …
>
> Using a router is precisely what I'm trying to avoid here.

I don't really know how cable broadband works, but AFAIU you need
exactly one system routing your traffic to the WAN, just like with DSL.
If this is the case, you need a router, period. That device doesn't need
to be the usual plastic junk, but it needs to, erm, route IP traffic.
That's why it's called a router. ;-)

> 1. Having had a router for the last 5 years or so, I've come to the
> conclusion that a single router, with a single configuration
> interface, can not accomodate our differing  needs (the other family
> uses strictly Windows and needs UPnP, we strictly use Debian and
> hate UPnP;

I must admit I never really understood what UPnP actually does, but if
you mean the automatic port-forwarding nonsense: it might be possible to
enable it for one network only with OpenWrt.

> resetting the shared router by one person has broken a
> download or some other internet-related task for the other person
> many a time;

The real problem here is that you need to reset the router in the first
place. I don't remember having done this with my WRT54GL ever. But my
network is not as busy as yours, of course.

> we both need so many ports forwarded that the 20
> available ports of an average router simply aren't enough;

No problem with OpenWrt.

> 2. Consumer grade routers are flaky at best; cramming a firewall,
> port forwarding, NATting, dhcp, routing and what not into a 32 MB
> device apparently wasn't such a great idea,

I wouldn't blame the low quality of the software on the low hardware
specs.

> I'd rather have a good switch or even hub than a consumer-grade router
> doing my broadband sharing.

It's just that a switch isn't capable of doing what you need.

> 3. Segment independence. I want to be able to use my network segment
> no matter what is happening on the other segment. With one central
> router serving both segments -- no matter how separated they may be
> internally to the device -- this can't be achieved.

With your proposed network layout, internal traffic doesn't reach the
shared router at all. Only cross-network (if allowed at all) and WAN
traffic goes through the shared router.

> What I had in mind is something like this:
> http://www.ehow.com/how_6823201_use-switch-hub-instead-router.html .

This only works if your cable modem is actually a router. As I said, I
don't know the specifics of cable broadband.

If your cable modem is able to connect to the WAN by itself, then you
can just go ahead any buy practically any switch on the market. The
cheaper it is, the less you have / are able to configure.

J.
-- 
I am on the payroll of a company to whom I owe my undying gratitude.
[Agree]   [Disagree]
 


signature.asc
Description: Digital signature

Re: Re: Why does Linux crash?

[Dr. Ed Morbius]

on 01:38 Wed 20 Apr, Borden Rhodes (j...@bordenrhodes.com) wrote:

> Thank you for your reply and your consolation that other people are
> equally miffed with Eclipse.  My question, though, is about Linux
> control systems.  Is one of the kernel's design goals to manage system
> resources to prevent a buggy program from crippling the system and
> forcing a hard restart?  

Generally:  via a memory segfault (a process requesting memory it
doesn't have access to), yes.

However, system resources can still be overcommitted (memory is
allocated in excess of system resources, look up OOM killer), there are
other resources (open file handles, open sockets, disk I/O, paging
rates), which can be exhausted or when occuring at too high a rate will
make a system unresponsive.  There's no surefire way of preventing this
(Google "halting problem"), though there are measures which can be taken
to reduce these risks.

> If so, the control failed and it needs to be reported... right?

Not necessarily.

If you can produce and provide specific debugging information, file a
bug via the Debian bugtracking system.

Given what you've posted so far, I'd suggest you not do so, as you don't
understand the problem and would be producing noise in the system.

-- 
Dr. Ed Morbius, Chief Scientist /|
  Robot Wrangler / Staff Psychologist| When you seek unlimited power
Krell Power Systems Unlimited|  Go to Krell!


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/20110420120355.gb22...@altaira.krellpowersys.exo

Re: [OT] Purchasing a wired switch; advice needed

[Tom Furie]

On Wed, Apr 20, 2011 at 01:08:22PM +0200, Klistvud wrote:

> Unfortunately, there are many posts on the Internet affirming that
> such a configuration can't and won't work, because a switch can't
> give out two IP's if your ISP just gives you one. So, in doubt, I

This is the crucial point. If your modem will only provide 1 IP, then
you need to use a router. If your modem will provide multiple IPs, then
you might be able to connect you LAN segments through a switch to the
modem, but this will cause further complexity and configuration issues
to your setup.

Cheers,
Tom



signature.asc
Description: Digital signature

Re: Why does Linux crash?

[Miles Fidelman]

Apropos this discussion:

The original poster does have a point:  resource allocation and process 
isolation are core o/s functions.


I've never really given a lot of thought to the details - for most of 
the servers I've set up over the years, everything has pretty much just 
worked (under Solaris for a while, then Redhat, these days Debian).  To 
the extent that I've tuned resource consumption, I've mostly done it 
across Xen virtual machines.


But now that someone's raised the question:  Can somebody suggest a good 
reference for an overview of the resource management philosophy, 
architecture, tools, and defaults for both the Linux kernel and for Debian?


Miles Fidelman

--
In theory, there is no difference between theory and practice.
In  practice, there is.    Yogi Berra



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4daed5c5.5020...@meetinghouse.net

Re: [OT] Purchasing a wired switch; advice needed

[Andrew McGlashan]

Hi,

Klistvud wrote:
I'm planning to purchase a wired (consumer grade) switch since I've 
heard they're inherently more robust than (consumer grade) routers, and 
I'm planning to connect it *directly* to our cable broadband modem. 
Then, the two families would connect their respective routers (we have 
some spare wireless routers) to this switch. The various computers and 
network printers would then be connected, in turn, to these routers.


Can a switch juggle two basically separate segments, plus a broadband 
connection, like that? What capabilities should I be looking for in such 
a switch?


Would it reduce the load on the two routers and do away with their 
lock-ups?


Would it make our two networks more independent, so that one locked-up 
router wouldn't bring the whole network down? I guess we should separate 
the shared LAN into two distinct IP subnets?


Firstly, if you have loads of connections via ANY device to the 
Internet, such as lots of torrents and you do that through NAT (which is 
how it is mostly done), then you'll have large NAT tables.  Routers will 
have to keep track of all the traffic that is current and it will time 
out traffic that is old (in it's tables).


It doesn't matter if it is a switch or a router, at the end of the day, 
you'll end up with the Internet router doing most of the real work.  The 
only way around this, splitting up the connection to two nets, is to 
have multiple IP addresses and have them assigned as one-to-one and no 
NAT in play.  Then each downstream router can manage it's own network 
based on the one [public] IP that is assigned to it.  The Internet 
facing device shouldn't do anything special except pass all traffic to 
the relevant router handling the public IP.


The other thing to consider is using VLANs so that both networks are as 
separated as possible; that will lessen the risk of any person's 
computer from either network being about to attack / infect any computer 
on the other family's network.


In a nutshell, I don't think your idea to use a switch has any worth in 
this case.  And if you can't get your ISP to provide an extra IP (or 
second distinct cable login to get it's own IP), then you'll have these 
huge NAT table issues with low memory consumer routers possibly 
requiring restarts to clear the tables and start again.


--
Kind Regards
AndrewM

Andrew McGlashan
Broadband Solutions now including VoIP


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4daef03d.90...@affinityvision.com.au

pbuilder and CONCURRENCY_LEVEL

[Niccolò Belli]

How can I set CONCURRENCY_LEVEL when using pbuilder?
CONCURRENCY_LEVEL=9 pbuilder build file.dsc doesn't work.

Thank you,
Darkbasic


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4daefc2a.3050...@gmail.com

Re: Re: Why does Linux crash?

[Borden Rhodes]

Thank you for your reply, Axel; perfect answer.  Now that I know that
these features of Linux exist I can go hunt them down and figure out
how to use them and stop this from happening again (like it did after
I sent my original e-mail).

With thanks again,


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/banlktim+xgp4ltk1hdyp2eacetzxs+i...@mail.gmail.com

Re: [OT] Purchasing a wired switch; advice needed [question]

[Paul E Condon]

On 20110421_003957, Andrew McGlashan wrote:
> Hi,
> 
> Klistvud wrote:
> >I'm planning to purchase a wired (consumer grade) switch since I've
> >heard they're inherently more robust than (consumer grade) routers,
> >and I'm planning to connect it *directly* to our cable broadband
> >modem. Then, the two families would connect their respective
> >routers (we have some spare wireless routers) to this switch. The
> >various computers and network printers would then be connected, in
> >turn, to these routers.
> >
> >Can a switch juggle two basically separate segments, plus a
> >broadband connection, like that? What capabilities should I be
> >looking for in such a switch?
> >
> >Would it reduce the load on the two routers and do away with their
> >lock-ups?
> >
> >Would it make our two networks more independent, so that one
> >locked-up router wouldn't bring the whole network down? I guess we
> >should separate the shared LAN into two distinct IP subnets?
> 
> Firstly, if you have loads of connections via ANY device to the
> Internet, such as lots of torrents and you do that through NAT (which
> is how it is mostly done), then you'll have large NAT tables.
> Routers will have to keep track of all the traffic that is current
> and it will time out traffic that is old (in it's tables).
> 
> It doesn't matter if it is a switch or a router, at the end of the
> day, you'll end up with the Internet router doing most of the real
> work.  The only way around this, splitting up the connection to two
> nets, is to have multiple IP addresses and have them assigned as
> one-to-one and no NAT in play.  Then each downstream router can
> manage it's own network based on the one [public] IP that is assigned
> to it.  The Internet facing device shouldn't do anything special
> except pass all traffic to the relevant router handling the public
> IP.
> 
> The other thing to consider is using VLANs so that both networks are
> as separated as possible; that will lessen the risk of any person's
> computer from either network being about to attack / infect any
> computer on the other family's network.
> 
> In a nutshell, I don't think your idea to use a switch has any worth
> in this case.  And if you can't get your ISP to provide an extra IP
> (or second distinct cable login to get it's own IP), then you'll have
> these huge NAT table issues with low memory consumer routers possibly
> requiring restarts to clear the tables and start again.

Andrew,

I'm lurking here, looking to better understand a problem that I've never
had to confront: NAT, I understand requires translation tables, one entry
for each active tcp connection. This takes RAM. It also takes enough 
CPU cycles to maintain this table --- set up new connections and find
and delete old connections that are no longer in use. I can see a number
of reasons why the old router would have to quit and need resetting.
I wonder if one of the uplist suggestions which was to set up an old
computer as a router might not work unless the hardware is powerful
enough to keep up with the heavy maintenance activity on the NAT tables.
Extra RMA also implies extra cpu cycles to find and remove old table
entries. It might be that for this service, a somewhat heavy duty
computer is needed. 

Coming from me, the above is little more than idle speculation. I'd be
interested in reading your comments on this speculation. Does it make
any sense?  How might OP go about specifying hardware that doesn't
need resetting in his environment? Would it be likely that some
'industrial strength' routers meet his needs, and others are really
expensive over kill?  Would the appropriate industrial grade router be
a bigger/smaller energy hog than something cobbled together out of an
old computer and junk box network cards? Or compare to new consumer
suitably sized for the job?

Whatever you have time to write will be interesting to me and
might be really useful to OP. 

-- 
Paul E Condon   
pecon...@mesanetworks.net


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110420172613.ga21...@big.lan.gnu

Re: Re: Troubleshooting boot

[Michael Grünewald]

Hi Jochen,
thank you for your input

Jochen Schulz  wrote:
> Not a kernel option, but you can edit /etc/default/rcS and set
> CONCURRENCY=none. That should prevent init scripts from being run in
> parallel.

it will unluckily be of little help there, since the kernel is not able 
to run init.

--
Best Regards,
Michael


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4daf1278.9030...@laposte.net

Re: [OT] Purchasing a wired switch; advice needed [question]

[PaulNM]

Paul E Condon wrote:

> 
> Andrew,
> 
> I'm lurking here, looking to better understand a problem that I've never
> had to confront: NAT, I understand requires translation tables, one entry
> for each active tcp connection. This takes RAM. It also takes enough 
> CPU cycles to maintain this table --- set up new connections and find
> and delete old connections that are no longer in use. I can see a number
> of reasons why the old router would have to quit and need resetting.
> I wonder if one of the uplist suggestions which was to set up an old
> computer as a router might not work unless the hardware is powerful
> enough to keep up with the heavy maintenance activity on the NAT tables.
> Extra RMA also implies extra cpu cycles to find and remove old table
> entries. It might be that for this service, a somewhat heavy duty
> computer is needed. 
> 

The problem home router have with the nat tables isn't really due to cpu
utilization, but RAM usage.  Bittorrent is particularly an issue due to
the many, many, connections and packets it creates in a short period of
time.  Popular distro releases can quickly create a huge amount of
traffic since there are many people worldwide seeding them. This is
especially true immediately after a release.

An old PII or PIII era machine with 128 MB or more should be plenty to
handle the load.  Either use smoothwall, or something else like the x86
version of openwrt.

PaulNM


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4daf1c79.6010...@paulscrap.com

Re: [OT] Purchasing a wired switch; advice needed

[Ron Johnson]

On 04/20/2011 06:08 AM, Klistvud wrote:
[snip]


What I had in mind is something like this:
http://www.ehow.com/how_6823201_use-switch-hub-instead-router.html .
Unfortunately, there are many posts on the Internet affirming that such
a configuration can't and won't work, because a switch can't give out
two IP's if your ISP just gives you one. So, in doubt, I turned to this
list for advice. I don't want to buy a switch (or the wrong switch) only
to discover that it won't work in that configuration.



As others have mentioned, this is the sticky wicket, and why you *do* 
need a router.  (Unless your broadband modem is also a router.)


An old PC running Smoothwall will give your a "router" with enough 
horsepower to handle the situation you describe.


--
"Neither the wisest constitution nor the wisest laws will secure
the liberty and happiness of a people whose manners are universally
corrupt."
Samuel Adams, essay in The Public Advertiser, 1749


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4daf1d6c.90...@cox.net

Re: Why does Linux crash?

[Ron Johnson]

On 04/19/2011 09:57 PM, Borden Rhodes wrote:

Good evening,

I do mean this earnestly and, despite my frustration, I am not trying
to flame the good people of Debian, GNU or Linux.  Nevertheless, I
have to ask: why is it that in 2011, the world's greatest operating
system lets Eclipse seize control of my computer, eat up 2 GB of RAM,
monopolise a 2.2 GHz, dual-core processor and flood my hard drive with
I/O?  I thought that a computer capable of processing over 4 billion
operations a second could sort itself out in 20 minutes but, alas, I
had to yank the power.



I'd run top(1) in a terminal window, with the window set to "always on 
top" and watch what happens when you do whatever you do which causes 
Eclipse to consume all resources.


--
"Neither the wisest constitution nor the wisest laws will secure
the liberty and happiness of a people whose manners are universally
corrupt."
Samuel Adams, essay in The Public Advertiser, 1749


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4daf2056.2050...@cox.net

Re: [OT] Purchasing a wired switch; advice needed

[Dom]

On 20/04/11 18:52, Ron Johnson wrote:

On 04/20/2011 06:08 AM, Klistvud wrote:
[snip]


What I had in mind is something like this:
http://www.ehow.com/how_6823201_use-switch-hub-instead-router.html .
Unfortunately, there are many posts on the Internet affirming that such
a configuration can't and won't work, because a switch can't give out
two IP's if your ISP just gives you one. So, in doubt, I turned to this
list for advice. I don't want to buy a switch (or the wrong switch) only
to discover that it won't work in that configuration.



As others have mentioned, this is the sticky wicket, and why you *do*
need a router. (Unless your broadband modem is also a router.)

An old PC running Smoothwall will give your a "router" with enough
horsepower to handle the situation you describe.


I run Smoothwall on an old Athlon (K6) system that I was given. I 
upgraded the memory to 128MB, which is fine for what I use it for. I 
have quite a few boxes (Linux and Windows) using it to access the 
internet with no problems at all.


The only issue I have is that the latest upgrade to Smoothwall Express 
(the free version) no longer supports this CPU. When I tried to update 
it I had to reinstall from the original CD, restore my settings and 
apply all updates apart from the latest one.


Other than that, Smoothwall is an excellent product.

--
Dom


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4daf250d.6000...@rpdom.net

Re: Why does Linux crash?

[Boyd Stephen Smith Jr.]

In <4dae819c.8020...@biotec.tu-dresden.de>, Alex Mestiashvili wrote:
>just a simple forkbomb will make your OS unusable ...

It shouldn't.  When I first got my current hardware up and running, I fork 
bombed myself and then recovered just to prove the OS and hardware could 
handle it.  I was able to login from another terminal (/sbin/login uses exec() 
to start your shell so it doesn't need another pid) and use the /proc file 
system as well as various shell built-ins to terminate all the fork bomb 
processes.

"Yanking the power" is rarely the correct thing to do.  Even if normal 
keyboard and mouse input appear to be ignored, there should be some attempt to 
use the "Magic" SysRq to take the keyboard back from X, terminate / kill 
various processes, sync / remount read-only file systems, produce a crash 
dump, etc. including doing a warm reboot or a soft power off.

If you were actively using lots of swap and have locked up X11, it can take 
quite a while to switch to a vt and kill eclipse.  It will normally be 
possible though.  In cases where that isn't possible, Alt+SysRq+K might help 
get rid of X and processes that are talking to it.  If that fails to give you 
back the system, Alt+SysRq+E will, but it'll also shut down all the system 
services, so you'll need to restart them.

In short, there's a ocean of possibilities between "wait for Linux/Eclipse to 
resolve things" and "yank the power" that you don't seem to be exploring.
-- 
Boyd Stephen Smith Jr.   ,= ,-_-. =.
b...@iguanasuicide.net   ((_/)o o(\_))
ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-'
http://iguanasuicide.net/\_/


signature.asc
Description: This is a digitally signed message part.

Please recommend an external HDD enclosure

[Panayiotis Karabassis]

Hi,

I made the mistake of purchasing an external HDD enclosure without
researching first. The enclosure wouldn't be recognized in Debian
Squeeze or Wheezy with errors that are not relevant since I've returned
it to the shop.

The drive I want to ...enclose is a 3.5 inch PATA HDD. Acceptable
interfaces are USB (any version) and it seems my motherboard also has an
eSata input.

I've spent quite some time googling but couldn't reach any solid
conclusions.

So please recommend an enclosure that is known to work in Debian,
preferably a popular brand, so that it stands a chance of being
available in my country.

-- 
Regards
Panayiotis Karabassis


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4daf3032.8080...@gmail.com

powerdown harddisk

[geertsky]

Hello group,
I want to powerdown my ide harddisk completly. 
I'm booting from a usb stick so the IDE disk is CONSUMING... I have an
old laptop so I'd like to preserver as much as possible.

I tried hdparm -Y /dev/sda but hdparm -C /dev/sda reports it's only in
standby...
And actually I can hear it spinning...
It must be possible to power it down completly no?

Any ideas?


Thanks!
Greetings,
Geert


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/c6590740-af86-4997-a486-54852aa93...@v16g2000vbq.googlegroups.com

Re: Please recommend an external HDD enclosure

[Mark]

Hi,

I've been using this one with Debian Lenny for a while.  Have a 250 GB IDE
hdd in it currently.

http://www.newegg.com/Product/Product.aspx?Item=N82E16817182154&cm_re=hdd_enclosure-_-17-182-154-_-Product

HTH.
Mark

On Wed, Apr 20, 2011 at 12:12 PM, Panayiotis Karabassis wrote:

> Hi,
>
> I made the mistake of purchasing an external HDD enclosure without
> researching first. The enclosure wouldn't be recognized in Debian
> Squeeze or Wheezy with errors that are not relevant since I've returned
> it to the shop.
>
> The drive I want to ...enclose is a 3.5 inch PATA HDD. Acceptable
> interfaces are USB (any version) and it seems my motherboard also has an
> eSata input.
>
> I've spent quite some time googling but couldn't reach any solid
> conclusions.
>
> So please recommend an enclosure that is known to work in Debian,
> preferably a popular brand, so that it stands a chance of being
> available in my country.
>
> --
> Regards
>Panayiotis Karabassis
>
>
> --
> To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact
> listmas...@lists.debian.org
> Archive: http://lists.debian.org/4daf3032.8080...@gmail.com
>
>

Re: powerdown harddisk

[Klistvud]

Dne, 20. 04. 2011 21:06:33 je geertsky napisal(a):


Any ideas?


IIRC (it's been some time) the combination of -B and -S parameters used  
to work for my IDE drives.
Of course, you must first make sure that *nothing whatsoever* (such as  
the Gnome automounter and the like) is accessing the drive, or it will  
never spin down. Test it in runlevel 1 first (init 1).


--
Cheerio,

Klistvud  
http://bufferoverflow.tiddlyspot.com
Certifiable Loonix User #481801  Please reply to the list, not to  
me.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1303331391.18252.1@compax

unscribe

[Leonardo Cuyar Morales]

Imagination is more important than knowledge
  Albert Einstein


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/7517620.6056.1303335154604.JavaMail.root@zimbra

Re: web page image alignment

[Dean Allen Provins, P. Geoph.]

Wayne and Weaver:

On Tue, Apr 19, 2011 at 08:02:43PM -0400, Wayne Topa wrote:
> On 04/19/2011 07:47 PM, Heddle Weaver wrote:
> >On 20 April 2011 07:33, Dean Allen Provins, P. Geoph.<
> >provi...@telusplanet.net>  wrote:
> >
> >>Hello:
> >>
> >>I'm trying to use the Writer program to put a web page together.
> >>
> >>I want to include some images, and I'd like to cut back on the
> >>document length.
> >>
> >>I tried placing the PNG images side by side (I used the LEFT alignment
> >>and RIGHT alignment controls on the menu bar), and they looked just fine
> >>in the Writer document.
> >>
> >>Sending them out to HTML (via export) and displaying them with "iceweasel"
> >>(and I also tried the most recent Firefox, and the installed Epiphany
> >>Web Browser) wasn't very satisfactory.  They overlaid each other!
> >>
> 
> I have not used it in a few years but I suggest you take at the
> Debian kompozer package.  I tried it when OO didn't cut the mustard
> and was very satisfied with thew results.
> 
> WT
> 
> >>Any ideas on how to correct this?
> >>
> >
> >In the repositories, in any distribution from Squeeze onward and even
> >available in Lenny through backports:
> >
> >*http://tinyurl.com/3o9acyn
> >*
> >Writer is essentially an html capable word processor but is far from a
> >mature web authoring app.
> >Regards,
> >
> >Weaver.

Thanks for the feedback.  I've installed "kompozer" and am now
experimenting with it.  So far it looks very good.

Dean

-- 
   Dean Provins, P. Geoph.
 dprov...@alumni.ucalgary.ca
  KeyID at at pgpkeys.mit.edu:11371: 0x9643AE65
  Fingerprint: 9B79 75FB 5C2B 22D0 6C8C 5A87 D579 9BE5 9643 AE65


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110420204710.GA4691@ve6wvc

Problem with udev device identification

[Sebastian Lara]

Hi everyone,

I have three different serial devices connected to three usb-serial
converters (FTDI USB-RS232). If I connect them, I get the generic
device names /dev/ttyUSBx as expected. When I try to write some udev
rules, udevadm shows me the same output for the three devices except
in some device serial id attributes.

Those serial id's will change on every device so if I use a different
usb-serial converter, my rules won't work anymore. So, I need a way to
communicate with the specific device connected to the serial
converters and identify it by its response.

"Writing udev rules" HOWTO, says the following about using PROGRAM
functionality to call an external program to name devices:

PROGRAM is used for running programs which produce device names (and
they shouldn't do anything other than that). When those programs are
being executed, the device node has not yet been created, so acting
upon the device in any way is not possible,

Rule example: SUBSYSTEM=="usb", PROGRAM="/bin/device_namer %k", SYMLINK+="%c"

so I can't write to /dev/ttyUSBx from there to identify the device.
Then, I have the RUN functionality, but it seems that it can't create
a new SYMLINK as udev won't get RUN's output.

Rule example: SUBSYSTEM=="usb", RUN="/bin/program"

Can I instruct udev from that RUN script to create the symlink? Or can
I create that symlink by hand from there?

Thanks

-- 
Sebastián Lara Menares
Ingeniería Civil Electrónica
Universidad de Concepción


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/BANLkTikPy+t3=jccytk6dm3xjsfteyo...@mail.gmail.com

Re: pbuilder and CONCURRENCY_LEVEL

[Niccolò Belli]

Il 20/04/2011 17:30, Niccolò Belli ha scritto:
> How can I set CONCURRENCY_LEVEL when using pbuilder?
> CONCURRENCY_LEVEL=9 pbuilder build file.dsc doesn't work.

Found it, there is an option to pass parameters to dpkg-buildpackage:
--debbuildopts "-j9"

Darkbasic


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4daf5bb5.3050...@gmail.com

Anyone using a StarTech PCI420USB card? (USB 2.0, 4-port, PCI card)

[Rick Thomas]

I just bought and installed a StarTech PCI420USB card for use in my  
PowerMac G4 AGP Graphics running Debian Lenny.


It's a PCI card that has 4 external USB 2.0 ports and one internal port.

When I try to plug a hub or a USB flash stick into it, I get the  
following messages in syslog:


kernel: [ 1608.271167] usb 5-2: new high speed USB device using  
ehci_hcd and address 9

kernel: [ 1623.383727] usb 5-2: device descriptor read/64, error -110
kernel: [ 1638.600348] usb 5-2: device descriptor read/64, error -110
kernel: [ 1638.816222] usb 5-2: new high speed USB device using  
ehci_hcd and address 10

kernel: [ 1653.928755] usb 5-2: device descriptor read/64, error -110
kernel: [ 1669.145281] usb 5-2: device descriptor read/64, error -110
kernel: [ 1669.361314] usb 5-2: new high speed USB device using  
ehci_hcd and address 11
kernel: [ 1679.769656] usb 5-2: device not accepting address 11,  
error -110
kernel: [ 1679.881645] usb 5-2: new high speed USB device using  
ehci_hcd and address 12
kernel: [ 1690.290003] usb 5-2: device not accepting address 12,  
error -110
kernel: [ 1690.290070] hub 5-0:1.0: unable to enumerate USB device  
on port 2
kernel: [ 1690.670048] usb 3-2: new full speed USB device using  
uhci_hcd and address 2

kernel: [ 1706.330581] usb 3-2: device descriptor read/64, error -110
kernel: [ 1722.203119] usb 3-2: device descriptor read/64, error -110
kernel: [ 1722.419123] usb 3-2: new full speed USB device using  
uhci_hcd and address 3
kernel: [ 1738.019719] hub 3-0:1.0: unable to enumerate USB device  
on port 2




The StarTech sales page says it works with Linux (and MacOS, and  
Windows, for what that's worth) but that probably means RedHat and  
x86, not Debian and PowerPC.


Anybody got any clues?

Thanks!

Rick


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/3d0b992f-b62d-4f5a-b831-ed480358f...@pobox.com

Re: Anyone using a StarTech PCI420USB card? (USB 2.0, 4-port, PCI card)

[Rick Thomas]

On Apr 20, 2011, at 6:46 PM, Rick Thomas wrote:




The StarTech sales page says it works with Linux (and MacOS, and  
Windows, for what that's worth) but that probably means RedHat and  
x86, not Debian and PowerPC.


Anybody got any clues?

Thanks!

Rick



In case it matters, StarTech's web page says it is using the "VIA  
VT6212" chip set.



Rick


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/0f130977-4c9f-4e7e-8d98-72b11afc3...@pobox.com

Mediawiki extensions for Debian

[John W Foster]

I'm looking for more Mediawiki extensions packaged for Debian..stable
Mediawiki 1.1xxx

Any suggestions? And of course I know I can install them myself, just
lazy & want to keep the system all Debian.
TIA:
John


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1303343483.5103.3.ca...@beast.home

Re: [OT] Purchasing a wired switch; advice needed

[Stan Hoeppner]

Dom put forth on 4/20/2011 1:25 PM:

> I run Smoothwall on an old Athlon (K6) system that I was given. I

The K7 is the original Athlon.  The K6 series chips had no marketing
name other than "K6", i.e. K6, K6-2, K6-2+, K6-3, K6-3+, etc.

> The only issue I have is that the latest upgrade to Smoothwall Express
> (the free version) no longer supports this CPU. When I tried to update
> it I had to reinstall from the original CD, restore my settings and
> apply all updates apart from the latest one.

TTBOMK Smoothwall Express uses generic i386 kernel and user space
binaries when should run fine on any CPU all the back to the Pentium
Pro, at minimum, and possibly all the way back to the 80386.  That's
what the documentation states.  An update should not have failed due to
your system having a K6 chip.  Any i386 and high 32 bit CPU is
supported.  Is it possible you selected the 64bit binary for the
upgrade, and that's what hosed your system?

-- 
Stan


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4daf8ebd.6020...@hardwarefreak.com

Ideas for mapping users to specific uids

[Brad Alexander]

I know the answer is going to be LDAP, but that's not really an option for
me.

I have, at work, a number of boxes with various users spread across our
network. And I have encountered this on several different jobs. In essence,
say you have four users, e.g. user1, user2, user3 and user4...and for the
sake of this problem, say there are 10 boxes, host 1 through host 5. User1
is the administrator for the network, so that person has accounts on all 5
boxes. User2 has an account on box 2, 4 and 5, user3 has accounts on 1, 2
and 4, user 4 has accounts on 2, 4 and 5, and box 3 is an NFS server.

What I would liike to do is set it up so user1 has uid 1000, user2, 1001,
and so forth. Is there an easy way to set this up on my network such that
when I create a user, that user can have the same uid across the board?
Ideally, I would like to be able to subdivide the uid space so that, say,
admins are always 1000-1050, dbas are 1051-1075, users are 2000-2100, etc.

I know LDAP can do this, but at work, it is not an option, at least in the
short term...

Thanks,
--b

Re: Cannot mount floppy drive in Squeeze

[Stephen Powell]

On Fri, 15 Apr 2011 11:56:47 -0800, Peter Easthope wrote:
> Bob, Camaleon, Dom, Stephen, Tom & others,
> 
> I am replying to the last message I found in the thread.  If there 
> is one later, I didn't ignore it deliberately.
> 
> * From: Stephen Powell 
> * Date: Mon, 6 Dec 2010 22:21:13 -0500 (EST)
>> No, the data is intact, as the mounting of the image file with
>> the loop option confirms.  Also, further experimentation seems
>> to suggest that if the floppy disk is physically mounted in the
>> floppy drive during boot, then I can logically mount it with the
>> mount command after boot.  This really is looking like a bug.
>> I intend to file a bug report, but I'd like to do some more
>> trial-and-error experimentation before I do.
> 
> This is relevant.
>   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=483781
>   
> The best explanation appears to be cited by A.E. Patrakov in the bug report.
>   http://www.linuxfromscratch.org/lfs/view/6.3/chapter08/fstab.html
> Scroll to the bottom of the page and read "Filesystems with MS-DOS ...".
> 
> My encounter with this has been an attempt to make a bootable 
> diskette for updating the BIOS on a Foxconn board.  Did Foxconn 
> have their filesystem mounted with UTF8 and the US codepage?  
> If they used strictly ASCII filenames, are the encoding and 
> codepage a concern?  
> 
> [Use MS or have more trouble than bargained for ... again!]
> 
> Regards, ... Peter E.

First of all, I accidentally deleted your e-mail, and my MUA does not
allow me to edit the headers to add an "In-reply-to" header; so this
reply may not be properly connected to your thread in the archives.
Sorry about that.

Second, thank you for responding.  Third, there was more to that original
thread.  But I can't find it either.  Maybe I accidentally replied
off-list?  Anyway, the resolution to my original thread was udisks.
I run GNOME, and GNOME has a udisks daemon that tries to automatically
mount things like USB thumb drives, CD-ROM disks, and yes, floppy disks,
when they are inserted.  I was able to work around the problem by using
the udisks command to mount and unmount floppy disks instead of the
native mount and umount commands.

-- 
  .''`. Stephen Powell
 : :'  :
 `. `'`
   `-


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1806358399.74644.1303352679315.javamail.r...@md01.wow.synacor.com

Re: Ideas for mapping users to specific uids

[Ron Johnson]

On 04/20/2011 09:16 PM, Brad Alexander wrote:

I know the answer is going to be LDAP, but that's not really an option
for me.

I have, at work, a number of boxes with various users spread across our
network. And I have encountered this on several different jobs. In
essence, say you have four users, e.g. user1, user2, user3 and
user4...and for the sake of this problem, say there are 10 boxes, host 1
through host 5. User1 is the administrator for the network, so that
person has accounts on all 5 boxes. User2 has an account on box 2, 4 and
5, user3 has accounts on 1, 2 and 4, user 4 has accounts on 2, 4 and 5,
and box 3 is an NFS server.

What I would liike to do is set it up so user1 has uid 1000, user2,
1001, and so forth. Is there an easy way to set this up on my network
such that when I create a user, that user can have the same uid across
the board? Ideally, I would like to be able to subdivide the uid space
so that, say, admins are always 1000-1050, dbas are 1051-1075, users are
2000-2100, etc.

I know LDAP can do this, but at work, it is not an option, at least in
the short term...



nis?

http://www.tldp.org/HOWTO/NIS-HOWTO/glossary.html#AEN120

--
"Neither the wisest constitution nor the wisest laws will secure
the liberty and happiness of a people whose manners are universally
corrupt."
Samuel Adams, essay in The Public Advertiser, 1749


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4daf9739.5010...@cox.net

Re: [OT] Purchasing a wired switch; advice needed [question]

[Andrew McGlashan]

Hi Paul,

Paul E Condon wrote:
 > I'm lurking here, looking to better understand a problem that I've never

had to confront: NAT, I understand requires translation tables, one entry
for each active tcp connection. This takes RAM. It also takes enough 
CPU cycles to maintain this table --- set up new connections and find

and delete old connections that are no longer in use. I can see a number
of reasons why the old router would have to quit and need resetting.
I wonder if one of the uplist suggestions which was to set up an old
computer as a router might not work unless the hardware is powerful
enough to keep up with the heavy maintenance activity on the NAT tables.
Extra RMA also implies extra cpu cycles to find and remove old table
entries. It might be that for this service, a somewhat heavy duty
computer is needed. 


As others have mentioned, RAM and CPU are the biggest issues.  And yes a 
PC is likely to use more power, however I think some newer PCs would be 
ideal -- those using low power CPUs, any of those will likely have 
enough memory too.


So, a cheap headless box that is power efficient with ample CPU and RAM 
would be a better choice.  Bridge the modem and let the headless box 
login via PPPoE.  Make sure the cheap box is running an x86 chip though, 
don't go with an ARM or ARMEL one as that will limit your options.


My choice of firewall is IPCop, it wins over Smoothwall options due to 
be "really free".  Smoothwall has limitations depending on version. 
IPCop is long overdue for a major upgrade, but the team is working on 
that.  Having said that, the current release version of IPCop works very 
well and there are a great many very happy users.  I also suggest to 
keep it as a firewall and don't be tempted to bolt on extras from third 
parties.


A cheap, lower power, headless NEW machine will likely not cost too much 
when compared to other options.  Although you need to make sure you have 
"enough" network connections available -- one for RED (Internet), one 
for GREEN (trusted LAN).


Splitting networks and having multiple GREEN networks is more of an 
issue, but it can be done.  Two IPCop boxes working each with their own 
public IP could be an option if the cable provider will supply two 
public IPs to use.  There is no point going too far down this track if 
the ISP won't play ball to start with though.


Having said all that, many around here would be using a Debian box as 
their firewall.  That might be the way to go, then you can choose from 
more architecture (ARMEL for instance).  But if you want a good 
replacement for the current situation without too much work, then IPCop 
is well worth looking into, but it will be better to use an x86 based 
CPU.  IPCop is optimized [and hardened by design] for it's job, a Debian 
install won't be, unless it is specifically a variant that is meant for 
firewall duties with appropriate hardening.



Coming from me, the above is little more than idle speculation. I'd be
interested in reading your comments on this speculation. Does it make
any sense?  How might OP go about specifying hardware that doesn't
need resetting in his environment? Would it be likely that some
'industrial strength' routers meet his needs, and others are really
expensive over kill?  Would the appropriate industrial grade router be
a bigger/smaller energy hog than something cobbled together out of an
old computer and junk box network cards? Or compare to new consumer
suitably sized for the job?


An older clunker machine will use more power, for sure, but a newer 
low-power machine would be much more efficient.  At one stage I kept a 
few PIIIs around because they were much less power hungry than P4 CPUs. 
 Anything from Core 2 Duo onwards is going to be more efficient, but an 
Intel based low power CPU should be fine.



Whatever you have time to write will be interesting to me and
might be really useful to OP. 


Thanks.

--
Kind Regards
AndrewM

Andrew McGlashan
Broadband Solutions now including VoIP


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4dafa99d.3030...@affinityvision.com.au

X EDID reads fail after recent sid update

[Joe Neal]

Hello.

I recently upgraded my sid box with a Gforce 8600.  Following a reboot
the binary nvidia drivers would only give me a screen resolution so
small it was unusable and nouveau failed to find any usable resolutions
at all.  Analysis of recent X logs indicated that in the case of both
drivers, EDID reads were failing, providing X with no data.  When
testing with a GRML live CD, EDID data was returned using nouveau, so
this isn't a hardware problem.  

Since the problem occurs with both nvidia drivers, it's not a driver
problem.  Since EDID reads work properly with the GRML live CD it's not
a hardware problem.  I don't think any X packages have been upgraded at
all since the last boot that could have caused the problem.  I'm at a
loss as to where the problem is coming from.

Any guesses?

Joe Neal.


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110420225409.6ef2e10f@Eliza

Re: file systems

[prad]

Stan Hoeppner  writes:

> you left out the best, most mature, highest performance Linux
> filesystem of them all:  XFS
>
that i did stan! i'd completely forgotten about it ever since i heard it
was good only for big files many years ago (i never really investigated
it either back then).

we don't have a need for better performance since we aren't running a
high production server. on the other hand, the possibility of learning
and using a different fs is appealing to both my son and myself since we
setup our system as an educational venture.

we want to run our servers through virtual box off usb drives which is a
total departure from what we've done over the years. so might as well
throw in a new fs too. :D

this is a wonderful list and there are so many informative answers to my
original post!

i very much appreciate yours in particular, stan!
we will do as you say and start reading up about fs and especially about
xfs!

-- 
in friendship,
prad


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/878vv4gqoq@towardsfreedom.com

Re: [OT] Purchasing a wired switch; advice needed

[Zoran Kolic]

Howdy!
I already found good advice in one of the posts on your
question. Personally, I'd stick with pair of linksys
wrt54gl, putting dd-wrt on both or three of them and ma-
king it do the work. The site has tutorials for all of
situations one might experience.
For another approach, there is better hardware, i.e. asus
n16. People tend to like it much and use it for everything.
There is also one netgear usb model, 3000 l, able to torrent
directly. You just have to put new firmware for that (or not,
I'm not aware if native could do). A lot of people made 
patches to such a circumstances.
Every wifi router works as wired one if you turn wireless off.
To have big bang for buck, get 54gl. First step anyway is to
know exactelly what you want to do.
Best regards

Zoran


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20110421043420.GA968@faust

Re: X EDID reads fail after recent sid update

[Sven Joachim]

On 2011-04-21 05:54 +0200, Joe Neal wrote:

> I recently upgraded my sid box with a Gforce 8600.  Following a reboot
> the binary nvidia drivers would only give me a screen resolution so
> small it was unusable and nouveau failed to find any usable resolutions
> at all.  Analysis of recent X logs indicated that in the case of both
> drivers, EDID reads were failing, providing X with no data.  When
> testing with a GRML live CD, EDID data was returned using nouveau, so
> this isn't a hardware problem.  
>
> Since the problem occurs with both nvidia drivers, it's not a driver
> problem.  Since EDID reads work properly with the GRML live CD it's not
> a hardware problem.  I don't think any X packages have been upgraded at
> all since the last boot that could have caused the problem.  I'm at a
> loss as to where the problem is coming from.
>
> Any guesses?

I don't really have an idea myself, but some people reported a problem
with corrupted EDID on https://bugs.freedesktop.org/show_bug.cgi?id=34554.
The information in that bug may be useful.

Sven


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87ipu8ch82@turtle.gmx.de

Re: file systems

[Stan Hoeppner]

prad put forth on 4/20/2011 11:43 PM:

> we want to run our servers through virtual box off usb drives which is a
> total departure from what we've done over the years. so might as well
> throw in a new fs too. :D

Why USB?

-- 
Stan


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4dafd06e.8040...@hardwarefreak.com

Re: [OT] Purchasing a wired switch; advice needed

[Dom]

On 21/04/11 02:56, Stan Hoeppner wrote:

Dom put forth on 4/20/2011 1:25 PM:


I run Smoothwall on an old Athlon (K6) system that I was given. I


The K7 is the original Athlon.  The K6 series chips had no marketing
name other than "K6", i.e. K6, K6-2, K6-2+, K6-3, K6-3+, etc.


The only issue I have is that the latest upgrade to Smoothwall Express
(the free version) no longer supports this CPU. When I tried to update
it I had to reinstall from the original CD, restore my settings and
apply all updates apart from the latest one.


TTBOMK Smoothwall Express uses generic i386 kernel and user space
binaries when should run fine on any CPU all the back to the Pentium
Pro, at minimum, and possibly all the way back to the 80386.  That's
what the documentation states.  An update should not have failed due to
your system having a K6 chip.  Any i386 and high 32 bit CPU is
supported.  Is it possible you selected the 64bit binary for the
upgrade, and that's what hosed your system?



I use the web interface to find and install the recommended updates.
The one that caused the problem is update8-i386.

The description is "This update greatly improves SmoothWall's driver 
support by updating the Linux kernel to 2.6.32, as well improving system 
secuirty through various version bumps."


The error I get upon boot is that the processor does not support CMOV, 
which is true of the K6 range.


--
Dom


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4dafd0f7.6070...@rpdom.net