Linux-Misc Digest #116, Volume #26               Mon, 23 Oct 00 01:13:04 EDT

Contents:
  Re: error msg in syslog (OTR Comm)
  Re: Controlling several boxes from one place (Jim Broughton)
  Re: Controlling several boxes from one place (Jim Broughton)
  Re: System spends too much time in X, but only sometimes...? (Hal Burgiss)
  diald, pppd,  and wvdial timeouts
  Re: Which Linux distro most 'generic' *nix ? (Bill Vermillion)
  Re: GPL questions? (Neil Cherry)
  Re: Emulation and writing to the code segment (Garry Knight)
  Star Office 6.0? (John Scudder)
  Re: Stupid question about telnet? (GD)
  FOR ALL VOTERS - PLS READ (Private User)
  Re: FOR ALL VOTERS - PLS READ (Hal Burgiss)
  Re: Screen Saver Question (Fred Mulharin)
  Re: rpm or debian packaging? (Phillip Deackes)

----------------------------------------------------------------------------

From: OTR Comm
Subject: Re: error msg in syslog
Date: Mon, 23 Oct 2000 02:23:30 GMT

It looks like our hardware clock is on the lam.  Can't get any response from it with 
/sbin/clock.

Do you by any chance know of any routines for a linux system that we can use to reset 
the cmos clock
without shutting the system down and going into the bios?

I hope that it is just some garbage that got into the hardware clock and not total 
failure, but I
would like to check this, if possible, without shutting the system down.

It may be new motherboard time.  URGH!!

Thanks for your reply,
Murrah Boswell

On 22 Oct 2000 22:21:24 GMT, [EMAIL PROTECTED] (Dances With Crows) wrote:

>On Sun, 22 Oct 2000 21:14:27 GMT, OTR Comm wrote:
>>Can some one tell me what is causing messages like this in my syslog:
>>Oct 22 13:35:23 sec kernel: set_rtc_mmss: can't update from 52 to 5
>
>Not completely sure, but it appears as if your CMOS clock isn't
>completely happy.  The relevant code is in /usr/src/linux/arch/XXXX/
>kernel/time.c about line 300.  Do you notice the time as reported by the
>CMOS clock drifting far off from the real time, or anything like that?
>
>-- 
>Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
>Brainbench MVP for Linux Admin /  Workin' in a code mine, hittin' Ctrl-Alt
>http://www.brainbench.com     /   Workin' in a code mine, whoops!
>-----------------------------/    I hit a seg fault....


------------------------------

From: Jim Broughton <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.hardware
Subject: Re: Controlling several boxes from one place
Date: Mon, 23 Oct 2000 02:21:10 GMT

Bo Berglund wrote:
> 
> Hi,
> I have a small home network consisting of 2 linux (RH) machines and 2
> WinNT and 1 Win95 PC. In my "office" I have very limited space so I
> can have only one monitor and keyboard/mouse. The Win95 is separate
> with its own kb/mon and one NT is a laptop, but I still have 3
> machines to access from my crammed desk. So now I have to crawl on the
> floor every time I want to switch and move monitor connections and
> bring up the keyboards from the floor to my desk. Very awkward...
> 
> Question:
> Is there some freeware software for WinNT (still my main work machine)
> that can emulate an X terminal so I can access the Linux boxes via the
> internal network?
> I have tested an eval version of "KEA! X" and it works good, but it is
> not freeware, instead rather expensive. I'd prefer some
> freeware/shareware solution if it can do the job.
> 
> Advice please...
> 
> Bo Berglund
> 
> [EMAIL PROTECTED]
> 
> PGP: My public key is available at the following locations:
> Idap://certserver.pgp.com
> http://pgpkeys.mit.edu:11371

 How about a combination of samba for shares on the linux boxes and
a telnet session to each linux box. If it has to be graphical then
getting an X client for NT is your only solution. I myself employ
a few different solutions. 1 NFS and 2 a KVM switch. I can also telnet
into my server box but that is slower than hitting the control key twice
to actualy get there (KVM). I am not certain how much NFS is for NT but for
linux it is free as is being a samba server/client. I sure you know by
now that samba is = to M$ network neighborhood. I do not use samba.
  If you already have the boxes connected with network wire then
telneting into the linux boxes is a snap. Login as a normal user then
su to root if you need too. This does not get you a graphical interface
but it does get you there in console from which you can do almost anything you
can do in graphical mode except play with the mouse.


-- 
Jim Broughton
(The Amiga OS! Now there was an OS)
If Sense were common everyone would have it!
Following Air and Water the third most abundant
thing on the planet is Human Stupidity.

------------------------------

From: Jim Broughton <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.hardware
Subject: Re: Controlling several boxes from one place
Date: Mon, 23 Oct 2000 02:22:32 GMT

Bo Berglund wrote:
> 
> Hi,
> I have a small home network consisting of 2 linux (RH) machines and 2
> WinNT and 1 Win95 PC. In my "office" I have very limited space so I
> can have only one monitor and keyboard/mouse. The Win95 is separate
> with its own kb/mon and one NT is a laptop, but I still have 3
> machines to access from my crammed desk. So now I have to crawl on the
> floor every time I want to switch and move monitor connections and
> bring up the keyboards from the floor to my desk. Very awkward...
> 
> Question:
> Is there some freeware software for WinNT (still my main work machine)
> that can emulate an X terminal so I can access the Linux boxes via the
> internal network?
> I have tested an eval version of "KEA! X" and it works good, but it is
> not freeware, instead rather expensive. I'd prefer some
> freeware/shareware solution if it can do the job.
> 
> Advice please...
> 
> Bo Berglund
> 
> [EMAIL PROTECTED]
> 
> PGP: My public key is available at the following locations:
> Idap://certserver.pgp.com
> http://pgpkeys.mit.edu:11371

 Oops one final note you do NOT have to place a monitor on a linux
box running as an X server.

-- 
Jim Broughton
(The Amiga OS! Now there was an OS)
If Sense were common everyone would have it!
Following Air and Water the third most abundant
thing on the planet is Human Stupidity.

------------------------------

From: [EMAIL PROTECTED] (Hal Burgiss)
Subject: Re: System spends too much time in X, but only sometimes...?
Reply-To: Hal Burgiss <[EMAIL PROTECTED]>
Date: Mon, 23 Oct 2000 02:34:42 GMT

On Sun, 22 Oct 2000 19:51:11 -0400, Jean-David Beyer
<[EMAIL PROTECTED]> wrote:
>Hal Burgiss wrote:
>
>
>> Ah, another one! It is an SMP issue. I've been chasing this for a
>> year now. If there is an answer to this mystery it is buried in some
>> crypt somewhere. I also have a G200, but not sure if that is a factor
>> or not. Also, not everyone with SMP sees this, so there are other
>> factors as well. You will notice some apps are bigger offenders than
>> others: realplayer, xmms, rxvt, some of the WindowMaker doc apps.
>>
>When I notice this, I am running none of those.

I just mention these as examples of apps that exacerbate the problem
that I see. Almost any X app, and really anything using unix domain
sockets, will show this effect. Some much more than others though. Not
sure we are seeing the same problem, but maybe. If you want to see if
this is it, just start xmms or realplayer, and watch top for a while.


>My CPUs peg at 100% most of the time because of the two setiathomes (at nice
>19) and my dbms server programs that are kind of busy right now:

Yea, you've got a lot going on.

This is me ATM:

 10:32pm  up 1 day, 22:17,  3 users,  load average: 0.26, 0.16, 0.09
103 processes: 102 sleeping, 1 running, 0 zombie, 0 stopped
CPU0 states:  0.4% user,  2.1% system,  0.0% nice, 96.4% idle
CPU1 states:  0.1% user,  2.0% system,  0.0% nice, 97.3% idle
Mem:   128040K av,  118380K used,    9660K free,   41640K shrd,    2768K buff
Swap:  321432K av,   33848K used,  287584K free                   29640K cached
  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
 9359 hal       11   0   984  984   756 R     1.5  0.7   0:01 top
 7574 root      18   0 47336  32M  3952 S     1.3 25.9   7:06 X
 7598 hal        8   0   460  460   224 S     1.3  0.3   2:59 wmsysmon
 7604 hal        3   0   260  144   100 S     0.5  0.1   1:22 wmSMPmon
 7603 hal        3   0   352  300   252 S     0.1  0.2   1:34 wmnet


A few mintutes before (without me doing anything, just watching). top
didn't catch this at the peak of the spike.

 10:22pm  up 1 day, 22:07,  3 users,  load average: 0.80, 0.25, 0.08
103 processes: 102 sleeping, 1 running, 0 zombie, 0 stopped
CPU0 states: 16.2% user,  9.4% system,  0.0% nice, 73.3% idle
CPU1 states: 23.2% user,  8.3% system,  0.0% nice, 67.4% idle
Mem:   128040K av,  111680K used,   16360K free,   36780K shrd,    2636K buff
Swap:  321432K av,   34032K used,  287400K free                   24516K cached
  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME COMMAND
 7574 root      14   0 47176  30M  2088 S    29.1 24.3   6:44 X
 9290 hal        3   0 19800  19M 10632 S     6.2 15.4   0:06 mozilla-bin
 7603 hal        1   0   352  300   252 S     3.5  0.2   1:29 wmnet
 7598 hal        1   0   460  460   224 S     2.1  0.3   2:52 wmsysmon
 9241 hal        0   0  7900 7884  3756 S     2.1  6.1   0:04 realplay
 7578 hal        1   0  1940 1940   972 S     1.9  1.5   0:08 wmaker
 7602 hal        0   0   480  336   280 S     1.9  0.2   0:19 wmWeather
 9267 hal        4   0  1000 1000   756 R     1.9  0.7   0:02 top
 7600 hal        0   0   288  288    80 S     1.7  0.2   0:08 mount.app
 7601 hal        0   0   448  448   228 S     1.3  0.3   0:21 wmmail
 9237 hal        0   0  1340 1340   420 S     1.3  1.0   0:04 xmms
 7599 hal        0   0   980  980   172 S     1.1  0.7   0:11 wmglobe


-- 
Hal B
 [EMAIL PROTECTED]
 [EMAIL PROTECTED]
 [EMAIL PROTECTED]
--

------------------------------

From: [EMAIL PROTECTED] ()
Subject: diald, pppd,  and wvdial timeouts
Date: 23 Oct 2000 02:47:34 GMT

Hoping for some help:

I'm running RedHat 7 and can successfully connect to my ISP using wvdial.
I'm trying to connect the house network via modem on the Linux box.
I've setup diald to accept connections. 
However, after diald initiates connection and I get an IP address, it dies, saying 
that pppd script times out. Afterwards, I can't get back to a "clean" system and have 
to kill diald, pppd process.

. 

var/log/messages error is: 

...diald[8847]: Connect script timed out. Killing script.

I suppose diald is killing pppd script. How can I tell diald that pppd has connected 
successfully or how shall I tell pppd to end its script because a successful 
connection was made?

Thanks a lot!
Jojo

------------------------------

Crossposted-To: alt.os.linux,comp.os.linux.development.apps,comp.unix.admin
From: [EMAIL PROTECTED] (Bill Vermillion)
Subject: Re: Which Linux distro most 'generic' *nix ?
Reply-To: [EMAIL PROTECTED]
Date: Mon, 23 Oct 2000 02:07:55 GMT

In article <[EMAIL PROTECTED]>,
John Hasler  <[EMAIL PROTECTED]> wrote:
>J.H.Delaney writes:
>> So I was wondering which distro comes closest to a generic commercial
>> *nix ?

>There is no generic commercial Unix. Commercial Unices differ more
>than Linux distributions do.

That's not been my experience.  I've found the Linux distributions
- at least from the way I view them - have more variations than the
commercial Unix systems I've used.  I've not used HP/UX and the
last AIX I used was years ago - and the latter was different to say
the least.

>Install Debian and FreeBSD and familiarize yourself with both and
>you will be better prepared to deal with a variety of Unices then
>someone who knows only Solaris.

Learn the Unix system and not the distribution and everything
starts to look somewhat alike.  

-- 
Bill Vermillion -   bv @ wjv . com

------------------------------

From: [EMAIL PROTECTED] (Neil Cherry)
Subject: Re: GPL questions?
Reply-To: [EMAIL PROTECTED]
Date: Mon, 23 Oct 2000 03:08:41 GMT

On Mon, 23 Oct 2000 00:17:57 GMT, John Hasler wrote:
>Neil Cherry writes:
>> if I assign my Copyrights over to the FSF how do they know that they have
>> them? Don't you have to tell them they have Copyrights on this code?
>
>I think you are a bit confused.  Notice the 'if' in your step 3.  Assigning
>copyright to the FSF is not a necessary (or usual) part of releasing a
>program under the terms of the GPL.

oh, yea, I didn't mean to imply all three and I'm more than a bit
confused lately. :-)

>When you release a program under the terms of the GPL, _you_ are licensing
>it to the recipients under _your_ license, which happens to be identical to
>the GPL.  The FSF is not involved except as author of the document you
>modeled your license on.
>
>If you really do want to assign your copyright to the FSF, I think you will
>find complete instructions at www.gnu.org.  I don't think that's what you
>want to do, though.

I think I agree, so it looks like the first 2 steps (the COPY file and
the text in the files to GPL) are all I need.

Thanks, I know it sounds like craziness but I don't want to revisit
this once the decission has been made. I'd rather be coding at that point.

-- 
Linux Home Automation           Neil Cherry             [EMAIL PROTECTED]
http://members.home.net/ncherry                         (Text only)
http://meltingpot.fortunecity.com/lightsey/52           (Graphics)
http://linuxha.sourceforge.net/                         (SourceForge)

------------------------------

From: Garry Knight <[EMAIL PROTECTED]>
Subject: Re: Emulation and writing to the code segment
Date: Mon, 23 Oct 2000 02:54:29 +0100

On Sun, 22 Oct 2000, Mihai Cartoaje wrote:
>Hello,
>
>I have heard of a technique which is supposed to make emulators
>three times faster. The emulator takes a piece of code without
>branches and translates it into native code. Then executes it. If at
>a later time, the emulation comes back to a translated piece of
>code, all the emulator has to do is to run it again.
>
>But this needs to be able to write to the code segment (or to
>execute the data segment.) Under Linux and gcc, an instruction like,
>
>    goto L2;
>L1:
>    a = 2.718281828;
>L2:
>    p = &&L1;
>    *(int *) p = 2;
>
>Gives a SIGINT.
>
>Is there any way?

Think about the underlying code that gets generated by the compiler in this
sequence. At L1, a sequence of machine-level instructions is generated to put a
floating point value into a variable. At L2 you're setting up a pointer to
point to the start of those instructions, then you're trying to write the
integer value 2 ***on top of those instructions***.

My guess is that you're trying to overwrite the value 2.718281828 with the
value of (int) 2. But the (int) 2 - if this worked at all - would overwrite the
first instruction, which is probably something like an LEA ebx,_a instruction
(I'm guessing here, since I haven't used gcc and don't know its register usage
protocols).

As for why you're getting a SIGINT rather than a SIGSEGV, I wouldn't care to
guess as I'm no expert on coding for Linux.

And personally I wouldn't rely on self-modifying code of this sort running on
any processor with an instruction cache.

>Should I ask this in another of the Linux newsgroups?

Umm... gnu.gcc maybe?

--
Garry Knight
[EMAIL PROTECTED]

------------------------------

From: John Scudder <[EMAIL PROTECTED]>
Subject: Star Office 6.0?
Date: Sun, 22 Oct 2000 23:26:09 -0500

Has anyone heard anything about Star Office 6.0?

I read somewhere that this will be a major upgrade.   Supposedly
StarWriter, StarDraw, etc.  will be independent applications so that the
whole Office Suite won't have to open for the sake of one application.

If I recall correctly,  it was to be available by the middle of
October.  Nothing on the Sun Site yet.

John


------------------------------

From: GD <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Subject: Re: Stupid question about telnet?
Date: Mon, 23 Oct 2000 03:43:29 GMT

Don't know anything about the Bastille hardening script, but check
/etc/hosts.deny  and /etc/hosts.allow.

GD

Bo Berglund wrote:

> I have installed RH6.2 on a new PC.
> I also downloaded and ran the Bastille hardening script, but I have
> not proceeded to do anything about the result of the Bastille script
> yet (I assume that I have to do something to enable the protection).
> This is partly because I did not really understand some of the
> questions asked so I decided to let that wait.
>
> Now I have connected the PC to my internal LAN in order to start
> working on the networking part of it. It answers ping OK.
>
> But I cannot telnet to the PC from my WinNT machine and when I look at
> the ipchains setup it lokks like this:
> # ipchains -L
> Chain input (policy ACCEPT):
> Chain forward (policy ACCEPT):
> Chain output (policy ACCEPT):
>
> To me this looks like it will accept anything for now (what I
> expected). So why does it refuse telnet? Where is that set up?
> Or do you have to explicitly *enable* telnet after the RH6.2 install?
>
> please advice
>
> Bo Berglund
>
> [EMAIL PROTECTED]
>
> PGP: My public key is available at the following locations:
> Idap://certserver.pgp.com
> http://pgpkeys.mit.edu:11371


------------------------------

From: Private User <See.Comment.Header@[127.1]>
Date: 20 Oct 2000 09:32:32 -0000
Subject: FOR ALL VOTERS - PLS READ
Crossposted-To: comp.os.linux.networking,comp.unix.admin


The state of Texas, under the leadership of Governor George W. Bush, is
ranked:

50th in spending for teachers' salaries

49th in spending on the environment

48th in per-capita funding for public health

47th in delivery of social services

42nd in child-support collections

41st in per-capita spending on public education

And ...

5th in percentage of population living in poverty

1st in air and water pollution

1st in percentage of poor working parents without insurance

1st in percentage of children without health insurance

1st in executions (average 1 every 2 weeks for Bush's 5 years as Governor)

Just think of what he could do for the country if he were president!

John R. Finnegan Jr., Ph.D.
Professor and Associate Dean For Academic Affairs
School of Public Health, University of Minnesota



------------------------------

From: [EMAIL PROTECTED] (Hal Burgiss)
Crossposted-To: comp.os.linux.networking,comp.unix.admin
Subject: Re: FOR ALL VOTERS - PLS READ
Reply-To: Hal Burgiss <[EMAIL PROTECTED]>
Date: Mon, 23 Oct 2000 04:38:15 GMT

On 20 Oct 2000 09:32:32 -0000, Private User <See.Comment.Header@[127.1]> wrote:

>Just think of what he could do for the country if he were president!
>
>John R. Finnegan Jr., Ph.D.
>Professor and Associate Dean For Academic Affairs
>School of Public Health, University of Minnesota

OK, because of this obnoxious off topic post, I now vote for Bush just
to spite your clueless, self righteous ass. If that's really who you are
since you don't give an email address and hide behind a remailer. At
least Bush supporters aren't spamming Linux newsgroups.

-- 
Hal B
 [EMAIL PROTECTED]
 [EMAIL PROTECTED]
 [EMAIL PROTECTED]
--

------------------------------

From: Fred Mulharin <[EMAIL PROTECTED]>
Crossposted-To: 
comp.os.linux,comp.os.linux.admin,comp.os.linux.help,comp.os.linux.networking,comp.os.linux.questions
Subject: Re: Screen Saver Question
Date: Mon, 23 Oct 2000 00:31:46 -0400

could it Be your Power Management shutting down your screen? ...just a
thought.

Lamar Thomas wrote:
> 
> How do you get the screen savers to work on RH 6.2 running X?  My monitor
> just goes blank after about 5 mins. no matter what screen saver I select!
> Thanks for your help.
> 
> Regards

-- 
Fred Mulharin
[EMAIL PROTECTED]

------------------------------

From: [EMAIL PROTECTED] (Phillip Deackes)
Subject: Re: rpm or debian packaging?
Date: Sun, 22 Oct 2000 19:42:11 GMT

In article <WTGI5.69927$[EMAIL PROTECTED]>, Database wrote:
>I want to start linux but I dont know if rpm or debian packaging is the
>best. If I were to go with rpm, I will get Red Hat 7. If I do debian system,
>i'll go with Storm.

I think most Linux users would agree that the Debian package system is
best - it is more thorough than rpm, especially in the post-install
configuration phase. The real advantage of using a Debian-based
distribution is apt-get. No messing around trying to find apps, just do
'apt-get install whatever' and the app will be downloaded and installed
along with any necessary dependencies. To upgrade your whole system, you
just do 'apt-get dist-upgrade' and all your apps, libraries etc. will be
upgraded where there is a later version on the 'net. You will not need
to buy another CD again.

I used to use rpm-based distros, but I could never go back.

Just don't bother trying Corel Linux - they messed up Debian big time. 

-- 
Phillip Deackes
Using Storm Linux 2000

------------------------------


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.misc) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Misc Digest
******************************

Reply via email to