[CentOS-docs] Suggestions on I Need the Kernel Source Wiki pages

2008-07-12 Thread Antti J. Huhtala
Hi all,

I am a newcomer to CentOS but slightly less so to Linux in general
(FC4-F9). In short, I am trying to make an old SB Vibra 16C (ISA) sound
card work in Centos 5. To achieve this, support for ISA bus and ISA PnP
is apparently needed in kernel. Therefore, a custom kernel needs to be
built. -- Yes, I have read all the warnings of lack of support, etc.

As a first step, I had to build the complete kernel tree according to I
Need the Kernel Source CentOS Wiki. I tried to meticulously follow the
directions given there but it took five or six attempts before I had a
complete tree in ~/rpmbuild/BUILD/. Would you please consider the
following questions in order to decide if any changes/additions are
necessary in the article:

a) In section 2.1 it is said: [EMAIL PROTECTED] yum install rpm-build. To
'yum install' something you need to be superuser (as confirmed by '#'
prompt before the command). Shouldn't the example read: 
[EMAIL PROTECTED] yum install rpm-build?

b) The next command on the page is rpm -i http://mirror.centos.org...;.
Two of my futile attempts to install the ...src.rpm were due to the fact
that 1) I did not understand and then 2) forgot to exit from root before
trying to install the source rpm. Shouldn't there be at least a $
prompt (and perhaps the directory, too) before the 'rpm -i' command?

c) One attempt of 'rpmbuild' failed because redhat-rpm-config is needed
by kernel-2.6.18-92.1.6.el5.i686. This may not apply to everyone but it
certainly applied to me who had only installed 5.1 in March and updated
it to 5.2 in July by just updating when told to do so. Perhaps it would
be advisable to add # yum install redhat-rpm-config to the article?

d) Two attempts of 'rpmbuild' perhaps failed because I inadvertently
used single quotes or apostrophes around 'uname -m' instead of `s
(accent grave?). The result in prep-err.log, however, was: Building
target platforms: uname -m and Building for target uname -m. 
I had to explicitly insert i686 in the 'rpmbuild' command like this:
[EMAIL PROTECTED] SPECS]$ rpmbuild -bp --target=i686 kernel-2.6.spec 2 
If this is what the user is supposed to do anyway, perhaps it could be
worded a little differently to make it perfectly clear? On the other
hand, if 'accent grave' characters around 'uname -m' really work in
replacing the proper architecture in the command, perhaps the text
should emphasize the fact? I foolishly did not cut-and-paste the
'rpmbuild' command but wrote it myself.

I realize there is another thread about these Wiki pages currently being
discussed on centos-docs list but I thought it better to start a new
thread.

TIA, Antti



___
CentOS-docs mailing list
CentOS-docs@centos.org
http://lists.centos.org/mailman/listinfo/centos-docs


Re: [CentOS-docs] Suggestions on I Need the Kernel Source Wiki pages

2008-07-12 Thread Akemi Yagi
On Sat, Jul 12, 2008 at 6:27 PM, Antti J. Huhtala [EMAIL PROTECTED] wrote:
 Hi all,

Hi Antti,

 I am a newcomer to CentOS but slightly less so to Linux in general
 (FC4-F9). In short, I am trying to make an old SB Vibra 16C (ISA) sound
 card work in Centos 5. To achieve this, support for ISA bus and ISA PnP
 is apparently needed in kernel. Therefore, a custom kernel needs to be
 built. -- Yes, I have read all the warnings of lack of support, etc.

 As a first step, I had to build the complete kernel tree according to I
 Need the Kernel Source CentOS Wiki. I tried to meticulously follow the
 directions given there but it took five or six attempts before I had a
 complete tree in ~/rpmbuild/BUILD/. Would you please consider the
 following questions in order to decide if any changes/additions are
 necessary in the article:

 a) In section 2.1 it is said: [EMAIL PROTECTED] yum install rpm-build. To
 'yum install' something you need to be superuser (as confirmed by '#'
 prompt before the command). Shouldn't the example read:
 [EMAIL PROTECTED] yum install rpm-build?

Yes, this should be root@ instead of [EMAIL PROTECTED]

 b) The next command on the page is rpm -i http://mirror.centos.org...;.
 Two of my futile attempts to install the ...src.rpm were due to the fact
 that 1) I did not understand and then 2) forgot to exit from root before
 trying to install the source rpm. Shouldn't there be at least a $
 prompt (and perhaps the directory, too) before the 'rpm -i' command?

The [EMAIL PROTECTED] would make it clearer, yes.

 c) One attempt of 'rpmbuild' failed because redhat-rpm-config is needed
 by kernel-2.6.18-92.1.6.el5.i686. This may not apply to everyone but it
 certainly applied to me who had only installed 5.1 in March and updated
 it to 5.2 in July by just updating when told to do so. Perhaps it would
 be advisable to add # yum install redhat-rpm-config to the article?

Maybe this can be added to or near the yum install rpm-build command
because both should be run as root.

 d) Two attempts of 'rpmbuild' perhaps failed because I inadvertently
 used single quotes or apostrophes around 'uname -m' instead of `s
 (accent grave?). The result in prep-err.log, however, was: Building
 target platforms: uname -m and Building for target uname -m.
 I had to explicitly insert i686 in the 'rpmbuild' command like this:
 [EMAIL PROTECTED] SPECS]$ rpmbuild -bp --target=i686 kernel-2.6.spec 2 
 If this is what the user is supposed to do anyway, perhaps it could be
 worded a little differently to make it perfectly clear? On the other
 hand, if 'accent grave' characters around 'uname -m' really work in
 replacing the proper architecture in the command, perhaps the text
 should emphasize the fact? I foolishly did not cut-and-paste the
 'rpmbuild' command but wrote it myself.

They are back ticks :-)  This is one of those things that might be
regarded as Unix/Linux basics (?).  It feeds the output of the command
inside the ticks to the shell.  In your case, `uname -m` produces i686
which then will be used as the parameter for the --target= option.

 I realize there is another thread about these Wiki pages currently being
 discussed on centos-docs list but I thought it better to start a new
 thread.

Yes, this is a new subject, so you did it just fine.

Thank you for writing this note to share what you have noticed.  The
Wiki article shall be modified accordingly.

Akemi

 TIA, Antti
___
CentOS-docs mailing list
CentOS-docs@centos.org
http://lists.centos.org/mailman/listinfo/centos-docs


Re: [CentOS-docs] Suggestions on I Need the Kernel Source Wiki pages

2008-07-12 Thread Scott Robbins
On Sat, Jul 12, 2008 at 08:39:46PM -0700, Akemi Yagi wrote:
 On Sat, Jul 12, 2008 at 6:27 PM, Antti J. Huhtala [EMAIL PROTECTED] wrote:
  Hi all,
 
 Hi Antti,
  d) Two attempts of 'rpmbuild' perhaps failed because I inadvertently
  used single quotes or apostrophes around 'uname -m' instead of `s
  (accent grave?). 
 
 They are back ticks :-)  This is one of those things that might be
 regarded as Unix/Linux basics (?).  It feeds the output of the command
 inside the ticks to the shell.  In your case, `uname -m` produces i686
 which then will be used as the parameter for the --target= option.

This confuses many newcomers (and old timers with bad vision if they're
not paying attention.)

Is there any merit to substituting that with $().  (To the OP, original
poster, both the back ticks, to the left of the numeral 1 on a QWERTY
keyboard and putting something inside $(), e.g., $(uname -m) mean to
execute a command. 

I write many pages for the beginner.  As that was a common error when
people would read my pages, I then began making a point of explaining
what they were.  Finally, I decided that simply using $() was easier
than adding a sentence or two specifying that those were backticks.  :)

-- 
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6

Buffy: I gotta stop him before he unleashes unholy havoc and 
it's just another Tuesday night in Sunnydale. 
___
CentOS-docs mailing list
CentOS-docs@centos.org
http://lists.centos.org/mailman/listinfo/centos-docs


Re: [CentOS-docs] Suggestions on I Need the Kernel Source Wiki pages

2008-07-12 Thread Scott Robbins
On Sat, Jul 12, 2008 at 09:11:10PM -0700, Akemi Yagi wrote:
 On Sat, Jul 12, 2008 at 9:03 PM, Scott Robbins [EMAIL PROTECTED] wrote:
  On Sat, Jul 12, 2008 at 08:39:46PM -0700, Akemi Yagi wrote:
 
 
  This confuses many newcomers (and old timers with bad vision if they're
  not paying attention.)
 
  Is there any merit to substituting that with $().  (To the OP, original
  poster, both the back ticks, to the left of the numeral 1 on a QWERTY
  keyboard and putting something inside $(), e.g., $(uname -m) mean to
  execute a command.
 
 
 Scott,
 
 There is one (minor) problem with using the $() notation.  That is, it
 does not work in *cough* csh.  I said minor because (t)csh users are
 minority.  Anyway, back ticks work regardless of the shell used.

Well, I'll be darned.  You're right. (I just tried it.)

As you might imagine, I've never used csh for scripts.  

 
 Akemi
 P.S. Please do not start the shell war (or c-shell bashing).  OK, Evolution?

Nope, not me.  FreeBSD, FWIW, still has it as root's default shell,

Perhaps then, it is worth adding a line or two explaining that these are
backticks, (and possible even mentioning where they are on the QWERTY
keyboard.)  The newcomer is often going to assume that they're single
quotes.


-- 
Scott Robbins
PGP keyID EB3467D6
( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 )
gpg --keyserver pgp.mit.edu --recv-keys EB3467D6

Buffy: Cool. You guys can do the brain thing. I'm gonna go to 
class. 
Oz: Which could also be construed as the brain thing. 

___
CentOS-docs mailing list
CentOS-docs@centos.org
http://lists.centos.org/mailman/listinfo/centos-docs


Re: [CentOS-docs] Suggestions on I Need the Kernel Source Wiki pages

2008-07-12 Thread Akemi Yagi
On Sat, Jul 12, 2008 at 9:31 PM, Scott Robbins [EMAIL PROTECTED] wrote:
 On Sat, Jul 12, 2008 at 09:11:10PM -0700, Akemi Yagi wrote:

 There is one (minor) problem with using the $() notation.  That is, it
 does not work in *cough* csh.  I said minor because (t)csh users are
 minority.  Anyway, back ticks work regardless of the shell used.

 Well, I'll be darned.  You're right. (I just tried it.)

 As you might imagine, I've never used csh for scripts.

 P.S. Please do not start the shell war (or c-shell bashing).  OK, Evolution?

 Nope, not me.  FreeBSD, FWIW, still has it as root's default shell,

Well, at my work place, a new Unix account is created with c-shell as
the default.

 Perhaps then, it is worth adding a line or two explaining that these are
 backticks, (and possible even mentioning where they are on the QWERTY
 keyboard.)  The newcomer is often going to assume that they're single
 quotes.

 Scott Robbins

When I come to think about it, building a custom kernel is not really
for people who require help at that level.  Not that I am discouraging
it but, especially in CentOS, this kind of practice should be done by
those who know very well what they are doing.  For one thing, they
cannot expect official support from CentOS.  For another, they may
well break their system.  If this sounds too harsh, I apologize.
However, I want to add that CentOS community members do help each
other for matters that may be outside the areas of official support by
the CentOS team.  The CentOS forums or mailing list are there for
everyone to participate in.

Akemi -- almost forgetting the fact this is the -docs mailing list...
___
CentOS-docs mailing list
CentOS-docs@centos.org
http://lists.centos.org/mailman/listinfo/centos-docs


[CentOS-announce] CESA-2008:0533 Important CentOS 5 i386 bind Update

2008-07-12 Thread Karanbir Singh

CentOS Errata and Security Advisory 2008:0533 Important

Upstream details at : https://rhn.redhat.com/errata/RHSA-2008-0533.html

The following updated files have been uploaded and are currently 
syncing to the mirrors: ( md5sum Filename ) 

i386:
b21b4ce7a86aba34be2a293350a79a42  bind-9.3.4-6.0.2.P1.el5_2.i386.rpm
e5bbae877fb1a9b508ca953cb1702363  bind-chroot-9.3.4-6.0.2.P1.el5_2.i386.rpm
f1b5d4f75a6c9adfc0f3154c91576658  bind-devel-9.3.4-6.0.2.P1.el5_2.i386.rpm
afc45322e4c1350c5b10d0ac656cc53e  
bind-libbind-devel-9.3.4-6.0.2.P1.el5_2.i386.rpm
3acab364668ce1a6ee5e10983a35e877  bind-libs-9.3.4-6.0.2.P1.el5_2.i386.rpm
30d366550a48f30a666eb3160d3a660a  bind-sdb-9.3.4-6.0.2.P1.el5_2.i386.rpm
b6600ed204c07ed684b7176b0a63d563  bind-utils-9.3.4-6.0.2.P1.el5_2.i386.rpm
03d39d4927b4eb3de2f17246097defe8  
caching-nameserver-9.3.4-6.0.2.P1.el5_2.i386.rpm

Source:
4fe2e56ebc53ac5e58d497117c300059  bind-9.3.4-6.0.2.P1.el5_2.src.rpm


-- 
Karanbir Singh
CentOS Project { http://www.centos.org/ }
irc: z00dax, [EMAIL PROTECTED]

___
CentOS-announce mailing list
CentOS-announce@centos.org
http://lists.centos.org/mailman/listinfo/centos-announce


[CentOS-announce] CESA-2008:0533 Important CentOS 5 x86_64 bind Update

2008-07-12 Thread Karanbir Singh

CentOS Errata and Security Advisory 2008:0533 Important

Upstream details at : https://rhn.redhat.com/errata/RHSA-2008-0533.html

The following updated files have been uploaded and are currently 
syncing to the mirrors: ( md5sum Filename ) 

x86_64:
8ce98ea9011cdf6edeb045530e50d65b  bind-9.3.4-6.0.2.P1.el5_2.x86_64.rpm
90dd43b806068a26544ecdc54d37a78a  bind-chroot-9.3.4-6.0.2.P1.el5_2.x86_64.rpm
c78acc6452141cb8113bb535ea1b88fc  bind-devel-9.3.4-6.0.2.P1.el5_2.i386.rpm
801eb10eb970c901b65965774862cd06  bind-devel-9.3.4-6.0.2.P1.el5_2.x86_64.rpm
cc731a37d6615140989af105185c734b  
bind-libbind-devel-9.3.4-6.0.2.P1.el5_2.i386.rpm
62f4a102991ee20d659723a58d9c88c6  
bind-libbind-devel-9.3.4-6.0.2.P1.el5_2.x86_64.rpm
dac13c403a35f9ede3fff86b3b3861cf  bind-libs-9.3.4-6.0.2.P1.el5_2.i386.rpm
7b4fc0b9c68e5a6154cb529e24e5f061  bind-libs-9.3.4-6.0.2.P1.el5_2.x86_64.rpm
afe442636646d8dc2f6d6d0a7b74cf27  bind-sdb-9.3.4-6.0.2.P1.el5_2.x86_64.rpm
83d74c041e84242c217a7221203e079b  bind-utils-9.3.4-6.0.2.P1.el5_2.x86_64.rpm
8072494205e1c7a180ef9a51b5e055ef  
caching-nameserver-9.3.4-6.0.2.P1.el5_2.x86_64.rpm

Source:
4fe2e56ebc53ac5e58d497117c300059  bind-9.3.4-6.0.2.P1.el5_2.src.rpm


-- 
Karanbir Singh
CentOS Project { http://www.centos.org/ }
irc: z00dax, [EMAIL PROTECTED]

___
CentOS-announce mailing list
CentOS-announce@centos.org
http://lists.centos.org/mailman/listinfo/centos-announce


[CentOS] looking for motherboard / temp / hdd monitor, other than lmsensors

2008-07-12 Thread Rudi Ahlers

Hi all

What other good temp / motherboard / fan speed / HDD / etc monitor can I 
use on CentOS 5.1 (or even 5.2 - still need to upgrade)? I know 
lmsensors works ok'ish, but it doesn't pickup my Gigabyte motherboard 
properly.


--

Kind Regards
Rudi Ahlers
CEO, SoftDux

Web:   http://www.SoftDux.com
Check out my technical blog, http://blog.softdux.com for Linux or other 
technical stuff, or visit http://www.WebHostingTalk.co.za for Web Hosting stuff


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


[CentOS] make rpm for kernel

2008-07-12 Thread Mag Gam
I am compiling a custom kernel. When I do a make oldconfig, and then
make rpm, I get a kernel rpm. The problem is I am not getting the
initrd image. Is this normal?

TIA
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] make rpm for kernel

2008-07-12 Thread Akemi Yagi
On Sat, Jul 12, 2008 at 5:37 AM, Mag Gam [EMAIL PROTECTED] wrote:
 I am compiling a custom kernel. When I do a make oldconfig, and then
 make rpm, I get a kernel rpm. The problem is I am not getting the
 initrd image. Is this normal?

Please try following this wiki:

http://wiki.centos.org/HowTos/Custom_Kernel

and do it CentOS way. :)

Akemi
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] make rpm for kernel

2008-07-12 Thread Mag Gam
Thanks

Actually, I am using Redhat distribution. I though CentOS would be
very similar to Redhat's kernel build.

TIA


On Sat, Jul 12, 2008 at 8:42 AM, Akemi Yagi [EMAIL PROTECTED] wrote:
 On Sat, Jul 12, 2008 at 5:37 AM, Mag Gam [EMAIL PROTECTED] wrote:
 I am compiling a custom kernel. When I do a make oldconfig, and then
 make rpm, I get a kernel rpm. The problem is I am not getting the
 initrd image. Is this normal?

 Please try following this wiki:

 http://wiki.centos.org/HowTos/Custom_Kernel

 and do it CentOS way. :)

 Akemi
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] make rpm for kernel

2008-07-12 Thread Akemi Yagi
On Sat, Jul 12, 2008 at 5:57 AM, Mag Gam [EMAIL PROTECTED] wrote:
 Thanks

 Actually, I am using Redhat distribution. I though CentOS would be
 very similar to Redhat's kernel build.

 TIA


 On Sat, Jul 12, 2008 at 8:42 AM, Akemi Yagi [EMAIL PROTECTED] wrote:
 On Sat, Jul 12, 2008 at 5:37 AM, Mag Gam [EMAIL PROTECTED] wrote:
 I am compiling a custom kernel. When I do a make oldconfig, and then
 make rpm, I get a kernel rpm. The problem is I am not getting the
 initrd image. Is this normal?

 Please try following this wiki:

 http://wiki.centos.org/HowTos/Custom_Kernel

 and do it CentOS way. :)

Or see post #2 of:

http://www.centos.org/modules/newbb/viewtopic.php?topic_id=14385forum=37

for some rpm hint.

Akemi
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] looking for motherboard / temp / hdd monitor, other than lmsensors

2008-07-12 Thread William L. Maltby

On Fri, 2008-07-11 at 13:40 +0200, Rudi Ahlers wrote:
 Hi all
 
 What other good temp / motherboard / fan speed / HDD / etc monitor can I 
 use on CentOS 5.1 (or even 5.2 - still need to upgrade)? I know 
 lmsensors works ok'ish, but it doesn't pickup my Gigabyte motherboard 
 properly.

I don't know if it's good or what you want, but I use gkrellm

$ rpm -qa|grep gkrel
gkrellm-2.2.7-0.el5.rf.i386

It's in the rpmforge repo.

It's highly configurable. I have a minimalist configuration. I've
attached a snapshot.

 

-- 
Bill
attachment: Gkrellm.png___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


RE: [CentOS] Latest samaba updates

2008-07-12 Thread John
 
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of David G. Mackay
Sent: Friday, July 11, 2008 8:24 AM
To: CentOS mailing list
Subject: RE: [CentOS] Latest samaba updates

On Thu, 2008-07-10 at 12:32 -0400, John wrote:
 Remember this. It is going to work when set to Permisive regardless!! 

Thanks, I will remember.  Now I just need to find out what it means by
read-only.  A find -perm 400 on the directory gives no hits, nor do most
other variations, like 444, etc.

Dave
---
See man ls. There is special SELinux Context Perms that it will show using
the special ls options. Otherwise with a regular ls /dev/null you want see.
I do remember seeing a few SE Linux know hows on the list maybe they can
chime in on this for you.

JohnStanley


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


RE: [CentOS] CUPS and system-config-printer question

2008-07-12 Thread John

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Al Sparks
Sent: Friday, July 11, 2008 7:17 PM
To: Centos List
Subject: [CentOS] CUPS and system-config-printer question

I was tasked with migrating a bunch of printer entries from one box to
another.

What I did is I got a list of printer names along with IP addresses and
using the CUPS
   lpadmin -p printer-name -E -v lpd://IP_ADDR/lp

I loaded them onto the new machine running RHEL 4.6.

Printer tests show that it works fine.  And if I use the CUPS web interface
at
http://localhost:631
Everything shows up.

But if I launch the RH (and presumably CentOS works the same way) 
system-config-printer
interface after adding the printers the CUPS way, nothing shows up in the RH
interface.

Probably not a big deal, but I have a colleague that will go nuts over this.
He's attached to his interfaces, but there was no way I was going to
manually enter those printers in that interface.

Where do I look to make that interface see the CUPS-only printers?
   === Al
--
Maybe copy the Printer Definitions

JohnStanley
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] command line configuration of DSL access

2008-07-12 Thread Alex White

Gergely Buday wrote:

Michel van Deventer wrote:


how can I configure a DSL access using the command line? The Red Hat
Certified Engineer Linux Study Guide does not describe this and I
could not find any description on the internet.

Maybe you can provide us with some more information, as like which
provider are you using in which country, what kind of hardware are you
using to connect to DSL (router/pci card?), which version of CentOS are
you using and maybe other info that might be of use ?


The case is that I used the GUI for my fedora box and now would like
to set the same to my centos server that would be the firewall. I use
eth0 to connect to my service provider's network. When I set up on
fedora I choose DSL and gave login name and password and worked out of
the box. I use CentOs 5.1 on my server.

I would like to use an editor better than the GUI to configure this.
My provider is Digi Tv in Hungary.

I wonder why do you need all this, isn't there a good description of
the config scripts somewhere?

- Gergely


We need all of that because how you set up your dsl depends upon how 
your provider provides you with access. pppoe or pppoa etc. etc.


At any rate, as mentioned by others you can use adsl-setup to configure 
your connection, or if you want to edit some text files manually you can 
have a peek in /etc/ppp/


The files of particular interest would be, pap-secrets, chap-secrets, 
and pppoe-server-options.


You may want to man adsl-setup and make note of the various references 
it makes.


HTH

Alex White
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Slow samba and Realtek card

2008-07-12 Thread Miguel Filho
On Thu, Jul 10, 2008 at 6:43 AM, Emmanuel Borlet
[EMAIL PROTECTED] wrote:
 Hello,

 We can see
 # lspci
 02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B
 PCI Express Gigabit Ethernet controller (rev 01)


Change the network card, period. I have a laptop with an R8168 and
even using kernel 2.6.25 the card just suddenly stops working. If I do
an rsync on a LAN, the card holds up to 300mb of data transfer and
stops working again. I always have to reboot. No errors, nothing. I
have seen many people complaining about this card on the IRC channel
too.

Miguel

-- 
http://osysadmin.blogspot.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


[CentOS] using new sysconfig file

2008-07-12 Thread Kai Schaetzl
I installed postgrey from rpmforge and wanted to use sysconfig to change 
options instead of overwriting the init file. However, I get a weird 
warning from postgrey. I'm not sure if this is a postgrey quirk or I use 
sysconfig the wrong way.

/etc/sysconfig/postgrey:
OPTIONS=--unix=/var/spool/postfix/postgrey/socket --delay=660

/etc/rc.d/init.d/postgrey:
OPTIONS=--unix=$SOCKET --delay=660

# Source an auxiliary options file if we have one, and pick up OPTIONS,
if [ -r /etc/sysconfig/$prog ]; then
. /etc/sysconfig/$prog
fi

Error:
 invalid for option delay (number expected)

(The quote marks are necessary!)

Using the additional option directly in the init file works just fine.

Kai

-- 
Kai Schätzl, Berlin, Germany
Get your web at Conactive Internet Services: http://www.conactive.com



___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


[CentOS] RE: CUPS and system-config-printer question

2008-07-12 Thread Timothy Murphy
John wrote:

 Where do I look to make that interface see the CUPS-only printers?

As a matter of interest, why do you want to use system-config-printer?

I've always found this completely useless,
while the CUPS web interface seems quite straightforward.


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/10/08, Rob Townley [EMAIL PROTECTED] wrote:
 why not use the dig command to query your isp dns system to see if
 they forward requests to opendns.  By the way, OpenDNS is a great way
 to help prevent phishing attacks.

 Lastly, you should use this opp to create a opendns signon, this will
 give you control over your dns request options.  You could block any
 domain via dns quikly.

Rob, I will go to opendns.com and see what I need to do, to use their
DNS Servers for web browsing, etc.,  temporarily, until I get Caching
DNS up and running. If my ISP is using them upstream, for their DNS
Service, it would be very easy for me to change the settings in my
ADSL modem, from the 2 DNS Servers at my ISP, to servers at
opendns.com and eliminate the delay at my local ISP.  I'd never heard
of opendns.com until I tried to connect to a secure server at irs.gov
about 10 days ago and got an SSL warning from Firefox, that the cert
belonged to opendns.com Another idea. Thanks! Lanny
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Ralph Angenendt
Lanny Marcus wrote:

[240kB png]

DON'T EVER DO THAT AGAIN.

You just sent out ~1GB of data.

As of now (as that already happened last week), the maximum message size
for this list is 50kB.

So people: Trim your mails :)

Ralph


pgpr6cPHknP7f.pgp
Description: PGP signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] pnm2ppa gone, in any repo?

2008-07-12 Thread Stephen John Smoogen
On Fri, Jul 11, 2008 at 5:48 PM, Ted Miller [EMAIL PROTECTED] wrote:
 Ted Miller wrote:

 I can't get my HP DeskJet 712C to print via cups.  I believe the reason is
 that according to
 http://www.redhat.com/archives/epel-devel-list/2007-May/msg2.html the
 pnm2ppa filter got dropped between RHEL 4 and RHEL 5, or between Fedora 6
 and RHEL 5, depending on how you look at it.

 Foomatic still generates the pnm2ppa.xml file, but there is no pnm2ppa
 binary filter installed, so any print attempt ends with an error message.

 I don't find that any repo I have installed has picked this up for x86_64
 architecture.  Do I need to add a repo?  Has this not been an issue for
 enough people that someone has made it available from a repo?

 Right now this has been a show-stopper on upgrading from Centos 4 to 5, as
 it is hard to use the workstation without a printer.  Any help appreciated.

 Since no repo seems to be interested in adding this printer driver, I
 installed the one included with Fedora 6 (rpm doesn't care what repo it came
 from, if you can get the URL right), and so far it is working.  At least I
 can start using my Centos 5 workstation.


Well I wouldn't say that 'no repo' is interested... its more that the
repos look for suggestions on their lists. Asking on this list can
sometimes do it, but most likely it will get missed.

 It would be nice if someone would add this to some repo somewhere, so it
 could be installed by yum.  Unfortunately, it only supports a half dozen
 printers, so I guess it is hard to get anyone motivated.


The first person to get motivated is yourself. Dag, atrpms, rpmrepo,
epel etc all look for people to do this... because well Dag, Axel, and
most of the EPEL people have other jobs and need more sleep than they
get.  If it fits your need 'get-er-done and help others'.

 Ted Miller
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos




-- 
Stephen J Smoogen. -- BSD/GNU/Linux
How far that little candle throws his beams! So shines a good deed
in a naughty world. = Shakespeare. The Merchant of Venice
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] RE: CUPS and system-config-printer question

2008-07-12 Thread Al Sparks



--- On Sat, 7/12/08, Timothy Murphy [EMAIL PROTECTED] wrote:

 From: Timothy Murphy [EMAIL PROTECTED]
 Subject: [CentOS] RE: CUPS and system-config-printer question
 To: centos@centos.org
 Date: Saturday, July 12, 2008, 9:32 AM
 John wrote:
 
  Where do I look to make that interface see
 the CUPS-only printers?
 
 As a matter of interest, why do you want to use
 system-config-printer?
 
 I've always found this completely useless,
 while the CUPS web interface seems quite straightforward.

I completely agree.  I'm just trying to make a colleague happy.

In fact, I am more of a CLI guy, and prefer to use xadmin to add, remove and 
configure printers.

But you're right, the CUPS interface is actually more straightforward
than the system-config-printer interface.
   === Al
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/12/08, Ralph Angenendt [EMAIL PROTECTED] wrote:
 Lanny Marcus wrote:
 [240kB png]
 DON'T EVER DO THAT AGAIN.
You just sent out ~1GB of data.
 As of now (as that already happened last week), the maximum message size
 for this list is 50kB.
 So people: Trim your mails :)

To: Ralph and everyone on the list: I apologize, sincerely. Bill sent
a .png attachment of the screen in his IPCop box and I sent mine back.
Please forgive me. I will not send an attachment like that to the list
again. Lanny
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


RE: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.comwhen browsing web?

2008-07-12 Thread Robert - elists

 
 As of now (as that already happened last week), the maximum message size
 for this list is 50kB.
 
 So people: Trim your mails :)
 
 Ralph

Ralph

Will the server notify us if we exceed that threshold ?

- rh

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/11/08, Scott Silva [EMAIL PROTECTED] wrote:
snip
  Question: Awhile ago, I got into the configuration settings for our
 ZTE ADSL Modem.
 For the change to me having my own Caching DNS Server, in the settings
 for the ADSL modem at this time, using the DNS servers at our ISP:
 Primary DNS Server   200.29.104.22
 Secondary DNS Server 200.29.96.22

 When I think I am ready to test the change I make to IPCop setting(s),
 should I set those to 0.0.0.0. so I can use my own DNS Server ? Or.
 leave those spaces blank? Or, leave them as they are now? Thank you,
 very much, for your time and help, which are greatly appreciated!

 It looks as if your ADSL modem is in NAT mode, so it is acting like a very
 simple router already. What settings does it actually have?

Scott: Which settings in the ADSL Modem are you interested in? There
are quite a few settings available in the web interface. If you tell
me which settings are of interest, I'll get them for you.

ADSL Port  Enable
Downstream Line Rate  2047
  Upstream Line Rate  507
  LAN IP Address  192.168.1.1
  Default Gateway  190.1.216.1
  Primary DNS Server  200.29.104.22
  Secondary DNS Server  200.29.96.22

ADSL line status

 Current adsl line status is as the below.

 Line Mode  ADSL2+   Line State  Show Time
  Line Up Time Duration  00:05:28:31   System Up Time  00:05:28:39
  Line Downstream Rate  2047   Line Upstream Rate  507
  Latency Type  Fast   Line Coding  Trellis On
  Noise Margin  31.6   Line Attenuation  19.5
  Output power  22.0   Attainable Line Rate  17628
  Line Up Count  1   Status  No Defect

If you note any problems in the quality of the line, the phone company
people were working in our subdivision a few weeks ago and they
detected a problem, with a long cable we have, underground, about 100
(?) meters in the street to their box. Apparently, 2 cables are
touching. They mentioned running a new cable in the air, instead of
underground. I was surprised that they found this problem, because at
the same time, on speedtest.net I  got a Download speed of 1780 from a
server in Orlando and our contract with our ISP is for 550, so I was
happy with the speed they were providing to us.

 I think you can leave those settings alone, as they only will be used if you
 point DNS settings at the modems ip address. If you set your IPcop box at
 127.0.0.1 it should seek out to the root servers by itself.

Cool. It sounds like all I need to do is change the one setting in the
IPCop box and if all goes well, my Caching DNS Server is up and
running. I will try it, ASAP, on our backup IPCop box. If I get up
*early* Sunday morning, I will try it then.

 As I posted earlier, you will have to poke around in the ipcop setup menu to
 get dhcp and custom DNS settings both working.

That's why I want to do it on the backup IPCop box. If it stops
working, my VIP users can continue using the IPCop box that works and
I don't have irate users.  :-) The IPCop box is our Production
server.  :-)

 I just played with one of my test vmware ipcop images and set it to dhcp on
 our internal network (which should simulate your natted connection through
 your adsl modem) for the red interface and I was able to dig +trace
 google.com
 with proper answers. So it is possible to get it working unless your ISP
 blocks DNS queries to anywhere else but their own servers.

Hoping they are not blocking those DNS queries or any other traffic.

I just SSH'd into the IPCop box:

[EMAIL PROTECTED]:~ # dig gmail.com

;  DiG 9.4.0  gmail.com
;; global options:  printcmd
;; Got answer:
;; -HEADER- opcode: QUERY, status: NOERROR, id: 29247
;; flags: qr rd ra; QUERY: 1, ANSWER: 3, AUTHORITY: 13, ADDITIONAL: 11

;; QUESTION SECTION:
;gmail.com. IN  A

;; ANSWER SECTION:
gmail.com.  27  IN  A   64.233.161.83
gmail.com.  27  IN  A   209.85.171.83
gmail.com.  27  IN  A   64.233.171.83

;; AUTHORITY SECTION:
com.152960  IN  NS  a.gtld-servers.net.
com.152960  IN  NS  f.gtld-servers.net.
com.152960  IN  NS  m.gtld-servers.net.
com.152960  IN  NS  b.gtld-servers.net.
com.152960  IN  NS  j.gtld-servers.net.
com.152960  IN  NS  g.gtld-servers.net.
com.152960  IN  NS  l.gtld-servers.net.
com.152960  IN  NS  i.gtld-servers.net.
com.152960  IN  NS  c.gtld-servers.net.
com.152960  IN  NS  e.gtld-servers.net.
com.152960  IN  NS  k.gtld-servers.net.
com.152960  IN  NS  h.gtld-servers.net.
com.152960  IN  NS  d.gtld-servers.net.

;; ADDITIONAL SECTION:
j.gtld-servers.net. 172736  IN  A   192.48.79.30
b.gtld-servers.net. 172737  IN

Re: [CentOS] using new sysconfig file

2008-07-12 Thread Ian Blackwell

Kai Schaetzl wrote:
I installed postgrey from rpmforge and wanted to use sysconfig to change 
options instead of overwriting the init file. However, I get a weird 
warning from postgrey. I'm not sure if this is a postgrey quirk or I use 
sysconfig the wrong way.


/etc/sysconfig/postgrey:
OPTIONS=--unix=/var/spool/postfix/postgrey/socket --delay=660

  

Hi Kai,

I got similar errors by corrupting my /etc/sysconfig/postgrey file, by 
putting text into the delay value - i.e. I replaced 660 with 66O.  I 
suggest you recreate the file (from scratch) to make sure you haven't 
got some odd binary data in their somehow (null's?).


BTW: Postgrey recommend a maximum delay of 300.  Is there a reason 
you're using 660?


Ian
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/11/08, Ian Blackwell [EMAIL PROTECTED] wrote:
 Scott Silva wrote:
 You would set the primary dns to 127.0.0.1 and if you want set the
 secondary
 dns to what your primary dns was set at. You might have to play with
 the
 options to have dhcp assigned red and still be able to set your
 nameserver
 settings.
 The ipcop boxes I have are all on static ip's, on either T1's or
 business
 class DSL, so the settings are a little different.

 For what it is worth, my IPCop box has the DNS values supplied by my ISP
 entered here instead of 127.0.0.1.  My dig +trace tests are all running
 fine.

You entered them there and you can dig +trace from there. That's
interesting.  I would like to discontinue using the DNS Servers at my
ISP, because: (a ) frequently slow (b) sometimes no DNS (c) the recent
problem where I get to opendns.com

 You can also create a backup using the web-interface.  The backup will
 be saved on your local machine and you can restore it from there if needed.

Thank you for reminding me about that! The IPCop box I am using now, I
backed up on 23 February. The Backup IPCop box, which I am going to
use to test this, will need to be updated and then I will backup,
before I try these changes.

snip
 For the change to me having my own Caching DNS Server, in the settings
 for the ADSL modem at this time, using the DNS servers at our ISP:
 Primary DNS Server   200.29.104.22
 Secondary DNS Server 200.29.96.22

 These are the number I would enter into the IPCop setup screen for DNS
 and Gateway.  My gateway value is the IP address of my ADSL modem.

Ian: Thank you for the information! Lanny
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/11/08, William L. Maltby [EMAIL PROTECTED] wrote:
snip
 Unless your IPCop box is assigned a dynamic IP address?

No. It has a Static IP address.

In that case,
 IIUC the DHCP server from the ISP/modem setup will provide the primary
 and secondary servers. I know they can be overridden if you massage the
 files though. But then if the ISP reassigns the servers' IP addresses
 you'll have to massage again - after the angst of it not working and you
 having to figure out that's what happened.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Ian Blackwell

Lanny Marcus wrote:

You entered them there and you can dig +trace from there. That's
interesting.  I would like to discontinue using the DNS Servers at my
ISP, because: (a ) frequently slow (b) sometimes no DNS (c) the recent
problem where I get to opendns.com
  
Generally your ISP's DNS should be quickest because they are closest.  
If you're not happy with them, google for public DNS and you'll find a 
plethora of publicly accessible DNS systems.

You can also create a backup using the web-interface.  The backup will
be saved on your local machine and you can restore it from there if needed.



Thank you for reminding me about that! The IPCop box I am using now, I
backed up on 23 February. The Backup IPCop box, which I am going to
use to test this, will need to be updated and then I will backup,
before I try these changes.
  
Don't forget to save the backup to your local system in case your IPCop 
box gets totally hosed.  You can then rebuild the IPCop system and 
restore the backup from your desktop.

snip
  
Ian: Thank you for the information! Lanny
  

You're welcome.

Ian
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/12/08, Ian Blackwell [EMAIL PROTECTED] wrote:
 Lanny Marcus wrote:
 You entered them there and you can dig +trace from there. That's
 interesting.  I would like to discontinue using the DNS Servers at my
 ISP, because: (a ) frequently slow (b) sometimes no DNS (c) the recent
 problem where I get to opendns.com

 Generally your ISP's DNS should be quickest because they are closest.
 If you're not happy with them, google for public DNS and you'll find a
 plethora of publicly accessible DNS systems.

We have been having problems with the DNS Servers at our ISP (the
phone company) for some time. Frequently, slow DNS or no DNS. I didn't
call them, until about 10 days ago, after I tried to get to a secure
server at irs.gov and I got a warning from Firefox, that the SSL
certificate belonged to opendns.com

I am going  to look at opendns.com first for public DNS.

 You can also create a backup using the web-interface.  The backup will
 be saved on your local machine and you can restore it from there if
 needed.

I will update the Backup IPCop box, and then backup, before I start
playing with the settings.

 Don't forget to save the backup to your local system in case your IPCop
 box gets totally hosed.  You can then rebuild the IPCop system and
 restore the backup from your desktop.

I will backup to floppy (it's an old box with a floppy drive) and also
to my Desktop.
 snip
 Hopefully, this will be a very quick and simple change and be up and running.

Good morning to you! It is 647 Saturday night here in Colombia.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Re: OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Ian Blackwell

Lanny Marcus wrote:

Good morning to you! It is 647 Saturday night here in Colombia.
___

  

9:34am Sunday morning here in Australia :)
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


[CentOS] 2.6.18-92.1.6.el5xen -- 8GB missing?

2008-07-12 Thread JJB
Hello,

Last night I upgraded a server to CentOS 5.2.  The server has 16GB of
RAM.  Now that it's running 2.6.18-92.1.6.el5xen only 8GB is reported to
exist.

# cat /proc/meminfo
MemTotal:  8818688 kB
MemFree:   3730124 kB
Buffers:202004 kB
Cached:4086788 kB
SwapCached:  0 kB
Active:1551480 kB
Inactive:  2958196 kB
HighTotal:   0 kB
HighFree:0 kB
LowTotal:  8818688 kB
LowFree:   3730124 kB
SwapTotal: 8388600 kB
SwapFree:  8388600 kB
Dirty:7348 kB
Writeback:   0 kB
AnonPages:  220788 kB
Mapped:  46948 kB
Slab:   137404 kB
PageTables:  29012 kB
NFS_Unstable:0 kB
Bounce:  0 kB
CommitLimit:  12797944 kB
Committed_AS:   756072 kB
VmallocTotal: 34359738367 kB
VmallocUsed:  5004 kB
VmallocChunk: 34359733235 kB

However, dmesg appears to report something else:

# dmesg | grep Memory
Memory: 15506152k/15899468k available (2417k kernel code, 384480k
reserved, 1350k data, 176k init)

I looked in the messages file for complaints and didn't see anything
that strikes me as unusual.

How much memory is available?  Is this a known issue?


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] 2.6.18-92.1.6.el5xen -- 8GB missing?

2008-07-12 Thread Victor Padro
On Sat, Jul 12, 2008 at 7:11 PM, JJB [EMAIL PROTECTED] wrote:

 Hello,

 Last night I upgraded a server to CentOS 5.2.  The server has 16GB of
 RAM.  Now that it's running 2.6.18-92.1.6.el5xen only 8GB is reported to
 exist.

 # cat /proc/meminfo
 MemTotal:  8818688 kB
 MemFree:   3730124 kB
 Buffers:202004 kB
 Cached:4086788 kB
 SwapCached:  0 kB
 Active:1551480 kB
 Inactive:  2958196 kB
 HighTotal:   0 kB
 HighFree:0 kB
 LowTotal:  8818688 kB
 LowFree:   3730124 kB
 SwapTotal: 8388600 kB
 SwapFree:  8388600 kB
 Dirty:7348 kB
 Writeback:   0 kB
 AnonPages:  220788 kB
 Mapped:  46948 kB
 Slab:   137404 kB
 PageTables:  29012 kB
 NFS_Unstable:0 kB
 Bounce:  0 kB
 CommitLimit:  12797944 kB
 Committed_AS:   756072 kB
 VmallocTotal: 34359738367 kB
 VmallocUsed:  5004 kB
 VmallocChunk: 34359733235 kB

 However, dmesg appears to report something else:

 # dmesg | grep Memory
 Memory: 15506152k/15899468k available (2417k kernel code, 384480k
 reserved, 1350k data, 176k init)

 I looked in the messages file for complaints and didn't see anything
 that strikes me as unusual.

 How much memory is available?  Is this a known issue?


 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos


I think there was a topic like this about a month ago, please search the
list archieves.

Cheers.

-- 
It is human nature to think wisely and act in an absurd fashion.

Todo el desorden del mundo proviene de las profesiones mal o mediocremente
servidas
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/10/08, Rob Townley [EMAIL PROTECTED] wrote:
snip
 Lastly, you should use this opp to create a opendns signon, this will
 give you control over your dns request options.  You could block any
 domain via dns quikly.

Rob: I just changed the DNS settings in the ADSL Modem to use the DNS
servers at opendns.com and not the DNS servers at our ISP in Cali. The
opendns.com servers will probably work a lot better. I still want to
get DNS Caching locally and with luck, possibly I will have that
running tomorrow. This should speed up our surfing, a lot. Lanny
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] OT: anything in CentOS 5.2 that uses opendns.com when browsing web?

2008-07-12 Thread Lanny Marcus
On 7/12/08, Lanny Marcus [EMAIL PROTECTED] wrote:
 On 7/10/08, Rob Townley [EMAIL PROTECTED] wrote:
 snip
 Lastly, you should use this opp to create a opendns signon, this will
 give you control over your dns request options.  You could block any
 domain via dns quikly.

 Rob: I just changed the DNS settings in the ADSL Modem to use the DNS
 servers at opendns.com and not the DNS servers at our ISP in Cali. The
 opendns.com servers will probably work a lot better. I still want to
 get DNS Caching locally and with luck, possibly I will have that
 running tomorrow. This should speed up our surfing, a lot. Lanny

Either our ISP (EMCALI) is using opendns.com intentionally, or, their
DNS servers have been corrupted. Probably, they are using it
intentionally and their Customer Support is not aware of that. In the
past 10 days, frequently, I ended up at the Open DNS Guide, when DNS
wasn't resolving.

Make life easier for you and your users.
OpenDNS Guide

The OpenDNS Guide is the page your users see when they go to a website
that doesn't exist or isn't resolving. We provide search results and
offer suggestions to help your users get back on their way. This
includes the use of our industry-leading domain spellchecking service
to save them time and make them more productive.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Is there a FC8 update repo out there?

2008-07-12 Thread Robert Moskowitz

Johnny Hughes wrote:

Robert Moskowitz wrote:

Ray Van Dolson wrote:

How do I find out if Python 2.4.3 contains abi?



I'm not sure what you're asking.  Python's ABI version would be 2.4.3,
but if you're referring to a module 'abi', I don't think one exists?

You can always fire up python and type 'import modulename' to see if
that module exists.
Well no such module exists in either 2.4.3 or 2.5.1, so what is being 
complained about in the yum install below? Only that python v2.5 is 
needed? (I have been trying to get my colleague to either set me up 
to compile the source or for him to build a Centos host to do it 
himself).


yum localinstall hip*
Loading fastestmirror plugin
Loading priorities plugin
Setting up Local Package Process
Loading mirror speeds from cached hostfile
* rpmforge: fr2.rpmfind.net
* base: medon.htt-consult.com
* updates: medon.htt-consult.com
* addons: mirror.steadfast.net
* extras: pubmirrors.reflected.net
272 packages excluded due to repository priority protections
Examining hipl-agent-1.0.4-1.i386.rpm: hipl-agent - 1.0.4-1.i386
Examining hipl-daemon-1.0.4-1.i386.rpm: hipl-daemon - 1.0.4-1.i386
Examining hipl-debuginfo-1.0.4-1.i386.rpm: hipl-debuginfo - 1.0.4-1.i386
Examining hipl-doc-1.0.4-1.i386.rpm: hipl-doc - 1.0.4-1.i386
Examining hipl-firewall-1.0.4-1.i386.rpm: hipl-firewall - 1.0.4-1.i386
Examining hipl-lib-1.0.4-1.i386.rpm: hipl-lib - 1.0.4-1.i386
Examining hipl-test-1.0.4-1.i386.rpm: hipl-test - 1.0.4-1.i386
Examining hipl-tools-1.0.4-1.i386.rpm: hipl-tools - 1.0.4-1.i386
Marking hipl-agent-1.0.4-1.i386.rpm to be installed
Marking hipl-daemon-1.0.4-1.i386.rpm to be installed
Marking hipl-debuginfo-1.0.4-1.i386.rpm to be installed
Marking hipl-doc-1.0.4-1.i386.rpm to be installed
Marking hipl-firewall-1.0.4-1.i386.rpm to be installed
Marking hipl-lib-1.0.4-1.i386.rpm to be installed
Marking hipl-test-1.0.4-1.i386.rpm to be installed
Marking hipl-tools-1.0.4-1.i386.rpm to be installed
Resolving Dependencies
-- Running transaction check
--- Package hipl-doc.i386 0:1.0.4-1 set to be updated
--- Package hipl-tools.i386 0:1.0.4-1 set to be updated
-- Processing Dependency: python(abi) = 2.5 for package: hipl-tools
--- Package hipl-firewall.i386 0:1.0.4-1 set to be updated
--- Package hipl-lib.i386 0:1.0.4-1 set to be updated
-- Processing Dependency: python(abi) = 2.5 for package: hipl-lib
--- Package hipl-agent.i386 0:1.0.4-1 set to be updated
--- Package hipl-debuginfo.i386 0:1.0.4-1 set to be updated
--- Package hipl-test.i386 0:1.0.4-1 set to be updated
--- Package hipl-daemon.i386 0:1.0.4-1 set to be updated
-- Finished Dependency Resolution
Error: Missing Dependency: python(abi) = 2.5 is needed by package 
hipl-lib
Error: Missing Dependency: python(abi) = 2.5 is needed by package 
hipl-tools



BTW, if you are interested, this is the hipl implementation of the 
HIP protocol.


It looks like those RPMS are not built for CentOS-5 ... but were 
instead built against a version of linux with a  newer version of 
python installed.
They were built against FC8.  My colleague has given me source compile 
instructions:


you can get the sources either via version control or the nightly tarball:

wget http://hipl.hiit.fi/hipl/hipl.tar.gz

you can build on CentOS by executing:

tar xvzf hipl.tar.gz
cd hipl--main--2.6
./autogen.sh
./configure
make rpm

(make install works too, but installs to /usr/local by default)

Software requirements are listed here:

http://infrahip.hiit.fi/hipl/manual/ch02.html


You can't install FC8 packages (for example) directly on CentOS-5 ... 
all the shared libraries are different, so the required things will 
not exist.
So I have learned.  I really expected this, but my colleague was away 
and pointed me to his FC8 build.


depending on the item in question, you might be able to BUILD it from 
the sources on CentOS-5 and then use it ... I have no idea what hipl 
is or does.

RFCs 4423 and 5201-5207

Read the charter at:  http://www.ietf.org/html.charters/hip-charter.html

Note the author of this protocol...




___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
  

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] Is there a FC8 update repo out there?

2008-07-12 Thread Robert Moskowitz

Ray Van Dolson wrote:

Robert Moskowitz wrote:


yum localinstall hip*
Loading fastestmirror plugin
Loading priorities plugin
Setting up Local Package Process
Loading mirror speeds from cached hostfile
* rpmforge: fr2.rpmfind.net
* base: medon.htt-consult.com
* updates: medon.htt-consult.com
* addons: mirror.steadfast.net
* extras: pubmirrors.reflected.net
272 packages excluded due to repository priority protections
Examining hipl-agent-1.0.4-1.i386.rpm: hipl-agent - 1.0.4-1.i386
  


Yeah, what is this hipl package?

  http://infrahip.hiit.fi/index.php?index=home

Is that it?  

Yes.

Although yours seems to be newer and has additional
packages.  Do they provide a src rpm which you could then just rebuild
against CentOS 5?
  

See my prior note. Time to do some compiling on my own.

I can design major changes to IP, but code it?

No way :)

And of course I had my hands in IPng. Now I get to actually implement it!

Ray
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

  

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] 2.6.18-92.1.6.el5xen -- 8GB missing?

2008-07-12 Thread Ruslan Sivak

Victor Padro wrote:



On Sat, Jul 12, 2008 at 7:11 PM, JJB [EMAIL PROTECTED] 
mailto:[EMAIL PROTECTED] wrote:


Hello,

Last night I upgraded a server to CentOS 5.2.  The server has 16GB of
RAM.  Now that it's running 2.6.18-92.1.6.el5xen only 8GB is
reported to
exist.
...snip...

How much memory is available?  Is this a known issue?


I think there was a topic like this about a month ago, please search 
the list archieves.


Cheers.


In case he didn't find it, I was the one that posted the original 
issue.  Turns out that Xen allocates memory to domU's and doesn't 
release it back to dom0 even if you stop the VM's.  Top only reports the 
memory available to dom0.  The xen tools will show the proper amount of 
memory. 


Russ
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


[CentOS] screen command

2008-07-12 Thread Ed Donahue
Anyone know which rpm give you the screen command?

Or tell me how to figure this out on my own :-)

Ed
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] screen command

2008-07-12 Thread Ray Van Dolson
On Sun, Jul 13, 2008 at 01:46:20AM -0400, Ed Donahue wrote:
 Anyone know which rpm give you the screen command?
 
 Or tell me how to figure this out on my own :-)

screen

yum provides /usr/bin/screen

Ray
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Re: [CentOS] screen command

2008-07-12 Thread John R Pierce

Ed Donahue wrote:

Anyone know which rpm give you the screen command?

Or tell me how to figure this out on my own :-)


# rpm -qf `which screen`
screen-4.0.3-1.el5


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos