date:20080821

 Fail2ban keeps up blacklists of ip:s that have failed authentication too
 many times. What is too many, and the duration of blacklisting can be
 configured easily in /etc/fail2ban.conf.

Kenneth Porter [EMAIL PROTECTED] escribio (20.8.2008 22:55)
 The installation is easy, but the configuration looks daunting. There's a
 bunch of jails and it looks like I need to study fail2ban's model to
 understand what those do.

What do you mean by jail? I didn't find that word in my fail2ban.conf
(dated 2005.06.21). I needed to change the default conf only very slightly
(mainly time constants, on/off switches for tasks, and or log file paths).

- Jussi

--
Jussi Hirvi  *  Green Spot
Topeliuksenkatu 15 C  *  00250 Helsinki  *  Finland
Tel.  fax +358 9 493 981  *  SMS +358 40 771 2098
[EMAIL PROTECTED]  *  http://www.greenspot.fi


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

Re: [CentOS] Re: Is there a way to save the routing table permanently?

2008-08-21 Thread Rob Townley

On Wed, Aug 20, 2008 at 7:52 AM, Filipe Brandenburger
[EMAIL PROTECTED]wrote:

 Hi,

 On Wed, Aug 20, 2008 at 01:22, Rob Townley [EMAIL PROTECTED] wrote:
  two NICs each that would use two different gateways,

 If you are configuring default gateways on each interface, you are
 probably doing something wrong.

 The only reason why you would want to do that is to balance your
 outgoing traffic between the two NICs, and this is better accomplished
 with bonding interfaces.


Bonding would defeat my purpose for this.  My registrar requires two dns
servers on two different IP addresses, but i only wanted to use one machine
for now.  The machine has two NICs that connect out through the same cable
modem.  One behind a soho firewall, one direct.



  ifup would not process either route.ethX nor ethX.route
  - at least not enough for it to show in route.  Had to set
  the routes in /etc/rc.local.

 I fail to see why SELinux would make any difference on that. Can you
 describe your issue better? What is the configuration you tried to set
 up, and why didn't it work? What version of CentOS are you using, 4 or
 5? What is in /var/log/messages and /var/log/audit/audit.log when you
 try to bring the interface up?


The interfaces would come up, the point was that that system-config-network
would not keep the static information for the two NICs after a reboot.  So
when the machine was rebooted, some part of IP, SM, GW, NS disappeared or
reverted back to DHCP even though it was explicitly set to static.  I was
using CentOS 5.0 / 5.1 when i had most problems.   No entries would have
appeared in the logs.  I modified ifup-route to add logging to it directly
and believe it was never called.  Maybe i will have to upgrade the machine
so i can run both the TUI and GUI more and monitor all files changed by
both.   Further, i turned off NetworkManager to get much further in keeping
static ip setup.

Couldn't tell you much about why seLinux may have caused problems except
that maybe there were mdac labels on files that broke some part of
system-config-network keeping static routes and dns servers.  I just
remember that uninstalling seLinux got me much much further on a different
machine when it came to static settings for multiple NICs.  Much Further.





 Regards,
 Filipe
 ___
 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

[CentOS] Disk quotas for Sendmail

Hi,

CentOS 5.x, Sendmail 8.13.8.

I would like to start using disk quotas on my Sendmail server, and optimally
they should be configurable by user or group. How can this be done? (I tried
to google, but no success.)

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

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

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Rudi Ahlers

On 8/21/08, Jussi Hirvi [EMAIL PROTECTED] wrote:

 Hi,

 CentOS 5.x, Sendmail 8.13.8.

 I would like to start using disk quotas on my Sendmail server, and
 optimally
 they should be configurable by user or group. How can this be done? (I
 tried
 to google, but no success.)

 - Jussi


If each user's mail is stored in his homefolder, then you just need to
enable user quotas :) man quota  man mtab for the options

-- 

Kind Regards
Rudi Ahlers
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] perl

Am trying to install perl module File::Find, but not able it gave the
following

cpan[1] install File::Find
CPAN: Storable loaded ok (v2.15)
Going to read /root/.cpan/Metadata
  Database was generated on Thu, 21 Aug 2008 02:03:21 GMT
Running install for module 'File::Find'
The most recent version 1.12 of the module File::Find
is part of the perl-5.10.0 distribution. To install that, you need to run
  force install File::Find   --or--
  install R/RG/RGARCIA/perl-5.10.0.tar.gz
CPAN: Time::HiRes loaded ok (v1.9715)
Running make test
  Can't test without successful make
Running make install
  Make had returned bad status, install seems impossible
Failed during this command:
 RGARCIA/perl-5.10.0.tar.gz   : make NO isa perl


cpan[2] quit
Lockfile removed.
[EMAIL PROTECTED] ftpsync-1.2.33]# perl -version

This is perl, v5.8.8 built for x86_64-linux-thread-multi

Copyright 1987-2006, Larry Wall

Perl may be copied only under the terms of either the Artistic License or
the
GNU General Public License, which may be found in the Perl 5 source kit.

Complete documentation for Perl, including FAQ lists, should be found on
this system using man perl or perldoc perl.  If you have access to the
Internet, point your browser at http://www.perl.org/, the Perl Home Page.


-- 
Your search - madunix - did not match any documents.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] General Linux query

2008-08-21 Thread lingu

HI,

  This is more general  linux  related topic ,  i want to know few  basic
things.

1) How file systeem get corrupted on linux?

2)  why,when and how fsck to be run without lossing data?

3)  Is there is any other tool  other than fsck available for file system
check
.
4)  what are all the precaution to be made to prevent file system
corruption.

5)  why  when the system will get hanged  and what are all the possible
reasons?

 If any one provide me suitable links providing all of the  above
information  in detail it will be quite usefull.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Rudi Ahlers

On 8/21/08, Jussi Hirvi [EMAIL PROTECTED] wrote:

  On 8/21/08, Jussi Hirvi [EMAIL PROTECTED] wrote:

  CentOS 5.x, Sendmail 8.13.8. http://8.13.8.

  I would like to start using disk quotas on my Sendmail server, and
 optimally
  they should be configurable by user or group. How can this be done? (I
 tried
  to google, but no success.)


 Rudi Ahlers ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 13:55):

  If each user's mail is stored in his homefolder, then you just need to
 enable
  user quotas :) man quota  man mtab for the options


 Hei,

 The mail is not stored in home folders but in /var/spool/mail. I would
 rather keep it that way - and I'm not sure if storing mail in home folders
 would work for procmail (Sendmail's local mailer), dovecot (pop server)
 etc.


 - Jussi



mm, then I don't know, sorry. Maybe someone else might shed some light on
the subject then.

-- 

Kind Regards
Rudi Ahlers
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Disk quotas for Sendmail

Rudi Ahlers ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 14:14):
 mm, then I don't know, sorry. Maybe someone else might shed some light on the
 subject then. 

Ok, thanks Rudi. If someone has a Sendmail system where email is stored in
the users' home folders, please chime in... Does that work ok?

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

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

Re: [CentOS] General Linux query

2008-08-21 Thread William L. Maltby


On Thu, 2008-08-21 at 16:43 +0530, lingu wrote:
 HI,
 
   This is more general  linux  related topic ,  i want to know few
 basic things.
 
 1) How file systeem get corrupted on linux?
 
 2)  why,when and how fsck to be run without lossing data?
 
 3)  Is there is any other tool  other than fsck available for file
 system check
 .
 4)  what are all the precaution to be made to prevent file system
 corruption.
 
 5)  why  when the system will get hanged  and what are all the
 possible reasons?
 
  If any one provide me suitable links providing all of the  above
 information  in detail it will be quite usefull.

http://www.google.com

 snip sig stuff

-- 
Bill

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

Re: [CentOS] General Linux query

2008-08-21 Thread Brett Serkez

I can only comment from my experience, which is primarily ext2 and ext3.

 1) How file systeem get corrupted on linux?

I've almost never seen corruption in a Linux file system, the primary
reason is usually a hardware issue, the secondary reason (by far) is
buggy code.

 2)  why,when and how fsck to be run without lossing data?

fsck can never guarantee data will not be lost, it does the best it
can, which is usually pretty good.  Normally fsck is run in single
user mode.  If you want to run it manually, you'd bring the system to
single user mode (init 1) or it would run automatically during boot.
There is a counter in the file system as to the last time fsck was
run, during boot if this exceeds a certain vaule, fsck is run.  There
is also a clean shutdown bit which is set during a normal
shutdown/dismount, if this is not se,t which indicates a possible
system crash, fsck is run during system boot.

 4)  what are all the precaution to be made to prevent file system
 corruption.

Run a production distribution such as Redhat/CentOS vs. more of a
bleeding edge distribution such as Fedora.  More than this, assume
there will be data loss or corruption at some point and take
precautions such as running RAID 1, backups and use a UPS to avoid
system crashes or hardware issues in the face of brown or black outs.
Lastly choose a journaling  file system such as ext3, use ext2 for
partitions were performance over consistency is more important.

 5)  why  when the system will get hanged  and what are all the possible
 reasons?

Usually hardware issues, particularly memory and hard drives.  Most
distributions come with memtest86 so that the hardware's memory can be
thoroughly tested prior to installation of the operating system.

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

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Alexander Dalloz

Jussi Hirvi schrieb:
 On 8/21/08, Jussi Hirvi [EMAIL PROTECTED] wrote:
 CentOS 5.x, Sendmail 8.13.8. http://8.13.8.
 I would like to start using disk quotas on my Sendmail server, and optimally
 they should be configurable by user or group. How can this be done? (I tried
 to google, but no success.)
 
 Rudi Ahlers ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 13:55):
 If each user's mail is stored in his homefolder, then you just need to enable
 user quotas :) man quota  man mtab for the options
 
 Hei,
 
 The mail is not stored in home folders but in /var/spool/mail. I would
 rather keep it that way - and I'm not sure if storing mail in home folders
 would work for procmail (Sendmail's local mailer), dovecot (pop server) etc.

http://freshmeat.net/projects/milterquota/

Other milters here:

http://www.sendmail.com/sm/partners/milter_partners/open_source_milter_partners/

I.e. MIMEdefang can rule quota decisions too, if you know a bit of perl.

An yes, using procmail as your LDA you can store user's mail spool in
their $HOME, either as mbox or maildir.

 - Jussi

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

[CentOS] Per-user limits for port access

Centos 5.x again...

Is there a way to limit port access on a per-user basis? My practical goal
is to limit, who is allowed to use SMTP AUTH in Sendmail. Any possible
solutions appreciated...

Obviously, limiting by ip will not work, since the very purpose of SMTP AUTH
is to enable mail relaying for trusted users from *anywhere*.

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

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

Re: [CentOS] General Linux query

2008-08-21 Thread Rainer Duffner


I can only answer some...



lingu schrieb:

HI,

  This is more general  linux  related topic ,  i want to know few  
basic things.


1) How file systeem get corrupted on linux?



Hardware-problems. Loss of power. Bugs in FS ;-)



2)  why,when and how fsck to be run without lossing data?

3)  Is there is any other tool  other than fsck available for file 
system check


Nope, you can just change filesystems.
(and platforms...)


.
4)  what are all the precaution to be made to prevent file system 
corruption.


UPS.
BBWC.



5)  why  when the system will get hanged  and what are all the 
possible reasons?



What do you mean by will get hanged?



 If any one provide me suitable links providing all of the  above 
information  in detail it will be quite usefull.



Is this your homework?
;-)




Rainer

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

Re: [CentOS] General Linux query

2008-08-21 Thread Warren Young


lingu wrote:


1) How file systeem get corrupted on linux?


The same way any file system gets corrupted: data gets damaged or lost 
on its way to the physical media.



2)  why,when and how fsck to be run without lossing data?


The purpose of fsck is to bring the file system back into a consistent 
state so you can continue to use it, not to guarantee to recover your 
data.  If you need guaranteed data recovery, ship the drive to a data 
recovery house.


3)  Is there is any other tool  other than fsck available for file 
system check


I'm not aware of any.  There may be other tools that incorporate fsck 
functionality, but nothing that's designed to do a better job, somehow.


4)  what are all the precaution to be made to prevent file system 
corruption.


1. Put the computer on a UPS.

2. Configure nut (http://networkupstools.org/) to bring the system down 
safely well before the UPS's battery quits.  You don't want to 
fully-discharge a lead-acid battery, which is what UPSes use.  Shut down 
before it drops below about 20%.


3. Use some form of redundant RAID.  (Not RAID-0!)

4. Run the RAID system's consistency check function periodically.

5. If the system's RAID card offers a battery backup option, get one.

6. If the system is experiencing kernel panics, fix the problem that's 
causing it.  Kernel panics are almost always due to broken hardware. 
Replace the broken hardware.  Very rarely, they may be due to a bad 
driver or a kernel bug, but when you hear hoof beats, think horses, not 
zebras.


7. Back up your system.

8. BACK UP YOUR SYSTEM.

9. BACK! UP! YOUR! SYSTEM!!!  (And then, back it up again.)

5)  why  when the system will get hanged  and what are all the possible 
reasons?


There are an infinite number of answers to this question.

You can configure Linux to dump the kernel's memory state and other 
debugging information to disk in case of a crash.  Do that, and convince 
someone to analyze the information for you.


P.S. Don't you think you could have thought up a better title for this 
thread?  Even fsck would have been better.

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

Re: [CentOS] Per-user limits for port access

2008-08-21 Thread Alexander Dalloz

Jussi Hirvi schrieb:
 Centos 5.x again...
 
 Is there a way to limit port access on a per-user basis? My practical goal
 is to limit, who is allowed to use SMTP AUTH in Sendmail. Any possible
 solutions appreciated...
 
 Obviously, limiting by ip will not work, since the very purpose of SMTP AUTH
 is to enable mail relaying for trusted users from *anywhere*.
 
 - Jussi

Pretty simple. If your intention is to define SMTP AUTH usage
independent from shell access (i.e. by SSH) - side node: really think
about whether it is a good idea to have user logins on your mail server
- you should use sasldb as your SASL backend. You have to create the
sasldb manually and this way you control who is permitted to relay remotely.

Alexander

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

Re: [CentOS] perl

2008-08-21 Thread John Thomas


Mad Unix wrote:
Am trying to install perl module File::Find, but not able it gave the 
following


cpan[1] install File::Find

snip

I made the mistake of trying to use cpan when I was young.  I am now 
older and wiser and have read things like:

http://wiki.centos.org/PackageManagement/SourceInstalls?highlight=(cpan)


--
Sincerely,
John Thomas
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Disk quotas for Sendmail

Alexander Dalloz ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 14:31):

 http://freshmeat.net/projects/milterquota/
 
 Other milters here:
 
 http://www.sendmail.com/sm/partners/milter_partners/open_source_milter_partner
 s/
 
 I.e. MIMEdefang can rule quota decisions too, if you know a bit of perl.

Thanks Alex!

Milterquota looks just perfect for me. It seems to be much better than using
disk quotas in that emails from the postmaster can override the quota limit
- that is, the user can be informed about the quota override!

I will give milterquota a try.

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

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

Re: [CentOS] Per-user limits for port access

Alexander Dalloz ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 14:46):
 Pretty simple. If your intention is to define SMTP AUTH usage
 independent from shell access (i.e. by SSH)

Yes, though ssh access is not used by many users on this mail server. The
important thing is to allow for *all users* relaying from port 25 without
AUTH (of course as limited by the access list etc.).

 - side node: really think
 about whether it is a good idea to have user logins on your mail server

Do you mean ssh logins? I could actually consider limiting that. The easiest
way would probably be to edit (or remove) the shell declatarions in
/etc/passwd.

 - you should use sasldb as your SASL backend. You have to create the
 sasldb manually and this way you control who is permitted to relay remotely.

Ok, thanks I will take a look at this soon.

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

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

Re: [CentOS] Disk quotas for Sendmail

Alexander Dalloz ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 14:31):
 http://freshmeat.net/projects/milterquota/

I didn't find milterquota with yum so I guess I have to build it... The
Makefile asks me to locate a file that I cannot find:

 # change this to the object directory in the sendmail build
 treeSENDMAIL_OBJ=/usr/src/packages/sendmail-8.12.9/obj.Linux.2.4.21.i586

I just installed sendmail source, but still I cannot find a obj.Linux
anywhere. (I searched with the command locate obj.Linux.)

How should I proceed?

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

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

Re: [CentOS] Disk quotas for Sendmail

Jussi Hirvi ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 15:56):
 I didn't find milterquota with yum so I guess I have to build it... The
 Makefile asks me to locate a file that I cannot find:
 
 # change this to the object directory in the sendmail build
 treeSENDMAIL_OBJ=/usr/src/packages/sendmail-8.12.9/obj.Linux.2.4.21.i586

Maybe I should add that my Sendmail is 8.13.8, not 8.12.9, as mentioned in
that Makefile sample. Maybe file names in the Sm build have changed.

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

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

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Joost Waversveld

Try to find the file through the find command, something like  find 
/usr -t f -name 'obj.Linux*' 


The command locate works with an database, which will be updated just 
once a day (If you enable it!), so that will not find any just 
installed files...


regards,

Joost Waversveld

Jussi Hirvi wrote:

Jussi Hirvi ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 15:56):
  

I didn't find milterquota with yum so I guess I have to build it... The
Makefile asks me to locate a file that I cannot find:



# change this to the object directory in the sendmail build
treeSENDMAIL_OBJ=/usr/src/packages/sendmail-8.12.9/obj.Linux.2.4.21.i586
  


Maybe I should add that my Sendmail is 8.13.8, not 8.12.9, as mentioned in
that Makefile sample. Maybe file names in the Sm build have changed.

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

___
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

[CentOS] Support policy CentOS 5

2008-08-21 Thread Patrick Derwael

Hi list

Can someone tell me what the exact CentOS policy for end-of-life is for 5.0?

Let me explain: I have been using RHEL for a number of years and had to
switch to Fedora for installing some packages only supported under F6.

As I find Fedora's policy for release life duration too stringent (and I
do no longer need the packages that made me choose F6), I'm considering
switching over to CentOS5.0
I know, 5.0 is not the most recent release, but that one is supported by
Parallel Pro (new package I need)

Thanks

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

[CentOS] YumRepo Warning: not using ftp, http[s], or file for repos, skipping - 5 is not a valid release or hasnt been released yet

2008-08-21 Thread Rudi Ahlers

Hi all

Does anyone know what causes this error?
I have setup a local CentOS repository, with mrepo, and can succesfully use
it for updates  installation of just about anything. Yet, this error comes
up.

Here's my /etc/yum.conf:

# PUT YOUR REPOS HERE OR IN separate files named file.repo
# in /etc/yum.repos.d

[base-local]
name=CentOS-$releasever - Base
#mirrorlist=
http://mirrorlist.centos.org/?release=$releaseverarch=$basearchrepo=os
baseurl=http://192.168.10.10/repo/centos5/$basearch/os
gpgcheck=1
enabled=1
gpgkey=http://mirror.centos.org/centos/RPM-GPG-KEY-CentOS-5

[updates-local]
name=CentOS-$releasever - Updates
#mirrorlist=
http://mirrorlist.centos.org/?release=$releaseverarch=$basearchrepo=updates
baseurl=http://192.168.10.10/repo/centos5/$basearch/updates
gpgcheck=1
enabled=1
gpgkey=http://mirror.centos.org/centos/RPM-GPG-KEY-CentOS-5

[addons-local]
name=CentOS-$releasever - Addons
#mirrorlist=
http://mirrorlist.centos.org/?release=$releaseverarch=$basearchrepo=addons
baseurl=http://192.168.10.10/repo/centos5/$basearch/addons
gpgcheck=1
enabled=1
gpgkey=http://mirror.centos.org/centos/RPM-GPG-KEY-CentOS-5

[extras-local]
name=CentOS-$releasever - Addons
#mirrorlist=
http://mirrorlist.centos.org/?release=$releaseverarch=$basearchrepo=addons
baseurl=http://192.168.10.10/repo/centos5/$basearch/extras
gpgcheck=1
enabled=1
gpgkey=http://mirror.centos.org/centos/RPM-GPG-KEY-CentOS-5


And when I browse to http://192.168.10.10/repo/centos5 - it looks similar to
the online CentOS repositories, but only with the x86 folder.

-- 

Kind Regards
Rudi Ahlers
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Alexander Dalloz

Jussi Hirvi schrieb:
 Jussi Hirvi ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 15:56):
 I didn't find milterquota with yum so I guess I have to build it... The
 Makefile asks me to locate a file that I cannot find:

 # change this to the object directory in the sendmail build
 treeSENDMAIL_OBJ=/usr/src/packages/sendmail-8.12.9/obj.Linux.2.4.21.i586
 
 Maybe I should add that my Sendmail is 8.13.8, not 8.12.9, as mentioned in
 that Makefile sample. Maybe file names in the Sm build have changed.
 
 - Jussi

The Sendmail objects directory is the one where the libmilter and libsm
subdirectories are inside. There should normally be no problem wit SM
8.12 vs SM 8.13 regarding the milter code.

What I would do is following: change this line in the Makefile

$(CC) $(OPTS) $(DEFINES) $(EFENCE) -o milterquota milterquota.c
$(SENDMAIL_OBJ)/libmilter/libmilter.a $(SENDMAIL_OBJ)/libsm/libsm.a -pthread

replace $(SENDMAIL_OBJ)/libmilter/libmilter.a
with /usr/lib64/libmilter.a
and
replace $(SENDMAIL_OBJ)/libsm/libsm.a
with /usr/lib64/libsm.a  OR  /usr/lib/libsm.a (if not x86_64 but i386
CentOS)

You need to yum install sendmail-devel to have these files. That
should be sufficient.

Alexander

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

Re: [CentOS] perl

2008-08-21 Thread David G. Miller


Mad Unix [EMAIL PROTECTED] wrote:


Am trying to install perl module File::Find, but not able it gave the
following

cpan[1] install File::Find
CPAN: Storable loaded ok (v2.15)
Going to read /root/.cpan/Metadata
  Database was generated on Thu, 21 Aug 2008 02:03:21 GMT
Running install for module 'File::Find'
The most recent version 1.12 of the module File::Find
is part of the perl-5.10.0 distribution. To install that, you need to run
  force install File::Find   --or--
  install R/RG/RGARCIA/perl-5.10.0.tar.gz
CPAN: Time::HiRes loaded ok (v1.9715)
Running make test
  Can't test without successful make
Running make install
  Make had returned bad status, install seems impossible
Failed during this command:
 RGARCIA/perl-5.10.0.tar.gz   : make NO isa perl


cpan[2] quit
Lockfile removed.
[EMAIL PROTECTED] ftpsync-1.2.33]# perl -version

This is perl, v5.8.8 built for x86_64-linux-thread-multi

Copyright 1987-2006, Larry Wall

Perl may be copied only under the terms of either the Artistic License or
the
GNU General Public License, which may be found in the Perl 5 source kit.

Complete documentation for Perl, including FAQ lists, should be found on
this system using man perl or perldoc perl.  If you have access to the
Internet, point your browser at http://www.perl.org/, the Perl Home Page.
That is just cpan's user friendly way of telling you that File:Find is 
already installed:


[EMAIL PROTECTED] ~]# rpm -q --whatprovides /usr/lib/perl5/5.8.8/File/Find.pm
perl-5.8.8-10.el5_2.3

The command man File::Find should succeed and provide you with the man 
page.


Cheers,
Dave

--
Politics, n. Strife of interests masquerading as a contest of principles.
-- Ambrose Bierce

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

RE: [CentOS] perl

John Thomas wrote:
 Mad Unix wrote:
  Am trying to install perl module File::Find, but not able it gave
  the following 
  
  cpan[1] install File::Find
 snip
 
 I made the mistake of trying to use cpan when I was young.  I am now
 older and wiser and have read things like:
 http://wiki.centos.org/PackageManagement/SourceInstalls?highlight=(cpan)

I attempted this once while building a new machine.  It was amazing how
fast I got bogged down in dependencies.  Almost none of the modules that
I needed were available from the repos and as soon as I packaged up one
and tried to install it, I found I needed 3 more for dependencies (which
also weren't available).

I finally gave up trying to install Perl modules via Yum/rpm.  I now use
CPAN to install Perl packages and Yum for everything else.  I find this
works well for me and saves an incredible amount of time.

If someone would come up with a way to automate the dependency
resolution so that I could grab and package everything at once, then I
would do it, but I just don't have time to go through the download,
package, install, find dependencies, download, package, install, ...
process.  Apparently, this is on the wishlist for cpan2rpm, but is not
implemented yet.

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

Re: [CentOS] Per-user limits for port access

2008-08-21 Thread Sergey Podushkin


Jussi Hirvi wrote:

Is there a way to limit port access on a per-user basis? My practical goal
is to limit, who is allowed to use SMTP AUTH in Sendmail. Any possible
solutions appreciated...

Obviously, limiting by ip will not work, since the very purpose of SMTP AUTH
is to enable mail relaying for trusted users from *anywhere*.
You could use new ability of milter-greylist (from 4.0 version), that 
provides method to limit connections by any criteria, such as remote ip, 
sender/recipient address, authentication status and many more.
You could extend configuration by using external helper, that can be 
reached by, for example, ldap or http protocols.

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

Re: [CentOS] perl

2008-08-21 Thread Karanbir Singh


Bowie Bailey wrote:

If someone would come up with a way to automate the dependency
resolution so that I could grab and package everything at once, then I
would do it, 


dag has a script that does just this.

- KB
--
Karanbir Singh : http://www.karan.org/  : [EMAIL PROTECTED]
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] perl

2008-08-21 Thread Steve Huff



On Aug 21, 2008, at 10:00 AM, Karanbir Singh wrote:


Bowie Bailey wrote:

If someone would come up with a way to automate the dependency
resolution so that I could grab and package everything at once,  
then I

would do it,


dag has a script that does just this.



can you post a link?  i've just been searching dag's site and  
rpmrepo.org for a while, and i can't find any such script.  this would  
be very useful to me as well.


thanks,
-steve

--
If this were played upon a stage now, I could condemn it as an  
improbable fiction. - Fabian, Twelfth Night, III,v




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

[CentOS] Re: Support policy CentOS 5

2008-08-21 Thread Michael Ekstrand

Patrick Derwael [EMAIL PROTECTED]
writes:
 Can someone tell me what the exact CentOS policy for end-of-life is
 for 5.0?

As far as I know, 5.0 itself is pretty much EOL; updates are not being
released for it.  5, on the other hand, won't be EOL for quite a few
years.

 Let me explain: I have been using RHEL for a number of years and had to
 switch to Fedora for installing some packages only supported under F6.

 As I find Fedora's policy for release life duration too stringent (and I
 do no longer need the packages that made me choose F6), I'm considering
 switching over to CentOS5.0
 I know, 5.0 is not the most recent release, but that one is supported by
 Parallel Pro (new package I need)

I imagine that Parallel Pro will probably work on 5.2.  5.2 isn't really
a new release of CentOS, but is rather an update set to CentOS 5.  Think
of it as CentOS 5 Service Pack 2, if you will.  The vast majority of it
(some desktop apps, such as Firefox, excluded) should be 100%
binary-compatible with 5.0.

Running 5.0 is not recommended, as you will not have security support.
If keeping as close to 5.0 as possible is desired, you could probably
install 5.0 and then only install the security-related updates.

- Michael

-- 
mouse, n: A device for pointing at the xterm in which you want to type.
Confused by the strange files?  I cryptographically sign my messages.
For more information see http://www.elehack.net/resources/gpg.


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

Re: [CentOS] perl

Am still not able to install Find_File

On Thu, Aug 21, 2008 at 4:28 PM, Karanbir Singh [EMAIL PROTECTED]wrote:

 Steve Huff wrote:

 If someone would come up with a way to automate the dependency
 resolution so that I could grab and package everything at once, then I
 would do it,

 dag has a script that does just this.

 can you post a link?  i've just been searching dag's site and rpmrepo.orgfor 
 a while, and i can't find any such script.  this would be very useful to
 me as well.


 Here we go:
 http://svn.rpmforge.net/svn/trunk/tools/dar/dar-perl.py

 I am sure Dag will appreciate feedback and info on the script.


 --
 Karanbir Singh : http://www.karan.org/  : [EMAIL PROTECTED]
 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos




-- 
Your search - madunix - did not match any documents.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

RE: [CentOS] iscsi and the last mile...

 Post the output from #fdisk -l

Done many times - just shows the default layout of the PE2950's native
hard drive.

Am I missing something here? So the OS doesn't see the volumes created by your
raid controller? Do you have the module for your raid card loaded?
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Xen bridged networking config

2008-08-21 Thread David Dyer-Bennet

I've got a Centos guest and a Windows 2003 server guest running in Xen
under Centos (5.2 in both cases), and they can get out to the network, and
I can ping them from dom0.  This is my first Xen install, and I haven't
used Linux as a router before (I'm very familiar with it as a webserver
and development platform) so I'm a bit weak on the bridging code and NAT /
IP masquerading.  (Amusingly, I was a voting member of IEEE 802.1 and
worked on the .d standard for bridging; that was back when it was new,
more than a decade ago, and my memories from then may be doing more harm
than good now.)

The problem is, I can't originate a connection to either guest from outside.

Dom0 is getting IP 192.168.1.91 (from our enterprise DHCP server), which
is a reasonable address.  There are two bridges created, xenbr0 and
virbr0.  Virbr0 has an IP of 192.168.122.1 assigned to it, which evidence
suggests did *not* come from our DHCP server.  The interfaces for the two
guest domains have 192.168.122.185 and .198 assigned, also not from our
DHCP server.  The IPs plus the behavior (can't connect in, can connect
out) make it look very much like NAT has been setup, but I can't find any
evidence like a MASQUERADE target in the dom0 iptables or anything (I've
never run NAT on Linux, I could be looking for the wrong thing in the
wrong place).

This setup does not look anything like the default bridging config I
find documented on various Xen websites (particularly
http://wiki.xensource.com/xenwiki/XenNetworking).  None of them have
virbr0.  I've seen a number of reports that people's working
configurations broke when the upgraded to Centos 5.2, so I'm suspecting
that what I'm seeing here is how Centos 5.2 chooses to package Xen.  And
hence I'm asking on a Centos list :-).

I created these guests in virt-manager.  I selected the bridging network
configuration.

Oh, ip forwarding is on in dom0; that was one of the things several places
I looked said was an obvious fix for my problem, but it was already on in
the default config, and it wasn't helping.

I'm currently not using either vlan or bonding, though I believe both are
in my future.  One thing at a time; and if I understand the basics I think
I just might be able to take it to the next stage.

This has gotta be simple; I want the most basic bridged configuration,
with the guests directly visible to the outside world.  Any ideas?

-- 
David Dyer-Bennet, [EMAIL PROTECTED]; http://dd-b.net/
Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/
Photos: http://dd-b.net/photography/gallery/
Dragaera: http://dragaera.info

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

Re: [CentOS] perl

2008-08-21 Thread Steve Huff



On Aug 21, 2008, at 11:14 AM, Mad Unix wrote:


Am still not able to install Find_File



if i may ask, why do you need to install this module?  as was noted  
above, it's already part of perl.  do you know for certain that you  
need a newer version of File::Find than is already installed?


-steve

--
If this were played upon a stage now, I could condemn it as an  
improbable fiction. - Fabian, Twelfth Night, III,v




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

RE: [CentOS] perl

Karanbir Singh wrote:
 Steve Huff wrote:
If someone would come up with a way to automate the dependency
resolution so that I could grab and package everything at once,
then I would do it,
   dag has a script that does just this.
  can you post a link?  i've just been searching dag's site and
  rpmrepo.org for a while, and i can't find any such script.  this
  would be very useful to me as well.
 
 Here we go:
 http://svn.rpmforge.net/svn/trunk/tools/dar/dar-perl.py
 
 I am sure Dag will appreciate feedback and info on the script.

It's not in an rpm???  :)

Thanks for the link.  I don't think I can start using it now since I
have everything installed via CPAN at the moment, but I'll keep it in
mind for the next installation.

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

Re: [CentOS] iscsi and the last mile...

2008-08-21 Thread Jim Wildman


Joseph L. Casale wrote:

Post the output from #fdisk -l

Done many times - just shows the default layout of the PE2950's native
hard drive.


Am I missing something here? So the OS doesn't see the volumes created by your
raid controller? Do you have the module for your raid card loaded?
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos


Yeah, you're missing the iscsi part.  Means the 'drives' are on a 
different system on the other end of a network link.



--
Jim Wildman, CISSP, RHCE   [EMAIL PROTECTED] http://www.rossberry.com
Society in every state is a blessing, but Government, even in its best
state, is a necessary evil; in its worst state, an intolerable one.
Thomas Paine
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] perl

Am trying to install ftpsync-1.2.33, prerequisite
[EMAIL PROTECTED] ftpsync-1.2.33]# cat README | grep ::
- File::Find, IO::Handle
- Net::FTP
- File::Listing


[EMAIL PROTECTED] ~]# perl -version

This is perl, v5.8.8 built for x86_64-linux-thread-multi

Copyright 1987-2006, Larry Wall

Perl may be copied only under the terms of either the Artistic License or
the
GNU General Public License, which may be found in the Perl 5 source kit.

Complete documentation for Perl, including FAQ lists, should be found on
this system using man perl or perldoc perl.  If you have access to the
Internet, point your browser at http://www.perl.org/, the Perl Home Page.

cpan[2] install File::Find
Running install for module 'File::Find'
The most recent version 1.12 of the module File::Find
is part of the perl-5.10.0 distribution. To install that, you need to run
  force install File::Find   --or--
  install R/RG/RGARCIA/perl-5.10.0.tar.gz
Running make test
  Can't test without successful make
Running make install
  Make had returned bad status, install seems impossible
Failed during this command:
 RGARCIA/perl-5.10.0.tar.gz   : make NO isa perl



On Thu, Aug 21, 2008 at 5:33 PM, Steve Huff [EMAIL PROTECTED] wrote:


 On Aug 21, 2008, at 11:14 AM, Mad Unix wrote:

  Am still not able to install Find_File



 if i may ask, why do you need to install this module?  as was noted above,
 it's already part of perl.  do you know for certain that you need a newer
 version of File::Find than is already installed?

 -steve

 --
 If this were played upon a stage now, I could condemn it as an improbable
 fiction. - Fabian, Twelfth Night, III,v



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




-- 
Your search - madunix - did not match any documents.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

RE: [CentOS] perl

Mad Unix wrote:
 Am trying to install ftpsync-1.2.33, prerequisite
 [EMAIL PROTECTED] ftpsync-1.2.33]# cat README | grep ::
 - File::Find, IO::Handle
 - Net::FTP
 - File::Listing
 
 cpan[2] install File::Find
 Running install for module 'File::Find'
 The most recent version 1.12 of the module File::Find
 is part of the perl-5.10.0 distribution. To install that, you need to
   run force install File::Find   --or--
   install R/RG/RGARCIA/perl-5.10.0.tar.gz

Cpan is trying to install the latest File::Find, but since File::Find is
part of the base Perl distribution, it can't upgrade it without
upgrading your Perl version as well (which is not allowed by default).

To determine if you have File::Find already installed (which you
should), do this:

perl -e 'use File::Find'

If that runs without errors, it is already installed and you don't have
to worry about it.

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

Re: [CentOS] perl

Thanks,

[EMAIL PROTECTED] ftpsync-1.2.33]# perl -e 'use File::Find'
[EMAIL PROTECTED] ftpsync-1.2.33]#

On Thu, Aug 21, 2008 at 5:54 PM, Bowie Bailey [EMAIL PROTECTED] wrote:

 Mad Unix wrote:
  Am trying to install ftpsync-1.2.33, prerequisite
  [EMAIL PROTECTED] ftpsync-1.2.33]# cat README | grep ::
  - File::Find, IO::Handle
  - Net::FTP
  - File::Listing
 
  cpan[2] install File::Find
  Running install for module 'File::Find'
  The most recent version 1.12 of the module File::Find
  is part of the perl-5.10.0 distribution. To install that, you need to
run force install File::Find   --or--
install R/RG/RGARCIA/perl-5.10.0.tar.gz

 Cpan is trying to install the latest File::Find, but since File::Find is
 part of the base Perl distribution, it can't upgrade it without
 upgrading your Perl version as well (which is not allowed by default).

 To determine if you have File::Find already installed (which you
 should), do this:

 perl -e 'use File::Find'

 If that runs without errors, it is already installed and you don't have
 to worry about it.

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




-- 
Your search - madunix - did not match any documents.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

RE: [CentOS] perl

Karanbir Singh wrote:
 Steve Huff wrote:
If someone would come up with a way to automate the dependency
resolution so that I could grab and package everything at once,
then I would do it,
   dag has a script that does just this.
  can you post a link?  i've just been searching dag's site and
  rpmrepo.org for a while, and i can't find any such script.  this
  would be very useful to me as well.
 
 Here we go:
 http://svn.rpmforge.net/svn/trunk/tools/dar/dar-perl.py
 
 I am sure Dag will appreciate feedback and info on the script.

I grabbed the file and tried it out to see what it would do.  It dumped
what looked like a spec file and then gave an error.

$ ./dar-perl.py Compress::Zlib

[normal spec information snipped]

%files
%defattr(-, root, root, 0755)
%doc Changes MANIFEST META.yml README examples/
%doc %{_mandir}/man3/Compress::Zlib.3pm*
%dir %{perl_vendorlib}/Compress/
#%{perl_vendorlib}/Compress/Zlib/
%{perl_vendorlib}/Compress/Zlib.pm

%changelog
Traceback (most recent call last):
  File ./dar-perl.py, line 679, in ?
print out, '* %s %s - %s-1' % (time.strftime('%a %b %d %Y',
time.localtime()), authorities[logname], version)
KeyError: 'bowieb'
$ 

Any idea what happened here?

What exactly is this program supposed to do?  Can someone give me an
example of how to use it to install a Perl module and its dependencies?

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

Re: [CentOS] perl

2008-08-21 Thread Karanbir Singh


Bowie Bailey wrote:

http://svn.rpmforge.net/svn/trunk/tools/dar/dar-perl.py


What exactly is this program supposed to do?  Can someone give me an
example of how to use it to install a Perl module and its dependencies?


There is an rpmforge mailing list for tools, thats the best place for 
this conversation.


--
Karanbir Singh : http://www.karan.org/  : [EMAIL PROTECTED]
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

RE: [CentOS] Support policy CentOS 5

2008-08-21 Thread Patrick Derwael

you'll find the end-of-life information of the centos releases at:

http://wiki.centos.org/FAQ/General?highlight=(end+of+life)#head-fe8a0be91ee
3e7dea812e8694491e1dde5b75e6d

that FAQ page may also answer other questions you may have, (e.g.,
the relationship between rhel and centos).

  - Rick



[Patrick Derwael] Rick

Thanks for the link, I now have the info I need !



  _  

avast! Antivirus http://www.avast.com : Inbound message clean. 


Virus Database (VPS): 080821-0, 21/08/2008
Tested on: 21/08/2008 17:52:33
avast! - copyright (c) 1988-2008 ALWIL Software.




  _  

avast! Antivirus http://www.avast.com : Outbound message clean. 


Virus Database (VPS): 080821-0, 21/08/2008
Tested on: 21/08/2008 18:21:37
avast! - copyright (c) 1988-2008 ALWIL Software.




  _  

avast! Antivirus http://www.avast.com : Outbound message clean. 


Virus Database (VPS): 080821-0, 21/08/2008
Tested on: 21/08/2008 18:23:53
avast! - copyright (c) 1988-2008 ALWIL Software.


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

RE: [CentOS] Re: Support policy CentOS 5

2008-08-21 Thread Patrick Derwael

Patrick Derwael [EMAIL PROTECTED]
writes:
 Can someone tell me what the exact CentOS policy for end-of-life is
 for 5.0?

As far as I know, 5.0 itself is pretty much EOL; updates are not being
released for it.  5, on the other hand, won't be EOL for quite a few
years.

snip

I imagine that Parallel Pro will probably work on 5.2.  5.2 isn't really
a new release of CentOS, but is rather an update set to CentOS 5.  Think
of it as CentOS 5 Service Pack 2, if you will.  The vast majority of it
(some desktop apps, such as Firefox, excluded) should be 100%
binary-compatible with 5.0.

Running 5.0 is not recommended, as you will not have security support.
If keeping as close to 5.0 as possible is desired, you could probably
install 5.0 and then only install the security-related updates.

- Michael

[Patrick Derwael] 

Michael,
The  point is that Parallel supports exactly CentOS 4.4 and 5.0, Fedora 4
and 6 and RHEL 4ES and 5ES.
I'm a bit stuck here!

As far as my other systems are concerned, (pen testing) I will reinstall
them under Centos 5.2 in the coming weeks

Thanks!



  _  

avast! Antivirus http://www.avast.com : Inbound message clean. 


Virus Database (VPS): 080821-0, 21/08/2008
Tested on: 21/08/2008 17:52:36
avast! - copyright (c) 1988-2008 ALWIL Software.




  _  

avast! Antivirus http://www.avast.com : Outbound message clean. 


Virus Database (VPS): 080821-0, 21/08/2008
Tested on: 21/08/2008 18:21:41
avast! - copyright (c) 1988-2008 ALWIL Software.




  _  

avast! Antivirus http://www.avast.com : Outbound message clean. 


Virus Database (VPS): 080821-0, 21/08/2008
Tested on: 21/08/2008 18:23:54
avast! - copyright (c) 1988-2008 ALWIL Software.


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

Re: [CentOS] Re: Support policy CentOS 5

2008-08-21 Thread Karanbir Singh


Patrick Derwael wrote:

*/[Patrick Derwael] /*
The  point is that Parallel supports exactly CentOS 4.4 and 5.0, Fedora 
4 and 6 and RHEL 4ES and 5ES.


This bit is interesting. I dont know anything about 'Parallel' so keep 
that in mind.


Most vendors when they say only 1 release and only that specific release 
is supported do so because they have kernel modules that are built 
against the kernel shipped in that version. Now when you install 
CentOS-5.0 you get the kernel that was shipped with 5.0 - when you do 
the yum updates, that kernel is not removed. The default policy for yum 
on CentOS is to not 'upgrade' the kernel, but to also install a newer 
one, if its available. Primary focus of this policy is to address the 
specific issue you are running up against, and also in cases where 
people have specific hardware support drivers compiled into the kernel 
or when they are running custom builds of the kernel.


Now, I dont know what parallel is or what you are doing there, but I 
dont belive the Vendor that only one update level is supported. Perhaps 
they have a dependancy on a specific package ( like the kernel ) - those 
are easy to lock into a version with things like yum-versionlock.



As far as my other systems are concerned, (pen testing) I will reinstall 
them under Centos 5.2 in the coming weeks


you dont need to reinstall to get from 5.0 to 5.2, if you do a yum 
update, that will bring in all the updated packages. ( remember 
5.0+updates == 5.2 ).


--
Karanbir Singh : http://www.karan.org/  : [EMAIL PROTECTED]
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

RE: [CentOS] perl

Karanbir Singh wrote:
 Bowie Bailey wrote:
   http://svn.rpmforge.net/svn/trunk/tools/dar/dar-perl.py
   
  What exactly is this program supposed to do?  Can someone give me an
  example of how to use it to install a Perl module and its
  dependencies? 
 
 There is an rpmforge mailing list for tools, thats the best place for
 this conversation.

In that case, I will move this conversation over there.

Thanks.

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

[CentOS] Re: Adaptec 2820 2gig+ partitions


on 8-20-2008 2:57 PM Brian Marshall spake the following:

Hi All,

I have some general questions about setting up partitions. I have been 
struggling to get an array to mount since I upgraded the drives and 
exceeded a 2 gig partition so now the logical drive won't mount after 
the install.


That being said I have a few questions.

Does anyone have any experience or advice for setting up large arrays on 
5.2 x86_64? My understanding is I need to setup a GPT partition but I am 
not having any luck getting a good configuration during the install 
process. Should I just not try during the install and use parted 
afterwards or is there a better way to go about this?


Also, during testing I just let it do a default configuration on the 
drives and noticed that it setup a 100mgb partition with the boot map on 
it and the rest as a LVM volume. Is this a preferred way to do it these 
days?


It is the default way, but there are arguments both ways for it being the 
preferred way.



--
MailScanner is like deodorant...
You hope everybody uses it, and
you notice quickly if they don't



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Re: Disk quotas for Sendmail


on 8-21-2008 4:10 AM Jussi Hirvi spake the following:

On 8/21/08, Jussi Hirvi [EMAIL PROTECTED] wrote:
CentOS 5.x, Sendmail 8.13.8. http://8.13.8.
I would like to start using disk quotas on my Sendmail server, and optimally
they should be configurable by user or group. How can this be done? (I tried
to google, but no success.)


Rudi Ahlers ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 13:55):

If each user's mail is stored in his homefolder, then you just need to enable
user quotas :) man quota  man mtab for the options


Hei,

The mail is not stored in home folders but in /var/spool/mail. I would
rather keep it that way - and I'm not sure if storing mail in home folders
would work for procmail (Sendmail's local mailer), dovecot (pop server) etc.

- Jussi


You can make /var/spool/mail a separate mounted partition, and set user and 
group quotas on that mount and it should work. It might work with it not being 
a separate partition, but that would be up to you to test.


But you can have other problems if you set the warning and maximum levels too 
close. If a message comes in and makes the quota go over the maximum and the 
warning level at the same time, the system won't be able to give the users any 
warning message as their quota will be over and will block the warning message.


You can set procmail to store the messages in other folders, and also set it 
to deliver to Maildir storage by fiddling with (or creating) an 
/etc/procmailrc file.


--
MailScanner is like deodorant...
You hope everybody uses it, and
you notice quickly if they don't



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Re: Disk quotas for Sendmail


on 8-21-2008 5:56 AM Jussi Hirvi spake the following:

Alexander Dalloz ([EMAIL PROTECTED]) kirjoitteli (21.8.2008 14:31):

http://freshmeat.net/projects/milterquota/


I didn't find milterquota with yum so I guess I have to build it... The
Makefile asks me to locate a file that I cannot find:


# change this to the object directory in the sendmail build
treeSENDMAIL_OBJ=/usr/src/packages/sendmail-8.12.9/obj.Linux.2.4.21.i586


I just installed sendmail source, but still I cannot find a obj.Linux
anywhere. (I searched with the command locate obj.Linux.)

How should I proceed?

- Jussi

--
Jussi Hirvi * Green Spot
Topeliuksenkatu 15 C * 00250 Helsinki * Finland
Tel.  fax +358 9 493 981 * Mobile +358 40 771 2098 (only sms)
[EMAIL PROTECTED] * http://www.greenspot.fi

You shouldn't need sendmail source, just install the sendmail-devel rpm.


--
MailScanner is like deodorant...
You hope everybody uses it, and
you notice quickly if they don't



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Re: Support policy CentOS 5


on 8-21-2008 9:24 AM Patrick Derwael spake the following:

Patrick Derwael [EMAIL PROTECTED]
writes:

 Can someone tell me what the exact CentOS policy for end-of-life is
 for 5.0?


As far as I know, 5.0 itself is pretty much EOL; updates are not being
released for it.  5, on the other hand, won't be EOL for quite a few
years.

snip

I imagine that Parallel Pro will probably work on 5.2.  5.2 isn't really
a new release of CentOS, but is rather an update set to CentOS 5.  Think
of it as CentOS 5 Service Pack 2, if you will.  The vast majority of it
(some desktop apps, such as Firefox, excluded) should be 100%
binary-compatible with 5.0.

Running 5.0 is not recommended, as you will not have security support.
If keeping as close to 5.0 as possible is desired, you could probably
install 5.0 and then only install the security-related updates.

- Michael

*/[Patrick Derwael] /*

Michael,
The  point is that Parallel supports exactly CentOS 4.4 and 5.0, Fedora 
4 and 6 and RHEL 4ES and 5ES.

I’m a bit stuck here!

As far as my other systems are concerned, (pen testing) I will reinstall 
them under Centos 5.2 in the coming weeks


Thanks!
Do you have a link to this application's website? Maybe we could determine why 
it might be stuck to a limited set of OS releases. If a software can't keep up 
with a limited subset of OS updates, maybe they are concerned more with costs 
then security.


--
MailScanner is like deodorant...
You hope everybody uses it, and
you notice quickly if they don't



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] General Linux query

2008-08-21 Thread Ignacio Vazquez-Abrams

On Thu, 2008-08-21 at 05:38 -0600, Warren Young wrote:
 lingu wrote:
  4)  what are all the precaution to be made to prevent file system 
  corruption.
 
 1. Put the computer on a UPS.
 
 2. Configure nut (http://networkupstools.org/) to bring the system down 
 safely well before the UPS's battery quits.  You don't want to 
 fully-discharge a lead-acid battery, which is what UPSes use.  Shut down 
 before it drops below about 20%.
 
 3. Use some form of redundant RAID.  (Not RAID-0!)
 
 4. Run the RAID system's consistency check function periodically.
 
 5. If the system's RAID card offers a battery backup option, get one.
 
 6. If the system is experiencing kernel panics, fix the problem that's 
 causing it.  Kernel panics are almost always due to broken hardware. 
 Replace the broken hardware.  Very rarely, they may be due to a bad 
 driver or a kernel bug, but when you hear hoof beats, think horses, not 
 zebras.
 
 7. Back up your system.
 
 8. BACK UP YOUR SYSTEM.
 
 9. BACK! UP! YOUR! SYSTEM!!!  (And then, back it up again.)

10. TEST the backups, preferably on another system. All the backup tapes
in the world won't help if you think you've been doing it right for the
past 5 months, but in reality all the tapes are either blank or damaged.

-- 
Ignacio Vazquez-Abrams [EMAIL PROTECTED]

PLEASE don't CC me; I'm already subscribed


signature.asc
Description: This is a digitally signed message part
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Re: perl


on 8-21-2008 8:41 AM Mad Unix spake the following:

Am trying to install ftpsync-1.2.33, prerequisite
[EMAIL PROTECTED] ftpsync-1.2.33]# cat README | grep ::
- File::Find, IO::Handle
- Net::FTP
- File::Listing


[EMAIL PROTECTED] ~]# perl -version

This is perl, v5.8.8 built for x86_64-linux-thread-multi

Copyright 1987-2006, Larry Wall

Perl may be copied only under the terms of either the Artistic License 
or the

GNU General Public License, which may be found in the Perl 5 source kit.

Complete documentation for Perl, including FAQ lists, should be found on
this system using man perl or perldoc perl.  If you have access to the
Internet, point your browser at http://www.perl.org/, the Perl Home Page.

cpan[2] install File::Find
Running install for module 'File::Find'
The most recent version 1.12 of the module File::Find
is part of the perl-5.10.0 distribution. To install that, you need to run
  force install File::Find   --or--
  install R/RG/RGARCIA/perl-5.10.0.tar.gz
Running make test
  Can't test without successful make
Running make install
  Make had returned bad status, install seems impossible
Failed during this command:
 RGARCIA/perl-5.10.0.tar.gz   : make NO isa perl


This still shows you trying to install a module that already exists. You were 
asked if you know for sure that you need a newer version of File::Find and you 
answered with a failed attempt to install it. If you want help, you will have 
to answer questions asked by those trying to help you.


--
MailScanner is like deodorant...
You hope everybody uses it, and
you notice quickly if they don't



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

RE: [CentOS] iscsi and the last mile...

Yeah, you're missing the iscsi part.  Means the 'drives' are on a
different system on the other end of a network link.

Heh, thought you were trying to export those discs *on* the iSCSI server
and couldn't see them there, my bad :)

'iscsiadm -m discovery -p ip of target' wont give the target name, you need
'iscsiadm --mode discovery --type sendtargets --portal ip of target'

You also need to pass tcp 3260 through the firewall on the target, I have never
used RH's target, does it do this for you? Iet allows you to define 
allowed/denied
clients, does RH's?

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

[CentOS] What fires logrotate

2008-08-21 Thread Al Sparks

I've been taking a look at how RedHat (and CentOS) handles logrotate.
According to the man page, logrotate is supposed to be fired by cron.
But when I look at root's crontab
   $ sudo crontab lu root
   no crontab for root

What exactly fires logrotate (and other scheduled events like
logwatch, which ends up in root's inbox)?
   === Al
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Dell Perc snmp

I am trying to help a friend with a 2850 and snmp monitoring of a controller 
and its arrays.
I don't have any dells, do they have a rpm to make easy work of this for the 
Percs? They have
a monitoring solution already. Any hints I could pass on would be appreciated!

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

[CentOS] VPN traceroute is not consistent/ Bind cannot transfer

2008-08-21 Thread Fajar Priyanto

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi all,
I've got this strange thing:
I setup a openvpn link between two openwrts.
NetA: 192.168.2.0/24
NetB: 192.168.0.0/24

In netA there is a mail server + DNS: 192.168.2.44, with domain name
branch.abc.com
In netB there is a mail server + DNS: 192.168.0.44, with domain name abc.com

I setup both DNS server to be a slave server of each other domain.
- From NetA, it is able to 'copy' the abc.com zone from NetB.
But from NetB, it isn't able to copy the branch.abc.com from NetA.
In /var/log/messages, the error is:
mail named[1566]: zone branch.abc.com/IN/external: Transfer started.
mail named[1566]: transfer of 'branch.abc.com/IN' from 192.168.2.44#53:
connected using 192.168.0.44#50125
mail named[1566]: transfer of 'branch.abc.com/IN' from 192.168.2.44#53:
failed while receiving responses: REFUSED
mail named[1566]: transfer of 'branch.abc.com/IN' from 192.168.2.44#53:
end of transfer
mail named[1566]: zone branch.abc.com/IN/internal: refresh: unexpected
rcode (REFUSED) from master 192.168.2.44#53 (source 0.0.0.0#0)

I've put allow-transfer in:
zone branch.abc.com {
type master;
allow-transfer { 192.168.0.44; };
file branch.abc.com.zone;
};

Adding to the confusion:
- From NetB, doing a traceroute to 192.168.2.44 result is OK. But, when I
restart named, the traceroute to 192.168.2.44 brings no response. But,
tracerouting to other PC in NetA succeed.
[EMAIL PROTECTED] ~]# traceroute 192.168.2.136
traceroute to 192.168.2.136 (192.168.2.136), 30 hops max, 40 byte packets
 1  192.168.0.2 (192.168.0.2)  0.630 ms  1.989 ms  2.381 ms
 2  192.168.0.211 (192.168.0.211)  121.073 ms  140.998 ms  160.724 ms
 3  192.168.2.136 (192.168.2.136)  179.393 ms * *
[EMAIL PROTECTED] ~]# traceroute 192.168.2.44
traceroute to 192.168.2.44 (192.168.2.44), 30 hops max, 40 byte packets
 1  192.168.0.2 (192.168.0.2)  0.976 ms  1.342 ms  1.733 ms
 2  192.168.0.211 (192.168.0.211)  136.118 ms  155.800 ms  175.573 ms
 3  192.168.2.44 (192.168.2.44)  209.216 ms  225.935 ms  245.866 ms
[EMAIL PROTECTED] ~]# traceroute 192.168.2.44
traceroute to 192.168.2.44 (192.168.2.44), 30 hops max, 40 byte packets
 1  * * *
 2  * * *
 3  * * *


I'm not sure what is causing this, either:
1. Bind configuration error
2. Inconsistent/strangeness in VPN

Any thoughts?
Thank you very much.
- --
Fajar Priyanto | Reg'd Linux User #327841 | Linux tutorial
http://linux2.arinet.org
13:10:54 up 5:02, 2.6.24-18-generic GNU/Linux
Let's use OpenOffice. http://www.openoffice.org
The real challenge of teaching is getting your students motivated to learn.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIra9V8TneBL/L6RoRAo4MAJ4jLoxHD5ZE7bH2m/66fdWkdY73iACeKVmA
qG7636EFa4/4dgQsXpZE+x4=
=HdQ/
-END PGP SIGNATURE-
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] What fires logrotate

2008-08-21 Thread Steve Huff



On Aug 21, 2008, at 2:06 PM, Al Sparks wrote:


What exactly fires logrotate (and other scheduled events like
logwatch, which ends up in root's inbox)?


look at /etc/cron.daily/logrotate and /etc/cron.daily/0logwatch.   
those scripts (and the others in /etc/cron.*ly) are invoked by the  
following code in /etc/crontab:


# run-parts
01 * * * * root run-parts /etc/cron.hourly
02 4 * * * root run-parts /etc/cron.daily
22 4 * * 0 root run-parts /etc/cron.weekly
42 4 1 * * root run-parts /etc/cron.monthly

-steve

--
If this were played upon a stage now, I could condemn it as an  
improbable fiction. - Fabian, Twelfth Night, III,v




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

Re: [CentOS] What fires logrotate

2008-08-21 Thread Stephen Harris

On Thu, Aug 21, 2008 at 11:06:19AM -0700, Al Sparks wrote:
 I've been taking a look at how RedHat (and CentOS) handles logrotate.
 According to the man page, logrotate is supposed to be fired by cron.
 But when I look at root's crontab
$ sudo crontab lu root
no crontab for root

See /etc/cron.daily
(see also other directories matching /etc/cron.*)

Look in /etc/crontab to see how they're called.

-- 

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

Re: [CentOS] Dell Perc snmp

2008-08-21 Thread Al Sparks

Try this:

   http://linux.dell.com/wiki/index.php/Repository/hardware

You can actually use yum to download Dell's rpm repositories.

Also, they have their own mailing list that I've found very helpful for these 
types of issues.

It's at [EMAIL PROTECTED]

   === Al



- Original Message 
From: Joseph L. Casale [EMAIL PROTECTED]
To: CentOS mailing list centos@centos.org
Sent: Thursday, August 21, 2008 10:07:39 AM
Subject: [CentOS] Dell Perc snmp

I am trying to help a friend with a 2850 and snmp monitoring of a controller 
and its arrays.
I don't have any dells, do they have a rpm to make easy work of this for the 
Percs? They have
a monitoring solution already. Any hints I could pass on would be appreciated!

Thanks,
jlc
___
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] What fires logrotate

2008-08-21 Thread Al Sparks

--- On Thu, 8/21/08, Stephen Harris [EMAIL PROTECTED] wrote:

 From: Stephen Harris [EMAIL PROTECTED]
 Subject: Re: [CentOS] What fires logrotate
 To: CentOS mailing list centos@centos.org
 Date: Thursday, August 21, 2008, 10:13 AM
 On Thu, Aug 21, 2008 at 11:06:19AM -0700, Al Sparks wrote:
  I've been taking a look at how RedHat (and CentOS)
 handles logrotate.
  According to the man page, logrotate is supposed to be
 fired by cron.
  But when I look at root's crontab
 $ sudo crontab lu root
 no crontab for root

 See /etc/cron.daily
 (see also other directories matching /etc/cron.*)

 Look in /etc/crontab to see how they're called.

 -- 

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

Thanks.  Very helpful.
   === Al
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] What fires logrotate

2008-08-21 Thread Joshua Baker-LePain


On Thu, 21 Aug 2008 at 11:06am, Al Sparks wrote


I've been taking a look at how RedHat (and CentOS) handles logrotate.
According to the man page, logrotate is supposed to be fired by cron.
But when I look at root's crontab
  $ sudo crontab lu root
  no crontab for root

What exactly fires logrotate (and other scheduled events like
logwatch, which ends up in root's inbox)?


Look in /etc/cron.{hourly,daily,monthly,weekly}.

--
Joshua Baker-LePain
QB3 Shared Cluster Sysadmin
UCSF
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Joseph L. Casale wrote:
I am trying to help a friend with a 2850 and snmp monitoring of a controller
and its arrays.
I don't have any dells, do they have a rpm to make easy work of this for the
Percs? They have
a monitoring solution already. Any hints I could pass on would be
appreciated!

I haven't tried to setup Dell's SNMP stuff, but I did download their
driver development kit and wrote a perl script to wrap around their
raidcfg utility to detect a bad(missing) disk.

[EMAIL PROTECTED] ~]# /usr/lib/nagios/plugins/check_dell_raid
OK: Both drives OK in disk subsystem
Controller: PERC 4e/Si
VDisk ID: 0
Size: 286080 MB (279 GB)
Type: RAID 1
Read Policy: Adaptive Read Ahead
Write Policy: Write Back
Cache Policy: Direct I/O
Stripe Size: 64
Drives: 0:0:0,0:1:0

The script assumes that there are only 2 disks, and if it doesn't
see two disks in the Drives: line will spit out a warning.

It does require root access so my nagios calls sudo in order to
call the script.

The script is here
http://portal.aphroland.org/~aphro/check_dell_raid.txt

I'm not a programmer so my scripts aren't the neatest by anyone's
standards I'm sure but it does work for me I have it deployed to
about 75 systems at the moment and it runs every 5 minutes. I
did find that the raidcfg tool that dell provides does seem to
crash some systems(2850s if I recall), but that could be because
the kernel was too old or something. The script checks the model
of the server and will refuse to run unless it matches.

I believe the source dell files are here:
http://support.dell.com/support/downloads/format.aspx?c=usl=ens=gendeviceid=8307libid=7releaseid=R159785vercnt=2formatcnt=0SystemID=PWE_1950servicetag=os=RHEL5osl=encatid=-1impid=-1

I install these RPMs on my system for comparison:
./i386/srvadmin-hapi-5.3.0-266.i386.rpm
./i386/srvadmin-hapi-dtk-5.3.0-266.i386.rpm
./i386/srvadmin-ipmi-5.3.0-266.rhel4.i386.rpm
./i386/srvadmin-omilcore-5.3.0-266.i386.rpm
./i686/dell-dtk-2.5-1.i686.rpm
./noarch/dkms-2.0.13-1.noarch.rpm
./noarch/openipmi-33.13.RHEL4-1dkms.noarch.rpm

(the dell-dtk rpm is a custom one I built with the dell tools, the
rest are direct from dell).

I've only tested these on CentOS 4.x and RHEL 4.x. Have not tested
on 5.x.

Certainly wasn't as easy as using HP's hpacucli utility that is
available to scan HP's storage arrays.

nate

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

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread MHR

On Thu, Aug 21, 2008 at 6:30 AM, Joost Waversveld [EMAIL PROTECTED] wrote:
 Try to find the file through the find command, something like  find /usr -t
 f -name 'obj.Linux*' 

 The command locate works with an database, which will be updated just once a
 day (If you enable it!), so that will not find any just installed files...


Or when you run updatedb (as root)

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

Re: [CentOS] Dell Perc snmp

2008-08-21 Thread nate

nate wrote:

 I haven't tried to setup Dell's SNMP stuff, but I did download their
 driver development kit and wrote a perl script to wrap around their
 raidcfg utility to detect a bad(missing) disk.

Now that I've looked at it more, it's the Dell Deployment Toolkit not
the Driver development kit(if there is one, not sure).

nate

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

Re: [CentOS] General Linux query

2008-08-21 Thread MHR

On Thu, Aug 21, 2008 at 4:13 AM, lingu [EMAIL PROTECTED] wrote:
 HI,

   This is more general  linux  related topic ,  i want to know few  basic
 things.

Google is your friend.  A good spell checker might not hurt, either.  :-)

 1) How file systeem get corrupted on linux?

Pretty much the same way as it might happen under any other OS, but
less frequently if you use a journaling file system.  Also, since
Linux is a significantly more reliable system than many others
overall, the odds are lower for such a thing to happen at all.

 2)  why,when and how fsck to be run without lossing data?

This really depends more on what kind of file system you are using
than fsck.  Fsck's purpose is to ensure file system integrity.  YMMV.

 3)  Is there is any other tool  other than fsck available for file system
 check

For file system integrity, not really.  For any other purpose, you're
on your own.

 4)  what are all the precaution to be made to prevent file system
 corruption.

Pretty much the same precautions you take under any other OS -
reliable hardware and backups.

 5)  why  when the system will get hanged  and what are all the possible
 reasons?

Pretty much the same as under any other OS - livelocks and deadlocks.
Linux tends to have fewer of both.

  If any one provide me suitable links providing all of the  above
 information  in detail it will be quite usefull.

http://www.google.com and http://centos.org are good places to start.

(IOW, don't expect to get all the answers handed to you.  If you're
going to play with computers in general, and Linux in particular, you
need to do your homework before asking a question on this, and most
any other, list, and then make the question specific if you want a
good answer.)

Regards,

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

[CentOS] Re: Support policy CentOS 5

2008-08-21 Thread Robert Nichols


Patrick Derwael wrote:

you'll find the end-of-life information of the centos releases at:

http://wiki.centos.org/FAQ/General?highlight=(end+of+life)#head-fe8a0be91ee
3e7dea812e8694491e1dde5b75e6d

that FAQ page may also answer other questions you may have, (e.g.,
the relationship between rhel and centos).

  - Rick



[Patrick Derwael] Rick

Thanks for the link, I now have the info I need !


But, read that table carefully. It says CentOS 5 not CentOS 5.0
CentOS 5 tracks the current point release and is now equivalent
to CentOS 5.2.  Running CentOS 5.0 means you do not receive any
updates from a subsequent point release and are locked into the
package versions that were current just before the release of
CentOS 5.1.

--
Bob Nichols NOSPAM is really part of my email address.
Do NOT delete it.

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

RE: [CentOS] Dell Perc snmp

I've only tested these on CentOS 4.x and RHEL 4.x. Have not tested
on 5.x.

I am sure you and Al's post will be sufficient!

Certainly wasn't as easy as using HP's hpacucli utility that is
available to scan HP's storage arrays.

nate

OT: I would sure love to get this working inside of busybox so I could use it
on my esx 3.5i server w/ an MSA20 on it :) Any tips?


Thanks everyone!
jlc
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Disk quotas for Sendmail

Joost Waversveld [EMAIL PROTECTED] escribio (21.8.2008 16:30)
 Try to find the file through the find command, something like  find /usr -t f
 -name 'obj.Linux*' 
 
 The command locate works with an database, which will be updated just once a
 day (If you enable it!), so that will not find any just installed files...

Thanks - I know that about locate. That's why I commanded updatedb before I
tried the locate command. Locate is so easy to use, in general I like it
more than find. 

- Jussi

--
Jussi Hirvi  *  Green Spot
Topeliuksenkatu 15 C  *  00250 Helsinki  *  Finland
Tel.  fax +358 9 493 981  *  SMS +358 40 771 2098
[EMAIL PROTECTED]  *  http://www.greenspot.fi


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

[CentOS] kickstart error on 5.2 exception

2008-08-21 Thread Jerry Geis


Hi,

I am trying to get my kickstart file that worked under 5.1 to work under 5.2 
centos x86_64.
This is the error that I get. 
On the screen it says Exception occured and gives me the option to save it. This is that file.


I dont see any odd that would cause it to crash. 
Can anyone help. My kickstart file is in the mix below.


Seems to be related to network, my line seems fine (I think) for network.

Thanks,

Jerry
-
Traceback (most recent call first):
 File /usr/lib/anaconda/network.py, line 341, in lookupHostname
   ret = isys.pumpNetDevice(dev.get('device'), dev.get('dhcpclass'))
 File /usr/lib/anaconda/network.py, line 531, in write
   ip = self.lookupHostname()
 File /tmp/treedir.1828/instimage/usr/lib/anaconda/yuminstall.py, line 1401, 
in doPreInstall
   anaconda.id.network.write(anaconda.rootPath)
 File /tmp/treedir.1828/instimage/usr/lib/anaconda/backend.py, line 180, in 
doPreInstall
   anaconda.backend.doPreInstall(anaconda)
 File /usr/lib/anaconda/dispatch.py, line 201, in moveStep
   rc = stepFunc(self.anaconda)
 File /usr/lib/anaconda/dispatch.py, line 124, in gotoNext
   self.moveStep()
 File /usr/lib/anaconda/gui.py, line 1018, in nextClicked
   self.anaconda.dispatch.gotoNext()
 File /usr/lib/anaconda/iw/progress_gui.py, line 243, in renderCallback
   self.intf.icw.nextClicked()
 File /usr/lib/anaconda/gui.py, line 1045, in handleRenderCallback
   self.currentWindow.renderCallback()
AttributeError: 'module' object has no attribute 'pumpNetDevice'

Local variables in innermost frame:
self: network.Network instance at 0x2b2109c8b680
myns: 65.24.0.168
dev: DEVICE=eth0
BOOTPROTO=dhcp
DHCPCLASS=
HWADDR=00:13:D4:FA:EB:53
ONBOOT=yes



Anaconda instance, containing members:
rootPath: /mnt/sysimage
rescue_mount: True
intf: InstallInterface instance, containing members:
 intf.ppw: InstallProgressWindow instance, containing members:
   intf.ppw.pixmaps: [rnotes/01-centos5-welcome.png, 
rnotes/02-centos5-donate.png, rnotes/03-centos5-yum.png, 
rnotes/04-centos5-repos.png, rnotes/05-centos5-centosplus.png, 
rnotes/06-centos5-support.png, rnotes/07-centos5-docs.png, 
rnotes/08-centos5-wiki.png, rnotes/09-centos5-virtualization.png]
   intf.ppw.intf: Already dumped
   intf.ppw.pixcurnum: 0
   intf.ppw.sizeComplete: 0
   intf.ppw.lastTimeEstimate: None
   intf.ppw.wrappedpixlist: 0
   intf.ppw.filesComplete: 0
   intf.ppw.pixtimer: None
   intf.ppw.initialTimeEstimate: None
   intf.ppw.adpix: gtk.Image object (GtkImage) at 0xec6cd20
   intf.ppw.adbox: gtk.EventBox object (GtkEventBox) at 0xec6ce10
   intf.ppw.curPackage: {'summary': gtk.Label object (GtkLabel) at 0xfb1d0a0, 
'package': gtk.Label object (GtkLabel) at 0xfb1d050}
   intf.ppw.estimateHistory: []
   intf.ppw.numComplete: 0
   intf.ppw.pkgstatus: None
   intf.ppw.ics: InstallControlState instance, containing members:
 intf.ppw.ics.prevEnabled: False
 intf.ppw.ics.cw: InstallControlWindow instance, containing members:
   intf.ppw.ics.cw.handle: 150
   intf.ppw.ics.cw.mainxml: glade.XML object (PyGladeXML) at 0xec6c5a0
   intf.ppw.ics.cw.reloadRcQueued: 0
   intf.ppw.ics.cw.currentWindow: Already dumped
   intf.ppw.ics.cw.window: gtk.Window object (GtkWindow) at 0xec6c550
   intf.ppw.ics.cw.installFrame: gtk.Frame object (GtkFrame) at 0xec6c690
   intf.ppw.ics.cw.anaconda: Already dumped
   intf.ppw.ics.cw.rnv: ReleaseNotesViewer instance, containing members:
 intf.ppw.ics.cw.rnv.textWin: gtk.Window object (GtkWindow) at 
0xec6ef50
 intf.ppw.ics.cw.rnv.vue: gtkhtml2.View object (HtmlView) at 
0x2b2109cc9f00
 intf.ppw.ics.cw.rnv.currentURI: /mnt/source/RELEASE-NOTES-en_US.html
 intf.ppw.ics.cw.rnv.topDir: /mnt/source
 intf.ppw.ics.cw.rnv.doc: gtkhtml2.Document object (HtmlDocument) at 
0x2b2109cc9fa0
 intf.ppw.ics.cw.rnv.height: 600
 intf.ppw.ics.cw.rnv.width: 800
 intf.ppw.ics.cw.rnv.opener: FancyURLopener instance, containing 
members:
   intf.ppw.ics.cw.rnv.opener._URLopener__unlink: built-in function 
unlink
   intf.ppw.ics.cw.rnv.opener._URLopener__tempfiles: []
   intf.ppw.ics.cw.rnv.opener.cert_file: None
   intf.ppw.ics.cw.rnv.opener.addheaders: [('User-agent', 
'Python-urllib/1.16')]
   intf.ppw.ics.cw.rnv.opener.tries: 0
   intf.ppw.ics.cw.rnv.opener.ftpcache: {}
   intf.ppw.ics.cw.rnv.opener.tempcache: None
   intf.ppw.ics.cw.rnv.opener.auth_cache: {}
   intf.ppw.ics.cw.rnv.opener.key_file: None
   intf.ppw.ics.cw.rnv.opener.proxies: {}
   intf.ppw.ics.cw.rnv.opener.type: file
   intf.ppw.ics.cw.rnv.opener.maxtries: 10
 intf.ppw.ics.cw.rnv.anaconda: Already dumped
 intf.ppw.ics.cw.rnv.htmlheader: htmlheadmeta http-equiv=Content-Type content=text/html; 
charset=utf-8/headbody bgcolor=whitepre
 intf.ppw.ics.cw.rnv.htmlfooter: /pre/body/html
 intf.ppw.ics.cw.rnv.is_showing:

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Les Mikesell


Jussi Hirvi wrote:

Joost Waversveld [EMAIL PROTECTED] escribio (21.8.2008 16:30)

Try to find the file through the find command, something like  find /usr -t f
-name 'obj.Linux*' 

The command locate works with an database, which will be updated just once a
day (If you enable it!), so that will not find any just installed files...


Thanks - I know that about locate. That's why I commanded updatedb before I
tried the locate command. Locate is so easy to use, in general I like it
more than find. 


The stuff needed to build milters should be in the sendmail-devel 
package and if you have it installed. 'rpm -q --list sendmail-devel' 
should show where it landed.


--
  Les Mikesell
   [EMAIL PROTECTED]
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Xen bridged networking config

2008-08-21 Thread David Dyer-Bennet


On Thu, August 21, 2008 10:23, David Dyer-Bennet wrote:


 The problem is, I can't originate a connection to either guest from
 outside.

Solved my own problem.  And found I'd forgotten the choices, so maybe my
description didn't fully make sense either.

I was indeed running my guests behind NAT, and eventually found that
iptables -t nat -L would show the MASQUERADE target I believed was
there.  So I confirmed that I'd ended up, unintentionally, behind NAT.

By creating a new guest, I found that the networking choices were not what
I had remembered.  Clearly my NAT situation came from selecting virtual
networking.  When I instead told it to associate the network with a
physical device, I got my new guest connected to xenbr0 as I had expected
all along.

I had tried to change my guests over to connecting to xenbr0 previously,
and it turns out I got the syntax slightly wrong -- the two arguments in
the config file to vif should have been in a single set of quotes,
separated by commas, rather than two separate arguments each quoted.  When
I had the config file for the new guest that wasn't behind NAT to compare
to, that became obvious, and making the change to the config files changed
the way they were connected as expected.  And so now all three guests, the
two old ones and the new one, are externally visible, and get IPs in our
normal range from the DHCP server.  (Well, actually I've made another
change since then, and went in and attached the static internal IPs I'd
been assigned to them.)

I've still got virbr0 and the NAT setup sitting around vestigially; I
haven't looked closely at how to get rid of that.  Not important for now,
because this is a scratch install (as I said, my first time using Xen), so
next time when I do it right I won't have that left over.

To recap, and put the search terms closer together for future users, the
presence of virbr0 with the IP address 192.168.122.1, and a NAT setup, was
a consequence of selecting virtual networking for my original Xen guest
OSs.

Win!

Hope this helps somebody in the future.
-- 
David Dyer-Bennet, [EMAIL PROTECTED]; http://dd-b.net/
Snapshots: http://dd-b.net/dd-b/SnapshotAlbum/data/
Photos: http://dd-b.net/photography/gallery/
Dragaera: http://dragaera.info

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

OT Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Ric Moore


On Thu, 2008-08-21 at 16:38 -0500, Les Mikesell wrote:

 The stuff needed to build milters should be in the sendmail-devel 
 package and if you have it installed. 'rpm -q --list sendmail-devel' 
 should show where it landed.

Hot dog! Hey Les! Old home week here on the CentOS list. Ric


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

Re: [CentOS] Disk quotas for Sendmail

2008-08-21 Thread Florin Andrei


Jussi Hirvi wrote:


I would like to start using disk quotas on my Sendmail server, and optimally
they should be configurable by user or group. How can this be done? (I tried
to google, but no success.)


Storing email in users' home folders makes sense in many ways. After 
all, those too are their files, not anybody else's.


As an alternative, pretty close but not quite the same, I use Cyrus 
IMAPd, and keep the main Cyrus spool in /home/cyrus. This way, the /home 
volume is just users' files except I don't keep user johndoe's email 
in /home/johndoe/mail but in /home/cyrus/../johndoe
And Cyrus already has quota limits at the application level. Also has 
built-in filtering with Sieve. Faster, more powerful, more scalable than 
procmail  friends.


It's just a suggestion, I'm not saying this is the ideal solution for 
every scenario.


--
Florin Andrei

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

Re: [CentOS] iscsi and the last mile...

2008-08-21 Thread Jim Wildman


Joseph L. Casale wrote:

Yeah, you're missing the iscsi part.  Means the 'drives' are on a
different system on the other end of a network link.


Heh, thought you were trying to export those discs *on* the iSCSI server
and couldn't see them there, my bad :)

'iscsiadm -m discovery -p ip of target' wont give the target name, you need
'iscsiadm --mode discovery --type sendtargets --portal ip of target'

You also need to pass tcp 3260 through the firewall on the target, I have never
used RH's target, does it do this for you? Iet allows you to define 
allowed/denied
clients, does RH's?

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


You're right.  I grabbed the wrong copy of my notes.

You have to manually configure the firewall on the target.

--
--
Jim Wildman, CISSP, RHCE   [EMAIL PROTECTED] http://www.rossberry.com
Society in every state is a blessing, but Government, even in its best
state, is a necessary evil; in its worst state, an intolerable one.
Thomas Paine
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] postfix mysql_pgsql update?

2008-08-21 Thread John Thomas


Should postfix-2.3.3-2.el5.centos.mysql_pgsql be updated for:
https://rhn.redhat.com/errata/RHSA-2008-0839.html
and, if so, may I humbly request it?
--
Sincerely,
John Thomas
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Yum Issues with Dev groups

Just my luck,
None of my CentOS systems or my Fedora systems can #yum groupinstall 
Development Tools Development Libraries?
Both are having dep issues? Is that a coincidence, or what's going on?

The CentOS boxes all complain:

-- Processing Dependency: glibc-common = 2.5-18.el5_1.1 for package: glibc
Error: No Package Matching glibc.i686

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

Re: [CentOS] Yum Issues with Dev groups

2008-08-21 Thread Jim Perrin

On Thu, Aug 21, 2008 at 10:26 PM, Joseph L. Casale
[EMAIL PROTECTED] wrote:
 Just my luck,
 None of my CentOS systems or my Fedora systems can #yum groupinstall 
 Development Tools Development Libraries?
 Both are having dep issues? Is that a coincidence, or what's going on?

 The CentOS boxes all complain:

 -- Processing Dependency: glibc-common = 2.5-18.el5_1.1 for package: glibc
 Error: No Package Matching glibc.i686


There are two times when this becomes an issue. One is on x86_64
systems where build deps can cross architectures, and the other is
when using systems like openvz/virtuozzo where the glibc is often
replaced or otherwise lobotomized.

Which one is yours?

-- 
During times of universal deceit, telling the truth becomes a revolutionary act.
George Orwell
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

RE: [CentOS] Yum Issues with Dev groups