Re: [CentOS] need to free space on a root partition.

[Yan Yu]

thanks for the tip!
On Jan 1, 2010, at 11:13 PM, Barry Brimer wrote:

>
>
> On Fri, 1 Jan 2010, Yan Yu wrote:
>
>> Hello, there,   Happy new year everyone!
>>   I am puzzled here. Could some expert help me shed some light on
>> this?
>> this may not be a centos specific Q..
>> i have a linux machine. and the root partition (/) is full..
>> so I moved a dir from /data to /var (/var is on a different
>> partition), and created a link from /data/fa to /var/xyz, the files
>> under /var/xyz is about 8G,
>> du  /var/fa
>> 8701248 /var/fa
>>
>> so I thought this should free some space for me, but when I used df,
>> it gave me similar results before and after the move..   and my rpm
>> installation would fail due to the lack of space..
>>
>> Filesystem   1K-blocks  Used Available Use% Mounted on
>> /dev/cciss/c0d0p6 15235040  15232452 0 100% /
>> /dev/cciss/c0d0p8other partitions have free space
>>
>> I was confused on why moving a 8G file to another partition did not
>> free some space for me?
>> and I did see that the disk usage for the destination partition/disk
>> increases after the move..
>>
>> any pointer is appreciated!
>
> If you do an "lsof" and look for anything that says "deleted" that is
> something that has been deleted from disk but is still open  
> somewhere and
> the disk space can not be reclaimed until the process that has the  
> file
> open lets go of the file.  You also may have hard links to files in  
> other
> directories that will not allow the disk space to be freed as well.
>
> Barry
> ___
> 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] need to free space on a root partition.

[Barry Brimer]

On Fri, 1 Jan 2010, Yan Yu wrote:

> Hello, there,   Happy new year everyone!
>I am puzzled here. Could some expert help me shed some light on
> this?
> this may not be a centos specific Q..
> i have a linux machine. and the root partition (/) is full..
> so I moved a dir from /data to /var (/var is on a different
> partition), and created a link from /data/fa to /var/xyz, the files
> under /var/xyz is about 8G,
> du  /var/fa
> 8701248 /var/fa
>
> so I thought this should free some space for me, but when I used df,
> it gave me similar results before and after the move..   and my rpm
> installation would fail due to the lack of space..
>
> Filesystem   1K-blocks  Used Available Use% Mounted on
> /dev/cciss/c0d0p6 15235040  15232452 0 100% /
> /dev/cciss/c0d0p8other partitions have free space
>
> I was confused on why moving a 8G file to another partition did not
> free some space for me?
> and I did see that the disk usage for the destination partition/disk
> increases after the move..
>
> any pointer is appreciated!

If you do an "lsof" and look for anything that says "deleted" that is 
something that has been deleted from disk but is still open somewhere and 
the disk space can not be reclaimed until the process that has the file 
open lets go of the file.  You also may have hard links to files in other 
directories that will not allow the disk space to be freed as well.

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

[CentOS] need to free space on a root partition.

[Yan Yu]

Hello, there,   Happy new year everyone!
I am puzzled here. Could some expert help me shed some light on  
this?
this may not be a centos specific Q..
i have a linux machine. and the root partition (/) is full..
so I moved a dir from /data to /var (/var is on a different  
partition), and created a link from /data/fa to /var/xyz, the files  
under /var/xyz is about 8G,
du  /var/fa
8701248 /var/fa

so I thought this should free some space for me, but when I used df,  
it gave me similar results before and after the move..   and my rpm  
installation would fail due to the lack of space..

Filesystem   1K-blocks  Used Available Use% Mounted on
/dev/cciss/c0d0p6 15235040  15232452 0 100% /
/dev/cciss/c0d0p8other partitions have free space

I was confused on why moving a 8G file to another partition did not  
free some space for me?
and I did see that the disk usage for the destination partition/disk  
increases after the move..

any pointer is appreciated!
yan
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mutt execution by daemon

[Joseph L. Casale]

>You could specify the muttrc file using the -F switch, maybe that will
>solve yer problem? Or are you saying that is what you're doing but it
>bombs? How about telling us the command/script and the error output?
>luck,
>Dave

Hi,
I am calling mutt like so:
mutt -F /path/muttrc -a /path/file -s "text here" d...@example.com < /dev/null

But as the user does not have a mailbox, it fails. I can't actually see what's
going on obviously but as I tail the maillog and nothing even wakes it I am just
making a WASG that I am pretty sure is right:) As there is and never was local 
mail
delivery, root doesn't even have a mailbox so if I run it, I see it prompts for 
the
mb...

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

Re: [CentOS] mutt execution by daemon

[Dave]

On Fri, Jan 1, 2010 at 6:21 PM, Joseph L. Casale
 wrote:
> I have a user w/o a shell that runs an app as a daemon that needs to use
> mutt to email an attachment from a script.
>
> During testing when the app was ran as root everything worked:) Now that
> we obviously are running as a user w/o a shell this little bit broke. I
> used mutt as I needed to send the attachment as mime, and the ability
> to specify a muttrc from the cli when executed was useful so we could tweak
> the mail headers etc...



Hmmm.. I call mutt from scripts called by cron all the time, can't see
why this is much different. Well, okay, cron uses the same user, so it
has a shell.

You could specify the muttrc file using the -F switch, maybe that will
solve yer problem? Or are you saying that is what you're doing but it
bombs? How about telling us the command/script and the error output?
luck,
Dave
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] yum centos repo dependency hell for matlab loaddap

[Dave]

Oops, I hadn't intended to hit send, but thanks for the response.

On Thu, Dec 31, 2009 at 1:24 PM, Ryan J M  wrote:
> maybe you can try the old version of loaddap, it depends on libdap
> 3.6.2, according to their descriptions.

Good suggestion, I shoulda thought of it.

> Or compile the new libdap
> 3.9.3 by your own.

I have a strong preference for rpms, especially since once I figure
this out I'll probably need to install it on 11 other machines. Guess
I could try making my own.


Where did the info below come from? Ah, thank you google.
http://opendap.org/download/ml-structs.html

> 
>
> Matlab Structs Tool 3.5.2 (28 April  2006)
> Binaries
>
>  Win32
>    Win32 package
>  Mac OS/X
>    OS/X 10.4(Tiger) package
>    Intel Mac
>  Linux
>     ix386/ix86_64/Fedora Core 4 .
>  Linux
>     i386/RedHat Enterprise3 .
>
> Required packages
>
> The Matlab structs tool requires the libdap library, version 3.6.2.
> 
>
> On Thu, Dec 31, 2009 at 3:26 PM, Dave  wrote:
>> I've got matlab installed:
>>                                                                   < M
>> A T L A B (R) >
>>                                                         Copyright
>> 1984-2009 The MathWorks, Inc.
>>                                                       Version
>> 7.8.0.347 (R2009a) 64-bit (glnxa64)
>>
>> February 12, 2009
>>
>> I want to install loaddap. I downloaded an rpm from
>> http://opendap.org/download/ml-structs.html.
>>
>> # yum --nogpgcheck localinstall loaddap-3.6.1-1.x86_64.rpm
>> Loaded plugins: fastestmirror, priorities, security
>> Setting up Local Package Process[snip]
>> --> Processing Dependency: libmex.so()(64bit) for package: loaddap
>> --> Processing Dependency: libmx.so()(64bit) for package: loaddap
>> --> Finished Dependency Resolution
>> loaddap-3.6.1-1.x86_64 from loaddap-3.6.1-1.x86_64.rpm has depsolving 
>> problems
>>  --> Missing Dependency: libmex.so()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> loaddap-3.6.1-1.x86_64 from loaddap-3.6.1-1.x86_64.rpm has depsolving 
>> problems
>>  --> Missing Dependency: libdapclient.so.3()(64bit) is needed by
>> package loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> loaddap-3.6.1-1.x86_64 from loaddap-3.6.1-1.x86_64.rpm has depsolving 
>> problems
>>  --> Missing Dependency: libdap.so.9()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> loaddap-3.6.1-1.x86_64 from loaddap-3.6.1-1.x86_64.rpm has depsolving 
>> problems
>>  --> Missing Dependency: libmat.so()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> loaddap-3.6.1-1.x86_64 from loaddap-3.6.1-1.x86_64.rpm has depsolving 
>> problems
>>  --> Missing Dependency: libmx.so()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> loaddap-3.6.1-1.x86_64 from loaddap-3.6.1-1.x86_64.rpm has depsolving 
>> problems
>>  --> Missing Dependency: libdap >= 3.8.0 is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> Error: Missing Dependency: libdapclient.so.3()(64bit) is needed by
>> package loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> Error: Missing Dependency: libmx.so()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> Error: Missing Dependency: libdap >= 3.8.0 is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> Error: Missing Dependency: libdap.so.9()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> Error: Missing Dependency: libmat.so()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>> Error: Missing Dependency: libmex.so()(64bit) is needed by package
>> loaddap-3.6.1-1.x86_64 (loaddap-3.6.1-1.x86_64.rpm)
>>
>> Apparently my libdap is too old:
>> # rpm -q libdap
>> libdap-3.7.8-1.el5.1
>>
>> And it is current with the centos repos,
>> # yum update libdap
>> Loaded plugins: fastestmirror, priorities, security
>> [snip]
>> No Packages marked for Update
>>
>> so I need a good rev source rpm
>> ___
>> CentOS mailing list
>> CentOS@centos.org
>> http://lists.centos.org/mailman/listinfo/centos
>>
>
>
>
> --
> FIXME if it is wrong.
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



-- 
———-
Q: Why should this email be 5 sentences or less?
A: http://five.sentenc.es
IPRC-help FAQ: https://wailua/wiki/index.php/Faq
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] mutt execution by daemon

[Joseph L. Casale]

>This may or may not be useful to you, but I use this to handle email
>from scripts and programs:
>
>http://www.cleancode.org/projects/email

Now that you mention this, I recall you suggesting it on this or the fedora
list before and I recall the squabbling over its name and hence exclusion
from the package repos:)

I don't want to worry about adding a package as I want to update all my
scripts no production boxes to use something in base or at worst, rpmforge.

I don't know python...

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

Re: [CentOS] mutt execution by daemon

[Frank Cox]

On Sat, 2010-01-02 at 04:21 +, Joseph L. Casale wrote:
> I have a user w/o a shell that runs an app as a daemon that needs to
> use
> mutt to email an attachment from a script.

This may or may not be useful to you, but I use this to handle email
from scripts and programs:

http://www.cleancode.org/projects/email

-- 
MELVILLE THEATRE ~ Melville Sask ~ http://www.melvilletheatre.com

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

Re: [CentOS] mutt execution by daemon

[Ray Van Dolson]

On Sat, Jan 02, 2010 at 04:21:02AM +, Joseph L. Casale wrote:
> I have a user w/o a shell that runs an app as a daemon that needs to use
> mutt to email an attachment from a script.
> 
> During testing when the app was ran as root everything worked:) Now that
> we obviously are running as a user w/o a shell this little bit broke. I
> used mutt as I needed to send the attachment as mime, and the ability
> to specify a muttrc from the cli when executed was useful so we could tweak
> the mail headers etc...
> 
> Since there is no local mb's at all on this server, are there any mutt gurus
> that know off hand how I can manipulate the behavior of mutt to simply send
> w/o looking for a mailbox?

How do you call mutt?  Couldn't you precreate a muttrc file and then
just include its path in however you make your call to mutt?

Regardless, seems like you'd be better off using a Python or PHP script
to send your email

Ray

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

[CentOS] mutt execution by daemon

[Joseph L. Casale]

I have a user w/o a shell that runs an app as a daemon that needs to use
mutt to email an attachment from a script.

During testing when the app was ran as root everything worked:) Now that
we obviously are running as a user w/o a shell this little bit broke. I
used mutt as I needed to send the attachment as mime, and the ability
to specify a muttrc from the cli when executed was useful so we could tweak
the mail headers etc...

Since there is no local mb's at all on this server, are there any mutt gurus
that know off hand how I can manipulate the behavior of mutt to simply send
w/o looking for a mailbox?

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

Re: [CentOS] unsuscribe

[Jon Moore]

On Fri, Jan 1, 2010 at 5:51 PM,   wrote:
> unsuscribe
>

Have a look at the URL in the footer of the message
  You might have
better luck there.

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

[CentOS] Tyan s4985 motherboard and lm_sensors

[Tom Bishop]

Trying to get cpu temps from lm_sensors for a Tyan s4985 motherboard, just
installed centos 5.4.  Tyan release some lm_sensors conf files and other
information which I have done, but I am not getting any cpu temps.  I was
wondering if anyone out there was running this board and if you have gotten
temps working for your cpu, here is my output...Thanks in advance.

w83627hf-isa-0c00
Adapter: ISA adapter
VCore 1:   +1.52 V  (min =  +0.00 V, max =  +4.08 V)
VCore 2:   +1.55 V  (min =  +0.00 V, max =  +4.08 V)
+3.3V: +3.49 V  (min =  +2.82 V, max =  +3.79 V)
+5V:   +5.24 V  (min =  +6.40 V, max =  +5.54 V)   ALARM
+12V:  +5.78 V  (min = +12.28 V, max = +12.28 V)   ALARM
-12V:  -6.93 V  (min =  +6.06 V, max =  +5.65 V)   ALARM
-5V:   -3.84 V  (min =  +2.49 V, max =  +0.28 V)   ALARM
V5SB:  +5.73 V  (min =  +6.69 V, max =  +5.86 V)   ALARM
VBat:  +3.09 V  (min =  +1.66 V, max =  +0.82 V)   ALARM
fan1:0 RPM  (min = 10227 RPM, div = 2)  ALARM
fan2:0 RPM  (min = 6250 RPM, div = 2)  ALARM
fan3:0 RPM  (min = 2789 RPM, div = 2)  ALARM
temp1:   +41°C  (high =   -54°C, hyst =   -54°C)   sensor = thermistor
ALARM
temp2: +43.0°C  (high =   +80°C, hyst =   +75°C)   sensor = thermistor

temp3: +45.5°C  (high =   +80°C, hyst =   +75°C)   sensor = thermistor

vid:  +0.000 V  (VRM Version 0.0)
alarms:
beep_enable:
  Sound alarm enabled
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] which program/command (perl, vi, or sed) is better

[Les Bell]

ann kok  wrote:

>>
Which program/command (perl, vi, or sed) is better to extract the data by
date?
<<

The book "Perl for System Administration ", (David N. Blank-Edelman,
O'Reilly, 2000) has a chapter on using Perl to analyse log files, rotate
and compress them, etc. Luckily, this chapter is the "free sample" they
provide online - see
http://oreilly.com/catalog/perlsysadm/chapter/ch09.html

Sorted.

I mean, problem solved.

Best,

--- Les Bell
[http://www.lesbell.com.au]
Tel: +61 2 9451 1144


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

Re: [CentOS] which program/command (perl, vi, or sed) is better

[Néstor]

I have done similar programs in Perl (you can use PHP also) where, when I
read the file in and using regex
I select the records that I want to keep.

Feliz Año  :-)

On Fri, Jan 1, 2010 at 9:14 AM, Les Mikesell  wrote:

> ann kok wrote:
> > Hi
> >
> > I have apache log file around 7.6G and record half year
> >
> > Which program/command (perl, vi, or sed) is better to extract the data by
> date? and finally I can remove that big file and still keep the record.
> >
> > I have problem when using vi and it uses up the server memory
> >
> > Thank you for your help
> >
>
> If the extraction is simple regex matches you could use sed, but I'd
> recommend
> perl because it has additional features that you might need if the program
> becomes more complex and it is easier to write in the first place.
>
> But first I'd check to see if any of the available log processing programs
> will
> already do what you want.  If you are looking for summaries with counts by
> page/time interval/client IP, etc. they may do what you want.  I like
> analog
> because it is very fast and can deal with multiple files even if the times
> overlap and can uncompress them on the fly http://www.analog.cx/ or
> download rpm
> from http://www.iddl.vt.edu/~jackie/analog/.   Or you might like webalizer
> if
> you just have one server/file.
>
> --
>   Les Mikesell
>lesmikes...@gmail.com
> ___
> 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] unsuscribe

[mvaras]

unsuscribe

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

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Kai Schaetzl]

Are you sure you read the OP's message(s)?

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

Re: [CentOS] Lost mdadm.conf

[Thomas Harold]

On 12/31/2009 11:27 AM, James Bensley wrote:
>
> I can't say this with 100% certainty but I would of thought that it
> would been fine. I've lost my mdadm.conf (reinstalled OS) with a
> separate 4 disk RAID 5 array and re-assembled the array and carried on
> as if nothing had happened.
>

Yes, in general, you don't need the mdadm.conf at all.  As long as the 
array is built out of partitions marked as type "fd: Linux raid autodetect".

However, whenever CentOS installs a new kernel and initrd image file, it 
creates (or uses?) an mdadm.conf file within the initial boot 
environment.  Back when I was migrating a server to a new environment, I 
had to unpack the image, edit that copy of mdadm.conf, and then repack 
it all in order to get a proper boot.

So I suspect (but am not certain) that the ARRAY lines in 
/etc/mdadm.conf are useless on a CentOS system but that the ARRAY lines 
inside the initrd image file are the real ones used.  But the former may 
be used to generate the latter when you install a new kernel.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Agile Aspect]

>
> "machines" do not have MX records, domains have MX records.


MX records are associated with the domain but point to machines with
IP addresses running a MTA.

If you don't assign a IP address of a machine with a MTA to the MX
record, then your MX record is useless.

It's essential for fault tolerance to have more than path for your
mailer if you consider your email to be mission critical.


-- 
  Enjoy global warming while it lasts.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] The future is here, but Spamassassin doesn't buy it

[Fabian Arrotin]

Kai Schaetzl wrote:
> Gilbert Sebenste wrote on Fri, 1 Jan 2010 13:52:03 -0600 (CST):
> 
>> X-Spam-Status: Yes, score=5.2 required=5.0 tests=FH_DATE_PAST_20XX,
>>  SUBJ_ALL_CAPS autolearn=no version=3.2.5
>> X-Spam-Report:
>>  *  3.4 FH_DATE_PAST_20XX The date is grossly in the future.
> 
> add
> 
> score FH_DATE_PAST_20XX 0
> 
> to your /etc/mail/spamassassin/local.cf
> And you should use Bayes. A well-trained Bayes would have balanced that 
> down (assuming it was ham).
> 
> Kai

  *nod* : my spamassasin machine didn't block any mail today (except 
real spam) thanks to bayes. The FH_DATE_PAST_20XX scored up but bayes 
scored down on the other hand so no False Positive here ;-)

-- 
--
Fabian Arrotin
test -e /dev/human/brain || ( echo 1 > /proc/sys/kernel/sysrq ; echo c > 
/proc/sysrq-trigger )



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

Re: [CentOS] The future is here, but Spamassassin doesn't buy it

[Kai Schaetzl]

Gilbert Sebenste wrote on Fri, 1 Jan 2010 13:52:03 -0600 (CST):

> X-Spam-Status: Yes, score=5.2 required=5.0 tests=FH_DATE_PAST_20XX,
>  SUBJ_ALL_CAPS autolearn=no version=3.2.5
> X-Spam-Report:
>  *  3.4 FH_DATE_PAST_20XX The date is grossly in the future.

add

score FH_DATE_PAST_20XX 0

to your /etc/mail/spamassassin/local.cf
And you should use Bayes. A well-trained Bayes would have balanced that 
down (assuming it was ham).

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

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Kai Schaetzl]

Jerry Geis wrote on Fri, 01 Jan 2010 15:05:35 -0500:

> All three machines have MX records. they all have the same domain name.

"machines" do not have MX records, domains have MX records. I think you 
told your provider some wrong information and that's why they set it up 
wrong.

> 
> It is a big deal to request changes from the provider so I was hoping 
> there was a way
> to just tell the two new machines that incoming email to them just send 
> on over to the the
> first machine.
> 
> Is there a way to do that - or am I going about this the wrong way?

What you attempt to do is just stupid (no offense intended) if you do not 
want these machines to handle mail from external. You will just attract a 
lot of spam and bounce a lot of spam for non-existent addresses to 
innocent victims. 
As a first thing stop sendmail on them or firewall port 25 on them. Then 
have your provider set up the correct records. Or use one of the many free 
or cheap DNS providers. Anyone of them is surely better than RoadRunner.

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

Re: [CentOS] The future is here, but Spamassassin doesn't buy it

[Gilbert Sebenste]

On Fri, 1 Jan 2010, Gilbert Sebenste wrote:

> Hey all,
>
> I am getting a number of incoming emails flagged as spam this morning.
> Here's why:
>
> X-Spam-Level: *
> X-Spam-Status: Yes, score=5.2 required=5.0 tests=FH_DATE_PAST_20XX,
>SUBJ_ALL_CAPS autolearn=no version=3.2.5
> X-Spam-Report:
> *  3.4 FH_DATE_PAST_20XX The date is grossly in the future.
>
> The date of 2010 in the emails is correct. I have typed "sa-update" and 
> restarted, to no avail. Anyone else seeing this?

Yep, it's universal. They caught this 5 months ago then never pushed the
update.

https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6269

***
Gilbert Sebenste 
(My opinions only!)  **
Staff Meteorologist, Northern Illinois University  
E-mail: seben...@weather.admin.niu.edu  ***
web: http://weather.admin.niu.edu  **
***
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Les Mikesell]

Jerry Geis wrote:
>
> 
> All three machines have the same FQDN.
> something like A.mydomain.com, B.mydomain.com, C.mydomain.com
> 
> The IP numbers are X.Y.Z.170, .171 and .172
> 
> All three machines have MX records. they all have the same domain name.
> 
> It is a big deal to request changes from the provider so I was hoping 
> there was a way
> to just tell the two new machines that incoming email to them just send 
> on over to the the
> first machine.
> 
> Is there a way to do that - or am I going about this the wrong way?
> 

You can forward by setting MAIL_HUB but it may cause problems where you accept 
mail at the rely, then reject at the delivery host because the user name is 
wrong or from spam checkes.  In that case, the intermediate host is required to 
construct a bounce message and try to deliver it - and for spam it probably 
won't be deliverable.

If you have some reasonably small number of users, you can make an alias file 
that forwards each user to the fqdn of the delivery server.  Or you can use a 
virtuser table for the same purpose with a default reject rule.

Or, as someone else suggested, just stop sendmail or block port 25 on the one's 
you don't want to work.  Outside deliveries will retry each MX until one 
succeeds, but it is slightly antisocial to knowingly make them do this extra 
work.


-- 
   Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] one server handled sendmail added two more servers messes things up

[John R Pierce]

Jerry Geis wrote:
> MAIL_HUB does the trick - thanks that was what I was looking for.
>   

you still should delete the extra MX records from the DNS


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

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Jerry Geis]

MAIL_HUB does the trick - thanks that was what I was looking for.

jerry

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

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Michel van Deventer]

Hi,


> All three machines have the same FQDN.
> something like A.mydomain.com, B.mydomain.com, C.mydomain.com
No, they don't have the same fqdn, the fqdn includes the hostname.

> 
> The IP numbers are X.Y.Z.170, .171 and .172
> 
> All three machines have MX records. they all have the same domain name.
There's your error, the new one shouldn't have mx records for the domain
if they are not designed to accept mail for that domain.

> 
> It is a big deal to request changes from the provider so I was hoping 
> there was a way
> to just tell the two new machines that incoming email to them just send 
> on over to the the
> first machine.
> 
> Is there a way to do that - or am I going about this the wrong way?
There is a way but maybe there's a quicker (and dirtier) way. Just
reconfigure sendmail on the two new machines to *not* listen on the
public ip on the smtp port. Otherwise block port tcp/25 with iptables
(iptables -I INPUT -j REJECT -p tcp --dport 25)
But you should really get your DNS fixed.

regards,

Michel


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

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Jerry Geis]

lists-centos wrote:
> What IPnumbers did you give the two new machines? What does your DNS
> look like.
>
> >From the outside, mail will be delivered to the MX- (or if that
> doesn't exist A-) record machine (based on the IPnumber in the DNS
> record) for the FQDN on the message. 
>
>   - Rick
>
>  Original Message 
>   
>> Date: Friday, January 01, 2010 02:27:12 PM -0500
>> From: Jerry Geis 
>> To: CentOS ML 
>> Subject: [CentOS] one server handled sendmail added two more
>> 
> servers messes things up
>   
>> Hi all,
>>
>> I had one server running centos 4.7 i686  everything was fine with 
>> incoming mail.
>>
>> I now added two more servers centos 5.4 x86_64 and gave them
>> machine  names all
>> part of the same domain as machine 1. I noticed that incoming
>> emails are  being
>> round robined to all three machines.  At this time I really just
>> want  the one original machine
>> to handle my incoming email... The other two new machines are for
>> other  purposes.
>>
>> How do I tell the two new machines that all incoming email send to 
>> machine 1.
>> Is there an easy way to do that? At this point I just turned off
>> the two  new machines
>> and all mail is coming into the original machine just like before.
>>
>> Thanks,
>>
>> Jerry
>> 
>   



All three machines have the same FQDN.
something like A.mydomain.com, B.mydomain.com, C.mydomain.com

The IP numbers are X.Y.Z.170, .171 and .172

All three machines have MX records. they all have the same domain name.

It is a big deal to request changes from the provider so I was hoping 
there was a way
to just tell the two new machines that incoming email to them just send 
on over to the the
first machine.

Is there a way to do that - or am I going about this the wrong way?

Jerry

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

Re: [CentOS] The future is here, but Spamassassin doesn't buy it

[Fabian Arrotin]

Gilbert Sebenste wrote:
> Hey all,
> 
> I am getting a number of incoming emails flagged as spam this morning.
> Here's why:
> 
> X-Spam-Level: *
> X-Spam-Status: Yes, score=5.2 required=5.0 tests=FH_DATE_PAST_20XX,
>  SUBJ_ALL_CAPS autolearn=no version=3.2.5
> X-Spam-Report:
>  *  3.4 FH_DATE_PAST_20XX The date is grossly in the future.
> 
> The date of 2010 in the emails is correct. I have typed "sa-update" and 
> restarted, to no avail. Anyone else seeing this?

Yes and they haven't fixed it (yet) : 
https://issues.apache.org/SpamAssassin/show_bug.cgi?id=6269
Change your score for that rule, or disable it, or change it like 
described in the bug entry
I like the fact that they said it was "fixed" in svn some months ago, 
but it's still in svn and not in the $current release so  ... :/
Nothing than a sa-update will not fix when they decide to update the 
base rules

-- 
--
Fabian Arrotin
test -e /dev/human/brain || ( echo 1 > /proc/sys/kernel/sysrq ; echo c > 
/proc/sysrq-trigger )



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

[CentOS] The future is here, but Spamassassin doesn't buy it

[Gilbert Sebenste]

Hey all,

I am getting a number of incoming emails flagged as spam this morning.
Here's why:

X-Spam-Level: *
X-Spam-Status: Yes, score=5.2 required=5.0 tests=FH_DATE_PAST_20XX,
 SUBJ_ALL_CAPS autolearn=no version=3.2.5
X-Spam-Report:
 *  3.4 FH_DATE_PAST_20XX The date is grossly in the future.

The date of 2010 in the emails is correct. I have typed "sa-update" and 
restarted, to no avail. Anyone else seeing this?

***
Gilbert Sebenste 
(My opinions only!)  **
***
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] one server handled sendmail added two more servers messes things up

[Les Mikesell]

Jerry Geis wrote:
> Hi all,
> 
> I had one server running centos 4.7 i686  everything was fine with 
> incoming mail.
> 
> I now added two more servers centos 5.4 x86_64 and gave them machine 
> names all
> part of the same domain as machine 1. I noticed that incoming emails are 
> being
> round robined to all three machines.  At this time I really just want 
> the one original machine
> to handle my incoming email... The other two new machines are for other 
> purposes.
> 
> How do I tell the two new machines that all incoming email send to 
> machine 1.
> Is there an easy way to do that? At this point I just turned off the two 
> new machines
> and all mail is coming into the original machine just like before.

This should be controlled by the DNS MX record for the domain name.  Just set 
it 
to the machine(s) you want to receive mail.  If you want more than one to 
receive mail but forward it all to one for delivery, set MAIL_HUB in the 
sendmail.mc file on the ones that should forward (usually that's not necessary, 
though).
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] one server handled sendmail added two more servers messes things up

[Jerry Geis]

Hi all,

I had one server running centos 4.7 i686  everything was fine with 
incoming mail.

I now added two more servers centos 5.4 x86_64 and gave them machine 
names all
part of the same domain as machine 1. I noticed that incoming emails are 
being
round robined to all three machines.  At this time I really just want 
the one original machine
to handle my incoming email... The other two new machines are for other 
purposes.

How do I tell the two new machines that all incoming email send to 
machine 1.
Is there an easy way to do that? At this point I just turned off the two 
new machines
and all mail is coming into the original machine just like before.

Thanks,

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

Re: [CentOS] which program/command (perl, vi, or sed) is better

[Les Mikesell]

ann kok wrote:
> Hi
> 
> I have apache log file around 7.6G and record half year
> 
> Which program/command (perl, vi, or sed) is better to extract the data by 
> date? and finally I can remove that big file and still keep the record.
> 
> I have problem when using vi and it uses up the server memory
> 
> Thank you for your help
> 

If the extraction is simple regex matches you could use sed, but I'd recommend 
perl because it has additional features that you might need if the program 
becomes more complex and it is easier to write in the first place.

But first I'd check to see if any of the available log processing programs will 
already do what you want.  If you are looking for summaries with counts by 
page/time interval/client IP, etc. they may do what you want.  I like analog 
because it is very fast and can deal with multiple files even if the times 
overlap and can uncompress them on the fly http://www.analog.cx/ or download 
rpm 
from http://www.iddl.vt.edu/~jackie/analog/.   Or you might like webalizer if 
you just have one server/file.

-- 
   Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] which program/command (perl, vi, or sed) is better

[Frank Cox]

On Fri, 2010-01-01 at 08:45 -0800, ann kok wrote:
> Hi
> 
> I have apache log file around 7.6G and record half year
> 
> Which program/command (perl, vi, or sed) is better to extract the data by 
> date? 

Your question is a little too general.  You want to extract what portion
of the data, into what?

If you want a complete record, you apparently have one already.  If you
want to keep everything, just run it through bzip2 before storing it.

If you just want reports, look at something like webalizer.

If you want something "custom", you will have to decide exactly what you
want.  Perl or sed or grep could probably extract it, or you could make
a program in python or C or whatever.

But the first task is to define exactly what you want.

-- 
MELVILLE THEATRE ~ Melville Sask ~ http://www.melvilletheatre.com

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

Re: [CentOS] stock openjdk vs. epel

[Les Mikesell]

John R Pierce wrote:
> Les Mikesell wrote:
>> Given that it could have been trivial to include Sun Java ages ago, or at 
>> least 
>> not intentionally break the jpackage installation methods, I think Red Hat 
>> has 
>> done more damage to java than any other company and don't see that turning 
>> around even if they try at this late date.
>>   
> 
> no, it wasn't trivial due to primarily licensing reasons.  

As it turned out, all they had to do to get the license of their choice was to 
ask.  Sun is (was...) responsible for more open source code than anyone.  But, 
Red Hat distributed Netscape back when they didn't like the license.  And 
Debian 
included Sun Java in the base distribution once redistribution was allowed 
where 
Red Hat only put it in the update stream for paid subscribers.  And in any 
case, 
it would have been trivial to remain compatible with the jpackage nosrc rpms or 
include them so users could deal with the license requirements without any 
other 
hassles.

But the worst damage to java was from distributing a non-standard and mostly 
non-working version.  I doubt if that damage can be undone even if they are 
actually willing now to distribute something that makes the OS irrelevant and 
applications portable.

-- 
   Les Mikesell
lesmikes...@gmail.com

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

[CentOS] which program/command (perl, vi, or sed) is better

[ann kok]

Hi

I have apache log file around 7.6G and record half year

Which program/command (perl, vi, or sed) is better to extract the data by date? 
and finally I can remove that big file and still keep the record.

I have problem when using vi and it uses up the server memory

Thank you for your help


  __
Get a sneak peak at messages with a handy reading pane with All new Yahoo! 
Mail: http://ca.promos.yahoo.com/newmail/overview2/
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] kickstart and logins.def question

[Kwan Lowe]

Hello all:

Happy New Year to everyone and thank you for all the knowledge this past year.

I have  a hopefully simple question about kickstart.   In the
authconfig section I can enable ldap, credential caching, etc.. Using
the GUI tool there's an option to create the user home directories on
first login. The docs don't show a similar option for authconfig in
kickstart.  For now I'm using a sed script in the %post section of the
kickstart to replace the entry in /etc/login.defs, but was curious if
there is a standard method via a system utility to change this option.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Lost mdadm.conf

[James Bensley]

2010/1/1 Luciano Rocha :
> On Thu, Dec 31, 2009 at 04:40:56PM +, Joseph L. Casale wrote:
>> >Not all rescue environments have MAKEDEV or have it in $PATH, so this 
>> >should always work. ;)
>>
>> Really, on my hp's as well? I don't have any /dev/hdx or /dev/sdx on the 
>> ones with software raid...
>>
>> Nothing is easier than simply:
>> # mdadm --detail --scan --verbose > /etc/mdadm.conf
>>
>> And that actually *does* work in all environments...
>
> Does that create the /dev/md* devices? I mean, a *scan* shouldn't change
> the filesystem, should it? That's the problem that line tries to fix:
> that there's no mdX device in /dev...

I think after adding those details to your mdadm.conf and restarting
your /dev/md* devices would be created?

-- 
Regards,
James ;)

Charles de Gaulle  - "The better I get to know men, the more I find
myself loving dogs." -
http://www.brainyquote.com/quotes/authors/c/charles_de_gaulle.html
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Update via yum fails with "needs libdbus-1.so.3, this is not available."

[Akemi Yagi]

On Fri, Jan 1, 2010 at 5:20 AM, Bart van Kuik  wrote:
> Hi list,
>
> I am running CentOS release 5.2. When I did a "yum -y update", I got a
> large list of packages to be updated (360 packages). Then I got the
> following error:
>
> Total download size: 499 M
> Downloading Packages:
> Running rpm_check_debug
> ERROR with rpm_check_debug vs depsolve:
> Package xulrunner needs libdbus-1.so.3, this is not available.
>  cut about 10 similar lines 
>
> What can I do to make the update roll through?

Please try again by following the recommended procedure in the Release Notes:

http://wiki.centos.org/Manuals/ReleaseNotes/CentOS5.4#head-29511ff6659f6463d444feb92326ed2232fc8c08

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

[CentOS] Update via yum fails with "needs libdbus-1.so.3, this is not available."

[Bart van Kuik]

Hi list,

I am running CentOS release 5.2. When I did a "yum -y update", I got a
large list of packages to be updated (360 packages). Then I got the
following error:

Total download size: 499 M
Downloading Packages:
Running rpm_check_debug
ERROR with rpm_check_debug vs depsolve:
Package xulrunner needs libdbus-1.so.3, this is not available.
 cut about 10 similar lines 

What can I do to make the update roll through?

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

Re: [CentOS] stock openjdk vs. epel

[Mathieu Baudier]

> As luck would have it, I have copies of the java-1.6.0 b12 EPEL RPMS
> that were offered before Centos added java-1.6.0 b09 as an "upgrade"
> on my home page.

A lot of luck (or foresight...) indeed!

I used these old EPEL SRPMs + my experience of building the OpenJDK on
CentOS (see previous mails) in order to adapt the latest Fedora 12
OpenJdk SRPMs (Java 1.6.0 b16, using IcedTea 1.6).

It basically boils down to:
- remove visualvm and its netbeans dependency
- remove X11 patch
- workaround the plugin compilation issue (see previous mails) on
x86_64 (the EPEL SRPM was really helpful here)

You can download an SRPM from here:
http://www.argeo.org/linux/argeo-el/5/plus/SRPMS/
http://www.argeo.org/linux/argeo-el/5/plus/SRPMS/java-1.6.0-openjdk-1.6.0.0-33.b16.el5.argeo.1.src.rpm

And rebuild it with:
rpmbuild --rebuild java-1.6.0-openjdk-1.6.0.0-33.b16.el5.argeo.1.src.rpm
(after having properly set up your RPM build environment:
http://wiki.centos.org/HowTos/SetupRpmBuildEnvironment)

You need to have the EPEL repo installed.

I could build it and test it on x86_64 but not on i386 (I'm abroad
with my x86_64 laptop, and changing the --target did not work, I'll
have a look at it when I'm back in the office next week).

I'm currently uploading the x86_64 RPMs but my connection is very bad,
so it may take the whole day (or have to wait until next week as well
if it fails)

The spec file can be seen here:
https://www.argeo.org/svn/dependencies/trunk/org.argeo.dep.rpm/centos/java-1.6.0-openjdk/java-1.6.0-openjdk.spec

As well as the two original spec files I merged:
https://www.argeo.org/svn/dependencies/trunk/org.argeo.dep.rpm/centos/java-1.6.0-openjdk/java-1.6.0-openjdk-fedora.spec
https://www.argeo.org/svn/dependencies/trunk/org.argeo.dep.rpm/centos/java-1.6.0-openjdk/java-1.6.0-openjdk-epel.spec

(you will have to accept our autogenerated SSL certificate)

A few disclaimers/comments:
- free software, no warranty, etc.
- this upgrades the base java-1.6.0-openjdk package thus you are not
in line with upstream anymore if you install it (I am considering to
package a version which could be installed in parallel)
- this is my first serious SRPM hacking, so comments/critics from more
experienced people are welcome!
- we are upgrading our infrastructure so the links above are subject
to change within the next few months. I'll keep the list posted about
what becomes of this (either contributing it to a third party repo, or
host it if nobody wants it)
- the tests failed and I deactivated hem (just as the EPEL package did
at the time). At first sight, it has to do with X11, so maybe part of
the X11 patch is needed after all. I'll try to have a look someday
(lesser priority though since we don't do much AWT, so help would be
welcome here if it is deemed important)

Many thanks again for your help!

Cheers,

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