Re: [CentOS] Re: Hard disk recomendation for a software raid 5 array. Does Linux Software Raid support/interacts well with TLER enabled disks.

[Peter Kjellstrom]

On Saturday 28 July 2007, Johnny Hughes wrote:
> Alexander Georgiev wrote:
...
> > Ok, and should I mix disks from different vendors? I have read
> > somewhere that this augments the reliability of the array, lowering
> > the probability of 2 disks and more go down simultaneously?
>
> Not necessarily from different vendors as much as from different build
> lots, etc.
>
> The theory is ... items build from the same components at the same time
> and the same place should fail/EOL at about the same time (all things
> being equal).
>
> In practice, I have not seen that.

I havn't seen that either. But what I have seen is a raid controller acting up 
as a function of some random micro property of a specific drive model. So, I 
would very much not want more than one type of drive in a raid as that would 
double the amount of strangeness the controller would have to deal with.

/Peter


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

Re: [CentOS] typo on a Wiki page

[Lanny Marcus]

On 28 July 2007, Akemi Yagi wrote:
> A nice catch.  As you know, typos are a hallmark for some CentOS
> developers and an integral part of the CentOS project because they
> make this OS stand out.  :-)

I laughed, when I read that! I believe they want to make it as perfect
as they possibly can. Actually, I wondered about reporting that typo,
because it seems like serious nit picking. 

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

Re: [CentOS] memory query

[Lanny Marcus]

On 28 July 2007, simon wrote:

> I have recently installed centOS 5 on DELL pentium 2.7ghz (model
> optiplex GX270) and have 512 memory

I do not have a CentOS 5 box yet, Below is on a CentOS 4.4 box. A low
end Dell Dimension 2400 (Celeron 2.6 GHz, with 512 MB RAM and integrated
video and sound). I hope someone running CentOS 5.0 can reply to you!

Check out the manual for the Optiplex (and the BIOS settings) and the
Support area of Dell.com and maybe you will find out where the missing
RAM is used. There's also a Linux area, in the Dell Support forums.
Lanny

> [EMAIL PROTECTED] ~]$ free
>  total   used   free sharedbuffers cached
> Mem:514440 511260   3180  0  22444 266204
> -/+ buffers/cache: 222612 291828
> Swap:  1044184  01044184

> [EMAIL PROTECTED] ~]$ cat /proc/meminfo
> MemTotal:   514440 kB
> MemFree:  2856 kB
> Buffers: 22560 kB
> Cached: 266208 kB
> SwapCached:  0 kB
> Active: 235784 kB
> Inactive:   180100 kB
> HighTotal:   0 kB
> HighFree:0 kB
> LowTotal:   514440 kB
> LowFree:  2856 kB
> SwapTotal: 1044184 kB
> SwapFree:  1044184 kB
> Dirty:   8 kB
> Writeback:   0 kB
> Mapped: 165624 kB
> Slab:23660 kB
> CommitLimit:   1301404 kB
> Committed_AS:   521484 kB
> PageTables:   2696 kB
> VmallocTotal:   499704 kB
> VmallocUsed:  3720 kB
> VmallocChunk:   495604 kB
> HugePages_Total: 0
> HugePages_Free:  0
> Hugepagesize: 4096 kB

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

[CentOS] Higher memory requirements for 64 bit?

[Steve Bergman]

I have a dual Xeon 3.2GHz XDMCP server serving out about 50 Gnome
desktops to thin clients.  I currently am running 32 bit with 4GB of
memory.  Performance is fine, normally, and the machine runs with about
3GB in swap.  However, if a user app suddenly sucks up the maximum
allowed rss and virtual or 256MB/process as specified by ulimit, the
resulting swap drags the machine down to the point that people think it
is locked up.

I have another 4GB of memory on order.  But I suspect that part of the
problem may be running out of memory in zone-normal, which I know will
cause swap storms. (An oops in the log file suggests that is where
memory was tight.)  I had low memory protection set to 150MB, and have
increased that to 256MB. (I had to increase it to at least 135MB
previously, else dd if=/dev/zero of=/tempfile.tmp would start a swap
storm that would bring the machine to its knees.)

More users will be added to this machine, and I am worried about having
an ever higher percentage of this (16GB maximum) machine in zone high.

So I am considering moving it from 32 bit CentOS4 to 64 bit CentOS5 or
64 bit Fedora 7.  But a cursory test of the 32 and 64 bit versions of
firefox suggest to me that, just opening them up and going to
google.com, the 64 bit version uses about 50% more memory than the 32
bit version.

Could someone please comment on how much of an additional memory hit I
might expect?  The VM risks and benefits that I might be taking?
Really, *any* commentary about this scenario would be appreciated.

Thanks,
Steve Bergman

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

Re: [CentOS] Re: Hard disk recomendation for a software raid 5 array. Does Linux Software Raid support/interacts well with TLER enabled disks.

[Patrick - South Valley Internet]

Peter Kjellstrom wrote:

On Saturday 28 July 2007, Johnny Hughes wrote:
  

Alexander Georgiev wrote:


...
  

Ok, and should I mix disks from different vendors? I have read
somewhere that this augments the reliability of the array, lowering
the probability of 2 disks and more go down simultaneously?
  

Not necessarily from different vendors as much as from different build
lots, etc.

The theory is ... items build from the same components at the same time
and the same place should fail/EOL at about the same time (all things
being equal).

In practice, I have not seen that.



I havn't seen that either. But what I have seen is a raid controller acting up 
as a function of some random micro property of a specific drive model. So, I 
would very much not want more than one type of drive in a raid as that would 
double the amount of strangeness the controller would have to deal with.


/Peter
  



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



No virus found in this incoming message.
Checked by AVG Free Edition. 
Version: 7.5.476 / Virus Database: 269.10.23/924 - Release Date: 7/28/2007 3:50 PM
  
Yeah don't mix the drive manufacturers - you are more prone to faults or 
"uknowns" because you are mixing two different drives from two different 
vendors.  Always stick with the same brand, and try to get them all at once.


And, if you can, try your hardest to think about shelling out the extra 
cash for a hardware raid controller.  If you are serious about your 
data, the extra $200 for an LSI Logic MegaRAID 150 4/6 port controller 
is a small investment for your data. 

Also, if this is for a Postfix server, you're going to want RAID10, if 
you can.  It's much faster for read/write access than RAID5.


HTH

Patrick


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

[CentOS] typo on a Wiki page

[R P Herrold]

On Sat, 28 Jul 2007, Akemi Yagi wrote:


A nice catch.  As you know, typos are a hallmark for some CentOS
developers and an integral part of the CentOS project because they
make this OS stand out.  :-)


hey -- I resemble that

-- Russ Herrold
orc_orc on IRC
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] typo on a Wiki page

[Akemi Yagi]

On 7/29/07, R P Herrold <[EMAIL PROTECTED]> wrote:
> On Sat, 28 Jul 2007, Akemi Yagi wrote:
>
> > A nice catch.  As you know, typos are a hallmark for some CentOS
> > developers and an integral part of the CentOS project because they
> > make this OS stand out.  :-)
>
> hey -- I resemble that
>
> -- Russ Herrold
> orc_orc on IRC

I was hoping you wouldn't see this post :P

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

[CentOS] Unsubscribe Me Please

[Robert J. Thompson]

Please unsubscribe me from the Centos mailing list.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Re: Mixing RPMforge and EPEL (Was: EPEL repo)

[Rex Dieter]

Dag Wieers wrote:

> On Fri, 27 Jul 2007, JC Júnior wrote:
> 
>> I received a message about EPEL repository, I would like to know if this
>> repo is long term support too.
> 
> Let me add that an effort to make sure EPEL is compatible with RPMforge
> failed as EPEL wants to become the only repository for RHEL and there is
> no interest to consider current RPMforge users.
...
> EPEL refused the repotag, so one cannot easily identify where a package
> comes from and mixing repositories becomes harder. Since compatibility is
> a 2 way interaction and EPEL shows no interest, it is certain that mixing
> EPEL with other repositories may break something.

Interesting you say that.  It's quite a stetch from "no repotags" to
conclude "EPEL has no interest" in compatibility.

In fact, epel (and fedora) repo is, by design and policy, supposed to be
compatible and considerate of other repos, e.g. most notably,
http://fedoraproject.org/wiki/PackageMaintainers/RepositoryCollaboration
(among other project policy documents).

When I posted the aforementioned repository collaboration document to the
rpmforge list(s) for comment, it received none.  

-- Rex


-- Rex

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

Re: [CentOS] Unsubscribe Me Please

[Alexander Dalloz]

Robert J. Thompson schrieb:

> Please unsubscribe me from the Centos mailing list.


Mail header content:

List-Unsubscribe: ,


Your turn to do it yourself in one of the 2 named fashions.

Alexander

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

[CentOS] Re: Mixing RPMforge and EPEL (Was: EPEL repo)

[Rex Dieter]

drew einhorn wrote:

> Dumb question.
> 
> Can't we identify the source of the package by looking at the signature.

Signature, vendor, etc... right.  Pretty much why epel (so far) didn't see
the need/value in the complexity/overhead of introducing repotags.

That said, my personal opinion is/was that epel *should* use them, if for no
other reason so that others can't use it to support arguments of
non-cooperation (too late).

-- Rex

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

Re: [CentOS] CentOS 3.9's up2date missing sources???

[Tru Huynh]

On Sat, Jul 28, 2007 at 07:16:00PM -0400, Lawrence Houston wrote:
> CentOS3.9 Users:
> 
> After applying all Updates to Upgrade from CentOS 3.8 CentOS 3.9 I find 
> up2date no longer works "correctly" (or shall I say NOT as before)... 
> Missing is the "sources" File!!!  Is this how V 4.5.5-7 is supposed to be 
> or is the "sources" File omission an "error"???

My bad, sorry.

The current up2date version is missing:
- a definition for newly installed up2date 
- a config(noreplace) tag for /etc/sysconfig/rhn/sources

I am pushing now a corrected version on the mirror.
i386:
up2date-gnome-4.5.5-7.centos.2.i386.rpm
up2date-4.5.5-7.centos.2.i386.rpm
x86_64:
up2date-gnome-4.5.5-7.centos.2.x86_64.rpm
up2date-4.5.5-7.centos.2.x86_64.rpm

howto fix:
--
If it's update from 3.8, you should have a /etc/sysconfig/rhn/sources.rpmsave
file just copy it to /etc/sysconfig/rhn/sources to fix up2date.  Or use yum
instead of up2date ;).

If up2date is broken due to the missing sources file, you can still use yum
to update up2date:
# yum update up2date

Last resort, I have attached the default /etc/sysconfig/rhn/sources, just
copy it there.

Once again, sorry for the error, and thanks for reporting it. :)

Tru
-- 
Tru Huynh (CentOS-3 i386/x86_64 Package Maintenance)
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xBEFA581B
### This describes the various package repositories (repos) that up2date will
### query for packages. It currently supports apt-rpm, yum, and "dir" repos.

### Format is one repository (repo) entry per line, # starts comments, the
### first word on each line is the type of repo.

### The default RHN (using "default" as the url means use the one in the 
### up2date config file). This is required.
#up2date default

### Note: when a channel label is required for the non up2date repos,
### the label is solely used as an internal identifier and is not
### based on the url or any other info from the repos.

### An apt style repo (the example is arjan's 2.6 kernel repo).
### The format is:
### type  channel-label  service:server pathrepo name
#apt arjan-2.6-kernel-i386 http://people.redhat.com ~arjanv/2.5/ kernel

### Note: for apt repos, there can be multiple repo names specified (space
### seperated).

### A yum style repo. The format is:
### type  channel-label url
#yum fedora-9-i386-stable 
http://download.fedora.us/fedora/redhat/9/i386/yum/stable
yum centos-3-base http://mirror.centos.org/centos/3/os/i386/
yum centos-3-addons http://mirror.centos.org/centos/3/addons/i386/
yum centos-3-updates http://mirror.centos.org/centos/3/updates/i386/
yum centos-3-extras http://mirror.centos.org/centos/3/extras/i386/
#yum centos-3-extras http://mirror.centos.org/centos/3/testing/i386/
#yum centos-3-centos+ http://mirror.centos.org/centos/3/centosplus/i386/

### A local directory full of packages (a "dir" repo). For example:
#dir my-favorite-rpms-i386-9 /var/spool/RPMS/

# Multiple versions of all repos except "up2date" can be used. Dependencies
# can be resolved "cross-repo" if need be.



pgpU2HUlQjMQR.pgp
Description: PGP signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Mixing RPMforge and EPEL (Was: EPEL repo)

[Lanny Marcus]

On 28 July 2007, Dag Wieers wrote:

> Let me add that an effort to make sure EPEL is compatible with
> RPMforge failed as EPEL wants to become the only repository for RHEL
> and there is no interest to consider current RPMforge users.
> 
> EPEL refused the repotag, so one cannot easily identify where a
> package comes from and mixing repositories becomes harder. Since
> compatibility is a 2 way interaction and EPEL shows no interest, it is
> certain that mixing EPEL with other repositories may break something.


Dag: I hope you will consider rewriting that post to the ML and putting
it into the CentOS Wiki. Rewriting may not be necessary. Your post was
pretty explicit and very informative.  IMHO, it will be a great addition
to the Repository Wiki. Lanny

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

[CentOS] MRTG with 14all.cgi on centos 5

[Indunil Jayasooriya]

Hi,

I installed mrtg on centos5 with 14all.cgi script.

I downloaded it from below URL.

http://my14all.sourceforge.net/14all-1.1.txt and copies it as 14all.cgi @
/var/www/cg-bin

I made a few changes to the script. Those changes can be seen in BOLDletteres.


# if MRTG_lib.pm (from mrtg) is not in the module search path (@INC)
# uncomment the following line and change the path appropriatly:
#use lib qw(/usr/local/mrtg-2/lib/mrtg2);
use lib qw(/usr/lib/mrtg2);

# if RRDs (rrdtool perl module) is not in the module search path (@INC)
# uncomment the following line and change the path appropriatly
# or use a LibAdd: setting in the config file
#use lib qw(/usr/local/rrdtool-1.0.38/lib/perl);
use lib qw(/usr/lib/perl5/vendor_perl/5.8.8/i386-linux-thread-multi);

### where the mrtg.cfg file is
# anywhere in the filespace
#$cfgfile = '/home/mrtg/mrtg.cfg';
#$cfgfile = '/etc/mrtg/mrtg.cfg';
# relative to the script
$cfgfile = ' mrtg.cfg';
# use this so 14all.cgi gets the cfgfile name from the script name
#(14all.cgi -> 14all.cfg)
#$cfgfile = 'mrtg.cfg';


It works fine. But When this server comes up , after rebooting,

the traffic of the First graph (i.e Daily' graph (5 Minute Average) goes up,
both incoming and outgoing traffic go up to 50-60 Mbits per secrond.

This happened, Whenever the system comes up after a reboot.

Any idea to solve this issue?

Unless it reboots, It works fine.

Also, I want to know, Where is this mtrg date stored? I want to back up
data. If the system crashes one day, then, I will be able to retrieve it.


-- 
Thank you
Indunil Jayasooriya
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos