Re: [CentOS] ClamAV help needed - Related problem

[Anne Wilson]

On Friday 20 June 2008 09:45:16 Anne Wilson wrote:
> On Friday 20 June 2008 06:13, Martin Garcia wrote:
> > Hi Anne, I have many servers running clamav, simply use the dag repos
> > http://dag.wieers.com/rpm/ and install it via yum "yum install clamav
> > clamd" then run freshclam, thats it. I presume your configuration is not
> > properly done.
>
> It turned out to be not exactly a clamav problem, but a clamtk problem.  I
> am using Dag's package, and logwatch has been telling me that everything is
> updating.  The problem showed up when I tried the context menu scan of a
> file, which clamtk provides.  I contacted the clamtk developer who has been
> very helpful.  This morning I have confirmed to him that the problem is
> fixed.
>
> For the sake of the archives, this is what he said:
> 
> Ok, so it dawned on me what the problem likely is... ClamAV has
> several methods for signatures it uses: daily.info folder, daily and
> main.cvd, daily and main.cld. So, I'm thinking you have more than one
> of those in your signatures directory.
>
> If you open up a terminal window and type "ls /var/clamav" (without
> quotes of course), I'm betting you'll see a variety of files and/or
> directories in there. If you're up for it, as root type
> rm /var/clamav/* -rf
> which will remove all the signatures. Don't worry, you'll get them
> back in the next step. As root, type
> freshclam -v
> And that will download all the necessary signatures again.
>
> The problem I have is there are a variety of ways the linux distros
> package ClamAV, and I have to decide which ones to gather the
> information from... I thought I had it right, but your email is making
> me reconsider. :)
> 
>
> I do wonder if something changed in clamav since I first installed it.  I
> remember that an update installed, and I did, for a couple of days, get a
> message that the database could not be notified of updated signatures.  I
> can't remember how that one got resolved.  Perhaps /var/clamav was left
> with an old version and a new version of the database, and was reading the
> old one.
>
This morning I wanted to make a change in the server's BIOS, so I had to 
reboot, and hit problems.  Bootup got to 'self-checking in 1800 seconds' and 
appeared to hang there.  I did leave it for some considerable time, but it 
didn't move on.  Eventually I used ssh to kill clamd and the boot continued.

I will be away from home for several days, so I need to get this sorted.  I 
presume that it is a mis-configuration somewhere that's causing it.  Can 
someone please advise me what to look for?  Thanks

Anne



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] how do I manage a Dell PE 860 server via IPMI?

[nate]

Rudi Ahlers wrote:

> Cool, this worked from my Windows Laptop to the Dell server, but I see
> it's only a very basic info type setup. Do you run any specific software
> on your Linux server to give you more control?
>
> I was actually hoping that I could use this as a KVM over IP type thing
> to manage the server if there's any OS / software problems, but it
> doesn't seem like it could do that.

Most IPMI implimentations support serial over LAN, basically a virtual
serial port for a serial console. For a linux system this works fine,
you just need to figure out which serial port to point your system
to. Basically the same as KVM over IP just text mode only.

Some management systems have additional options which include full
VGA KVM abilities, I have a demo Dell box with DRAC 5 which has this
ability though I haven't been able to get it to work. HP systems
have this ability as well but require an additional ~$300 license
add on(iLO Advanced), whereas remote serial console is free. That
and the remote KVM stuff usually requires a browser or a special
client, whereas remote serial console can be done in text mode
(the before mentioned system running DRAC5 allows me to SSH into
it and connect to the serial console that way).

nate

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

Re: [CentOS] error

[Alexander Dalloz]

[EMAIL PROTECTED] schrieb:

hi
this is my named.conf


//
// named.conf for Red Hat caching-nameserver
//

options {
directory "/var/named";
dump-file "/var/named/data/cache_dump.db";
statistics-file "/var/named/data/named_stats.txt";
/*
 * If there is a firewall between you and nameservers you want
 * to talk to, you might need to uncomment the query-source
 * directive below.  Previous versions of BIND always asked
 * questions using port 53, but BIND 8.1 uses an unprivileged
 * port by default.
 */
 // query-source address * port 53;
//  forward first;
//  forwarders {
//   172.18.1.212;
//  };
forward first;
forwarders {
 203.193.139.150;
};
};

//
// a caching only nameserver config
//
controls {
inet  127.0.0.1 allow { localhost; } keys { rndckey; };
};

key "rndckey" {
algorithm hmac-md5;
secret "8Ik4b9wwIirsJU0AVy5gy29V3F7VpEpNV6TqlEZVz6OSCoOMAb";
 };

zone "." IN {
type hint;
//file "named.ca";
file "named.root";
};

zone "localdomain" IN {
type master;
file "localdomain.zone";
allow-update { none; };
};
zone "localhost" IN {
type master;
file "localhost.zone";
allow-update { none; };
};

zone "0.0.127.in-addr.arpa" IN {
type master;
file "named.local";
allow-update { none; };
};

zone
"0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.ip6.arpa"
IN {
type master;
file "named.ip6.local";
allow-update { none; };
};

zone "255.in-addr.arpa" IN {
type master;
file "named.broadcast";
allow-update { none; };
};

zone "0.in-addr.arpa" IN {
type master;
file "named.zero";
allow-update { none; };
};

zone "suresh.com" IN {
type master;
file "open-ims.dnszone";
notify no;
};



this is my rndc.conf

key "rndckey" {
algorithm hmac-md5;
secret "8Ik4b9wwIirsJU0AVy5gy29V3F7VpEpNV6TqlEZVz6OSCoOMAb";
 };

options {
default-key "rndckey";
default-server 127.0.0.1;
default-port 953;
};



this is the log message
# named -g -p 53
21-Jun-2008 07:19:35.630 starting BIND 9.3.3rc2 -g -p 53
21-Jun-2008 07:19:35.631 found 1 CPU, using 1 worker thread
21-Jun-2008 07:19:35.635 loading configuration from '/etc/named.conf'
21-Jun-2008 07:19:35.637 listening on IPv4 interface lo, 127.0.0.1#53
21-Jun-2008 07:19:35.639 listening on IPv4 interface eth0, 127.0.0.1#53
21-Jun-2008 07:19:35.640 could not configure root hints from 'named.root':
file not found
21-Jun-2008 07:19:35.641 loading configuration: file not found
21-Jun-2008 07:19:35.641 exiting (due to fatal error)

this is the error

# rndc reload
rndc: decode base64 secret: bad base64 encoding


Different errors now, compared with your previous postings.

> 21-Jun-2008 07:19:35.640 could not configure root hints from 
>'named.root': file not found


> //file "named.ca";
> file "named.root";

Create the file if you have it renamed or deleted.

> rndc: decode base64 secret: bad base64 encoding

Build up a valid secret key. Use `rndc-confgen -a' for this.

Alexander


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

[CentOS] CentOS-announce Digest, Vol 40, Issue 9

[centos-announce-request]

Send CentOS-announce mailing list submissions to
[EMAIL PROTECTED]

To subscribe or unsubscribe via the World Wide Web, visit
http://lists.centos.org/mailman/listinfo/centos-announce
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]

You can reach the person managing the list at
[EMAIL PROTECTED]

When replying, please edit your Subject line so it is more specific
than "Re: Contents of CentOS-announce digest..."


Today's Topics:

   1. CESA-2008:0556 Important CentOS 3 i386 freetype - security
  update (Tru Huynh)
   2. CESA-2008:0556 Important CentOS 3 x86_64 freetype - security
  update (Tru Huynh)
   3. CESA-2008:0556 Important CentOS 3 ia64 freetype - security
  update (Pasi Pirhonen)
   4. CESA-2008:0556 Important CentOS 4 ia64 freetype - security
  update (Pasi Pirhonen)
   5. CESA-2008:0556 Important CentOS 3 s390(x) freetype - security
  update (Pasi Pirhonen)
   6. CESA-2008:0556 Important CentOS 4 s390(x) freetype - security
  update (Pasi Pirhonen)


--

Message: 1
Date: Fri, 20 Jun 2008 17:05:03 +0200
From: Tru Huynh <[EMAIL PROTECTED]>
Subject: [CentOS-announce] CESA-2008:0556 Important CentOS 3 i386
freetype -  security update
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"

CentOS Errata and Security Advisory CESA-2008:0556

freetype security update for CentOS 3 i386:
https://rhn.redhat.com/errata/RHSA-2008-0556.html

The following updated file has been uploaded and is currently syncing to
the mirrors:

i386:
updates/i386/RPMS/freetype-2.1.4-8.el3.i386.rpm
updates/i386/RPMS/freetype-devel-2.1.4-8.el3.i386.rpm

addons/i386/RPMS/freetype-demos-2.1.4-8.el3.i386.rpm
addons/i386/RPMS/freetype-utils-2.1.4-8.el3.i386.rpm

source:
updates/SRPMS/freetype-2.1.4-8.el3.src.rpm

You may update your CentOS-3 i386 installations by running the command:

yum update freetype

Tru
-- 
Tru Huynh (mirrors, CentOS-3 i386/x86_64 Package Maintenance)
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xBEFA581B
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : 
http://lists.centos.org/pipermail/centos-announce/attachments/20080620/0b1dcee1/attachment-0001.bin

--

Message: 2
Date: Fri, 20 Jun 2008 17:06:34 +0200
From: Tru Huynh <[EMAIL PROTECTED]>
Subject: [CentOS-announce] CESA-2008:0556 Important CentOS 3 x86_64
freetype- security update
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"

CentOS Errata and Security Advisory CESA-2008:0556

freetype security update for CentOS 3 x86_64:
https://rhn.redhat.com/errata/RHSA-2008-0556.html

The following updated file has been uploaded and is currently syncing to
the mirrors:

x86_64:
updates/x86_64/RPMS/freetype-2.1.4-8.el3.i386.rpm
updates/x86_64/RPMS/freetype-2.1.4-8.el3.x86_64.rpm
updates/x86_64/RPMS/freetype-devel-2.1.4-8.el3.x86_64.rpm

addons/x86_64/RPMS/freetype-demos-2.1.4-8.el3.x86_64.rpm
addons/x86_64/RPMS/freetype-utils-2.1.4-8.el3.x86_64.rpm

source:
updates/SRPMS/freetype-2.1.4-8.el3.src.rpm

You may update your CentOS-3 x86_64 installations by running the command:

yum update freetype

Tru
-- 
Tru Huynh (mirrors, CentOS-3 i386/x86_64 Package Maintenance)
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xBEFA581B
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : 
http://lists.centos.org/pipermail/centos-announce/attachments/20080620/71f5de91/attachment-0001.bin

--

Message: 3
Date: Fri, 20 Jun 2008 22:41:14 +0300
From: Pasi Pirhonen <[EMAIL PROTECTED]>
Subject: [CentOS-announce] CESA-2008:0556 Important CentOS 3 ia64
freetype -  security update
To: [EMAIL PROTECTED]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="us-ascii"

CentOS Errata and Security Advisory 2008:0556

https://rhn.redhat.com/errata/RHSA-2008-0556.html

The following updated files have been uploaded and are currently
syncing to the mirrors:

ia64:
updates/ia64/RPMS/freetype-2.1.4-8.el3.ia64.rpm
updates/ia64/RPMS/freetype-demos-2.1.4-8.el3.ia64.rpm
updates/ia64/RPMS/freetype-devel-2.1.4-8.el3.ia64.rpm
updates/ia64/RPMS/freetype-utils-2.1.4-8.el3.ia64.rpm


-- 
Pasi Pirhonen - [EMAIL PROTECTED] - http://pasi.pirhonen.eu/
Top-postings silently ignored
-- next part --
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : 
http://lists.centos.org/pipermail/centos-announce/attachments/20080620/fa164481/attachment-0001.bin

--

Message: 4
Date: Fr

[CentOS] Re: centos 4.6 - 586 install - how to get that to a 486 level if possible

[Jerry Geis]

Sorry, I have no clue about your question. However, I have several
AMD K6 chips (pentium-equivalent) lying around should you trip over
a motherboard that'll run them. That would solve your 486 problem, I
think, so let me know if such a chip would help you out.
  
Fred - thanks for the offer - however I do have a specific device I am 
targeting. Ebox2300sx.
I have no options on the hardware. It would be a great solution hardware 
wise but software is

being a bit of a pain at this time.

Johnny - thanks for the info. I actually have debian booted on the 
device but then EVERYTHING
is so much different on debian than centos. I was also trying a redhat 9 
install and put a 2.6 kernel on it.
However, I realize this is not the best solution. BUT - to try it out 
and get something working with
the other programs I need I thought that might be a valid short term 
path to check things out.


Thanks guys,

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

Re: [CentOS] how do I manage a Dell PE 860 server via IPMI?

[John R Pierce]

nate wrote:

Some management systems have additional options which include full
VGA KVM abilities, I have a demo Dell box with DRAC 5 which has this
ability though I haven't been able to get it to work. HP systems
have this ability as well but require an additional ~$300 license
add on(iLO Advanced), whereas remote serial console is free. That
and the remote KVM stuff usually requires a browser or a special
client, whereas remote serial console can be done in text mode
(the before mentioned system running DRAC5 allows me to SSH into
it and connect to the serial console that way).
  




DRAC 4 or 5 remote console shoudl work just fine, I've used it quite 
regularlly... the client needs a web browser with java 1.4 or 1.5 
enabled as they use a java VNC implementation to 'remote' the ATI 
VGA.  this gives you a remote CD and floppy as well as graphical 
VGA.   the remote CD can either be the client systems actual CD drive, 
or it can be an ISO file.


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

[CentOS] recommendations for copying large filesystems

[Mag Gam]

I need to copy over 100TB of data from one server to another via network.
What is the best option to do this? I am planning to use rsync but is there
a better tool or better way of doing this?

For example, I plan on doing
rsync -azv /largefs /targetfs

/targetfs is a NFS mounted filesystem.

Any thoughts?

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

Re: [CentOS] recommendations for copying large filesystems

[John R Pierce]

Mag Gam wrote:
I need to copy over 100TB of data from one server to another via 
network. What is the best option to do this? I am planning to use 
rsync but is there a better tool or better way of doing this?


For example, I plan on doing
rsync -azv /largefs /targetfs

/targetfs is a NFS mounted filesystem.

Any thoughts?


rsync would probably work better if you ran it in client server mode 
rather than over NFS, especially if you have to restart it.

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

Re: [CentOS] how do I manage a Dell PE 860 server via IPMI?

[Les Mikesell]

Rudi Ahlers wrote:

>
So, how do I actually access these servers remotely? If possible, I'd 
like todo it via another CentOS server at the DC using SSH.


Dell's usually come with their openmange suite, so you might google 
for that.  I think a java-enabled browser should connect and get an 
applet that looks more or less like vnc for console access.  But you 
only need it when installing remotely or if something is wrong so you 
can't ssh to the main interface.


Dell usually has all their documentation and updates online. If this 
doesn't help you can probably find the full manual.

 ftp://ftp.us.dell.com/sysman/readme_160_A00.txt

Thanx, that link was very helpful. I see they mention the word KVM, but 
I doubt it will give me a full "Keyboard / Video / Mouse" as a KVMoverIP 
device would, or am I wrong? I would basically like to access the server
remotely, as with a KVMoverIP device if there's any problems with the 
OS, like a kernel update that ran into problems, or something like that. 
Is it possible with these, out of your experience, or not?


I haven't used that exact one, but the ones I have seen would download 
java applet to a browser that worked like vnc and does what you want. 
If you run their openmanage agent you can do it without the management 
NIC, but then the OS has to be running and there's not much advantage 
over VNC.  The management NIC will work across a reboot of the machine.



Dell's site is just too big for me to get any proper answers on this matter


I've found their web support to be excellent if you can get past the 
place where they sort out the individual/small/large business users. 
Try to find support/downloads and the place where you put in the model 
number or their tag number which identifies it completely.  Then you go 
directly to everything they have for that device, which might be flash 
updates, all the manuals, and any utilities they provide.  I think they 
maintain this for everything they have ever sold.


--
   Les Mikesell
[EMAIL PROTECTED]

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

Re: [CentOS] how do I manage a Dell PE 860 server via IPMI?

[Lanny Marcus]

On 6/21/08, Les Mikesell <[EMAIL PROTECTED]> wrote:

> I've found their web support to be excellent if you can get past the
> place where they sort out the individual/small/large business users.
> Try to find support/downloads and the place where you put in the model
> number or their tag number which identifies it completely.  Then you go
> directly to everything they have for that device, which might be flash
> updates, all the manuals, and any utilities they provide.  I think they
> maintain this for everything they have ever sold.

There is also a huge community of users on the Dell support site and
many of them have the answers. I find it much easier to use the Dell
support site, for our 4 Dimension boxes, than the Compaq support site,
for our Compaq EVO.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recommendations for copying large filesystems

[Les Mikesell]

Mag Gam wrote:

I need to copy over 100TB of data from one server to another via network.
What is the best option to do this? I am planning to use rsync but is there
a better tool or better way of doing this?

For example, I plan on doing
rsync -azv /largefs /targetfs

/targetfs is a NFS mounted filesystem.


The only problem you are likely to have is that rsync reads the entire 
directory contents into RAM before starting, then walks the list fixing 
the differences.  If you have a huge number of files and a small amount 
of RAM, it may slow down due to swapping.  'cp -a ' can be faster if the 
target doesn't already have any matching files.  Also, the -v to display 
the names can take longer than the file transfer on small files. 
Running rsync over ssh instead of nfs has a tradeoff in that the remote 
does part of the work but you lose some speed to ssh encryption. If the 
filesystem is live, you might make an initial run copying the larger 
directories with rsync or cp, then do whatever you can to stop the files 
from changing and make another pass with 'rsync --av --delete' which 
should go fairly quickly and and fix any remaining differences.


--
  Les Mikesell
[EMAIL PROTECTED]


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

RE: [CentOS] centos 4.6 - 586 install - how to get that to a 486 level if possible

[Kevin K]

I tried RHEL 3 (I believe) on a Pentium class system, and that did not work
well.

For everything but the kernel, there was a 386 version available (glibc,
ssl, etc), so we recompiled the kernel for the 586.

Unfortunately, glibc, when you compile it for anything below a 686, did not
support NTPL threads, causing nothing but heartache.  Luckily, before we got
too far, we learned we didn't have to target the Pentium based computer
anymore.  I am happy we didn't have to validate everything still worked when
something as major as glibc is rebuilt (since I started looking whether it
was possible to rebuild glibc for 586 and NTPL).  Kernels are less risky,
since people rebuild them more frequently.

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Johnny Hughes
Sent: Friday, June 20, 2008 8:49 PM
To: CentOS mailing list
Subject: Re: [CentOS] centos 4.6 - 586 install - how to get that to a 486
level if possible

Jerry Geis wrote:
> Is there a method to get a 586 (i586 text) install  to a 486 level?
> I am looking for information guidance on this.
> I have looked into using debian/386 which stinks in my opinion, 
> slackware doesnt quite have it either.
> 
> So I am wishing/hoping there is a NOT TO painful way to get a 586 
> install to run on a 486 chip.
> 

This would be almost impossible with centos.

The reason is that there are things that glibc will compile in that are not
i486 compatible

You would be much better off trying to do this with debian I think.

If it were possible, it would be by installing the i386 glibc, i386 openssl
and editing the kernel spec file and the config file.

You can try to change the processor type to i386 or i486 and see if it will
work ... however I do not think it is possible on other than
centos-3 to get an i386 compile.

> I have been searching but havent found anything useful. If anyone 
> knows what might need to be done I would appreciate it.
> 



No virus found in this incoming message.
Checked by AVG. 
Version: 8.0.100 / Virus Database: 270.4.1/1511 - Release Date: 6/20/2008
11:52 AM

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

Re: [CentOS] Pulling Hair Out - TWiki 4.2 on CentOS 5

[Les Mikesell]

MHR wrote:

On Thu, Jun 19, 2008 at 5:11 AM, Chuck <[EMAIL PROTECTED]> wrote:

Once I build a system and bring it to our defined baseline, I rarely use rpm
from that point forward...I custom roll almost everything -- especially
apache. (red hat's layout makes my skin crawl) When did CPAN become so bad?
It was the defacto standard and source of truth for perl modules 10 years
ago. I trust CPAN over any rpm provided by red hat. Maybe things have
changed, it has been several years since I got down and dirty with perl
modules...



If your distrust of Red Hat is so high, one has to wonder why you're
using CentOS at all.


Depending on how far back you go, the approach may or may not be 
warranted.  If you used anything before RH7.3 you were pretty much 
forced to roll your own apache/perl to get a working mod_perl.  The 
stock version finally was built right in RH7.3, then broken again in 8.0 
until something much later, not sure if it was RHEL3 or 4 - or maybe it 
was broken different in each of those and fixed in updates.  As I recall 
the main symptom of the stock one was that it leaked memory whenever a 
perl page was updated, but there may have been other things wrong too.



No insult or deprecation intended, it's just that there are many Linux
options around, enough of them free.


I'm not sure if any of them got mod_perl right, and since 
apache/perl/CPAN are relatively easy to replace it wouldn't be worth 
switching for.



Personally, I like the stability and reliability of CentOS (RHEL)
enough to put up with any inconveniences I have found so far.
Besides, the support on this list is sublime.


I have to agree - for several years there have been few surprises. 
However, many commonly needed perl modules aren't in the stock repo so 
if you use applications like twiki, RT, MimeDefang, etc. you were forced 
to use CPAN until Rpmforge and EPEL added them.  (And java stuff is even 
worse...).


--
   Les Mikesell
[EMAIL PROTECTED]

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

[CentOS] cannot unmount volume "xxx"

[李晖]

Hi all:
I am using Centos 5.1. But now I have a problem when unmount a removable
usb hard disk with right click and choose unmount volume command. When I did
that, system reminds me with a message like this:Cannot unmount volume
"xxx", Detail: Cannot remove directory, "xxx" represents a temporary
directory made when the system auto mount the disk in /media, and "xxx" is
simple-chinese. But in fact, the volume has been removed because there're no
files or folder in xxx.
Because the file system type of the hard disk is NTFS, Ｉ install ntfs-3g
and add a script named "mount.ntfs" in /sbin, the content of the script is
#!/bin/sh
export LANG=zh_CN.UTF-8
exec /sbin/mount.ntfs-3g "$@"
I googled this problem for many days and didn't find any way to solve
this problem. Maybe it is a bug. I hope someone can help me. Thank you.
-- 
Sunny Lee
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recommendations for copying large filesystems

[Rainer Duffner]

Am 21.06.2008 um 15:33 schrieb Mag Gam:

I need to copy over 100TB of data from one server to another via  
network. What is the best option to do this? I am planning to use  
rsync but is there a better tool or better way of doing this?


For example, I plan on doing
rsync -azv /largefs /targetfs

/targetfs is a NFS mounted filesystem.





What network link is there between these hosts?

Are these 1 or 2 million small files or bigger ones?

Does the data change a lot?

Is it a SAN or JBOD?


cheers,
Rainer
--
Rainer Duffner
CISSP, LPI, MCSE
[EMAIL PROTECTED]


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

[CentOS] Re: centos 4.6 - 586 install - how to get that to a 486 level if possible

[Scott Silva]

on 6-21-2008 6:17 AM Jerry Geis spake the following:


Sorry, I have no clue about your question. However, I have several
AMD K6 chips (pentium-equivalent) lying around should you trip over
a motherboard that'll run them. That would solve your 486 problem, I
think, so let me know if such a chip would help you out.
  
Fred - thanks for the offer - however I do have a specific device I am 
targeting. Ebox2300sx.
I have no options on the hardware. It would be a great solution hardware 
wise but software is

being a bit of a pain at this time.

Johnny - thanks for the info. I actually have debian booted on the 
device but then EVERYTHING
is so much different on debian than centos. I was also trying a redhat 9 
install and put a 2.6 kernel on it.
However, I realize this is not the best solution. BUT - to try it out 
and get something working with
the other programs I need I thought that might be a valid short term 
path to check things out.


Thanks guys,

Jerry
I still think Debian, Ubuntu, Gentoo, or one of the BSD's will be your best 
option on that hardware. But Gentoo would take forever to compile/install on 
that hardware.


--
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] Re: centos 4.6 - 586 install - how to get that to a 486 level if possible

[nate]

Scott Silva wrote:

> I still think Debian, Ubuntu, Gentoo, or one of the BSD's will be your best
> option on that hardware. But Gentoo would take forever to compile/install on
> that hardware.

I thought I read at some point that Debian etch dropped 486 support.

Another option may be CentOS 3.x or CentOs 2.x, I'd think one of them
would support 486, and in theory at least they are still getting
security updates. Older hardware often needs older software to run,
at least there's an option to run older software that's still
supported.

nate

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

Re: [CentOS] cannot unmount volume "xxx"

[nate]

ÀîêÍ wrote:

> I googled this problem for many days and didn't find any way to solve
> this problem. Maybe it is a bug. I hope someone can help me. Thank you.


You may lose data doing this but at least for NFS volumes if something
is stuck I umount the file systems with the "-l -f" options. I'm not
sure if it'll work for NTFS but if you really want to force unmount
try

umount (path to mounted file system) -l -f

You may want to check to see if any files are in use on that file system
first, for that I'd use lsof

lsof | grep (path to mounted file system)

nate

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

Re: [CentOS] Re: centos 4.6 - 586 install - how to get that to a 486 level if possible

[Johnny Hughes]

nate wrote:

Scott Silva wrote:


I still think Debian, Ubuntu, Gentoo, or one of the BSD's will be your best
option on that hardware. But Gentoo would take forever to compile/install on
that hardware.


I thought I read at some point that Debian etch dropped 486 support.

Another option may be CentOS 3.x or CentOs 2.x, I'd think one of them
would support 486, and in theory at least they are still getting
security updates. Older hardware often needs older software to run,
at least there's an option to run older software that's still
supported.


The real issue is that glibc compiled for --target i386 is not really 
100% i386 compatible ... also NTPL doesn't work (is turned off).  This 
will potentially affect many other packages.


Upstream just made many assumptions in glibc and, the bottom line is 
that centos is just not designed to work on less than an i586 machine 
... well really i686, but i586 SHOULD work with the proper mods.


However, IMHO, i486 is impossible to get working (reliably) without 
rebuilding everything with the new glibc/gcc.





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

Re: [CentOS] recommendations for copying large filesystems

[Mag Gam]

Network is a 10/100
1 million large files
No SAN, JBOD



On Sat, Jun 21, 2008 at 1:19 PM, Rainer Duffner <[EMAIL PROTECTED]>
wrote:

>
> Am 21.06.2008 um 15:33 schrieb Mag Gam:
>
>
>  I need to copy over 100TB of data from one server to another via network.
>> What is the best option to do this? I am planning to use rsync but is there
>> a better tool or better way of doing this?
>>
>> For example, I plan on doing
>> rsync -azv /largefs /targetfs
>>
>> /targetfs is a NFS mounted filesystem.
>>
>>
>>
>
> What network link is there between these hosts?
>
> Are these 1 or 2 million small files or bigger ones?
>
> Does the data change a lot?
>
> Is it a SAN or JBOD?
>
>
> cheers,
> Rainer
> --
> Rainer Duffner
> CISSP, LPI, MCSE
> [EMAIL PROTECTED]
>
>
>
> ___
> 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] recommendations for copying large filesystems

[Rainer Duffner]

Am 21.06.2008 um 21:51 schrieb Mag Gam:


Network is a 10/100




You're kidding?




1 million large files
No SAN, JBOD




Move the data by moving the storage itself.
It will take months to transfer 100 TB via FastEthernet.



cheers,
Rainer
--
Rainer Duffner
CISSP, LPI, MCSE
[EMAIL PROTECTED]


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

Re: [CentOS] recommendations for copying large filesystems

[John R Pierce]

Mag Gam wrote:

Network is a 10/100
1 million large files
No SAN, JBOD


assuming 100baseT wire speed of about 10Mbyte/sec, moving 100TB will 
take a minimum of 100TB/10MB/s = 10,000,000 seconds or 2900 hours, or 
about 4 months.   even on a gigE network, this would still take about 2 
weeks or more.






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

Re: [CentOS] how do I manage a Dell PE 860 server via IPMI?

[Rudi Ahlers]

Lanny Marcus wrote:

On 6/21/08, Les Mikesell <[EMAIL PROTECTED]> wrote:

  

I've found their web support to be excellent if you can get past the
place where they sort out the individual/small/large business users.
Try to find support/downloads and the place where you put in the model
number or their tag number which identifies it completely.  Then you go
directly to everything they have for that device, which might be flash
updates, all the manuals, and any utilities they provide.  I think they
maintain this for everything they have ever sold.



There is also a huge community of users on the Dell support site and
many of them have the answers. I find it much easier to use the Dell
support site, for our 4 Dimension boxes, than the Compaq support site,
for our Compaq EVO.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

  

Thanx, I'll give the community a try :)

--

Kind Regards
Rudi Ahlers
CEO, SoftDux

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

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

Re: [CentOS] recommendations for copying large filesystems

[nightduke]

Can add fiber network card to each server? fiber switch? if not try to
plugin to each server giga ethernet card put a crossover cable and
start rsync... i did that with 1tb of photos and takes a lot of
timekeep power supply working and cross the fingers

I hope this can help

2008/6/21 John R Pierce <[EMAIL PROTECTED]>:
> Mag Gam wrote:
>>
>> Network is a 10/100
>> 1 million large files
>> No SAN, JBOD
>
> assuming 100baseT wire speed of about 10Mbyte/sec, moving 100TB will take a
> minimum of 100TB/10MB/s = 10,000,000 seconds or 2900 hours, or about 4
> months.   even on a gigE network, this would still take about 2 weeks or
> more.
>
>
>
>
>
> ___
> 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] recommendations for copying large filesystems

[Matt Morgan]

On Sat, Jun 21, 2008 at 5:12 PM, nightduke <[EMAIL PROTECTED]> wrote:

> Can add fiber network card to each server? fiber switch? if not try to
> plugin to each server giga ethernet card put a crossover cable and
> start rsync... i did that with 1tb of photos and takes a lot of
> timekeep power supply working and cross the fingers
>
> I hope this can help
>
> 2008/6/21 John R Pierce <[EMAIL PROTECTED]>:
> > Mag Gam wrote:
> >>
> >> Network is a 10/100
> >> 1 million large files
> >> No SAN, JBOD
> >
> > assuming 100baseT wire speed of about 10Mbyte/sec, moving 100TB will take
> a
> > minimum of 100TB/10MB/s = 10,000,000 seconds or 2900 hours, or about 4
> > months.   even on a gigE network, this would still take about 2 weeks or
> > more.
>

Then if you get the network sorted out, the fastest & most reliable way I
know to copy lots of files is

star --copy

You can get star with

yum install star

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

Re: [CentOS] recommendations for copying large filesystems

[Rainer Duffner]

Am 21.06.2008 um 23:44 schrieb Matt Morgan:


O
Then if you get the network sorted out, the fastest & most reliable  
way I know to copy lots of files is


star --copy

You can get star with

yum install star




Now that I know the details - I don' think this is going to work. Not  
with 100 TB of data. It kind-of-works with 1 TB.

Can anybody comment on the feasibility of rsync on 1 million files?
Maybe DRBD would be a solution.
If you can retrofit DRDB to an existing setup...

If not it's faster to move the drives physically - believe me, this  
will create far less problems.
In a SAN, you would have the possibility of synching the data outside  
of the filesystem, during normal operations.



100 TB is a lot of data.
How do you back that up, BTW?
What is your estimated time to restore it from the medium you  back  
it up to?




cheers,
Rainer
--
Rainer Duffner
CISSP, LPI, MCSE
[EMAIL PROTECTED]


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

Re: [CentOS] Network FS w/o user setup

[Ted Miller]

Johnny Hughes wrote:

Ted Miller wrote:

Johnny Hughes wrote:

Ted Miller wrote:
Is there a file system + configuration that will let me share a 
directory, and anyone who has access to something in that directory 
on the server will also have access (and lack of access) to the same 
files from the client? Clients will be Centos5, Win2K, WinXP.  
Server is Centos5.


To put it another way, all users have accounts on the server.  I 
don't want to have to set up ANY user information on the server, 
other than what I set up to control local access.  I just want to 
say "Share /vmware" and have it available, to the same users who can 
access it locally.


With Samba I have to maintain duplicate user lists, password lists, 
and share access lists.  I have not been able to find a clear 
instructions on how NFS4 handles this, but what I found didn't seem 
any better than Samba.


I don't mind implementing ACLs on the server if it will do what I 
need, but I can't find anything that says it will save me any work 
either.


Well, since you want to set up shares ... and since you want to share 
between Windows and Linux machines, and to share for windows you will 
need to use samba.


Since you can also set up linux to use a samba client, that would 
probably be the best method to "share these files" ... if you expect 
to just oepn them via a file manager on all platforms.


Is there a way to set up samba so that it "just uses" ACL information 
for permissions, instead of having to spell everything out for each 
share and each user?


Well ... you would need to Join the "Samba Server" to your "Windows 
Domain".  If that domain is ADS (Active Directory Services) then it is a 
different procedure than if it is a WinNT type Windows Domain.


This is getting well outside the range of complexity that I am looking for. 
 If I add more detail, maybe something more suitable to my situation will 
suggest itself to members of the list.


1. This is a very small network, only one primary file server (office2). A 
second file server (RAIDer1) has only one shared directory, so is not 
really an issue.


2. Users log in primarily from Linux boxes, but have to run virtual Windows 
machines for some software, and also log in from Windows laptops.


3. office2 is set up with logins and home directories for all users, and 
directories are permissioned such that users can run programs on office2 
(if needed) and directory permissions work right.


4. Some users don't have physical machines, but only have virtual 
machine(s) running on office2, which also need "network" access to office2 
files.


Because all the users and permissions already exist on office2, I would 
like those existing permissions to be reflected when the file system is 
shared, just the same as when it is accessed locally.  To restate: my 
desire is that users, logins, and permissions be identical whether a user 
is logged into office2 or whether that user is using a network file share 
from another virtual or physical machine, running Linux or Windows.  I 
would think there would be a "market" for a network file system where 
sharing a directory tree involved no more than assigning a network share 
name to it.  If (and only if) you had access to the file locally, you now 
have access to it on the network.  Very simple to administer, very simple 
to understand--one set of permissions (kept locally) works everywhere.


From everything I have heard, a windows domain controller would be more 
work than it is worth for this size of project, as I am looking for 
something machine-scale, not enterprise scale.


I hope this more clearly expresses my desires, even if only so that 
everyone can tell me to keep dreaming, because what I want doesn't 
exist--or in the open source tradition, quit dreaming and start coding. 
(Unfortunately I am still working on my first C++ lesson book.)


Sorry I neglected this (and all other) threads for a week or more, as I had 
to learn how to do video editing to rescue an otherwise disastrously 
unusable video project for my employer.


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

Re: [CentOS] Network FS w/o user setup

[Les Mikesell]

Ted Miller wrote:

Johnny Hughes wrote:


Well ... you would need to Join the "Samba Server" to your "Windows 
Domain".  If that domain is ADS (Active Directory Services) then it is 
a different procedure than if it is a WinNT type Windows Domain.


This is getting well outside the range of complexity that I am looking 
for.  If I add more detail, maybe something more suitable to my 
situation will suggest itself to members of the list.


1. This is a very small network, only one primary file server (office2). 
A second file server (RAIDer1) has only one shared directory, so is not 
really an issue.


2. Users log in primarily from Linux boxes, but have to run virtual 
Windows machines for some software, and also log in from Windows laptops.


Virtual windows machines should be no different in terms of network 
connections, so you can ignore that distinction.


3. office2 is set up with logins and home directories for all users, and 
directories are permissioned such that users can run programs on office2 
(if needed) and directory permissions work right.


Is samba running there? If so, you are mostly done.

4. Some users don't have physical machines, but only have virtual 
machine(s) running on office2, which also need "network" access to 
office2 files.


Again, nothing different.

Because all the users and permissions already exist on office2, I would 
like those existing permissions to be reflected when the file system is 
shared, just the same as when it is accessed locally.  To restate: my 
desire is that users, logins, and permissions be identical whether a 
user is logged into office2 or whether that user is using a network file 
share from another virtual or physical machine, running Linux or 
Windows.  I would think there would be a "market" for a network file 
system where sharing a directory tree involved no more than assigning a 
network share name to it.  If (and only if) you had access to the file 
locally, you now have access to it on the network.  Very simple to 
administer, very simple to understand--one set of permissions (kept 
locally) works everywhere.


This mostly "just works" if you deal with a few complications that on a 
small scale can be worked around without too much trouble.  The first 
complication is that you need to maintain passwords separately for Linux 
and Windows because they are stored with different encryption.  If you 
aren't already using samba, you need to 'smbpasswd -a username' for each 
user and input the password (or go around and let them type it 
themselves).  After this, a windows user mapping a samba-shared 
directory from your office2 machine will have the same access as the 
same user logged in locally.  There are the same issues with directories 
that users share with group permissions, but samba offers some extra 
options to force owner/group/permissions on newly created files that 
will help. Windows/samba connections are treated as single users with 
all access through that connection treated with the permissions of the 
matching linux login.  With samba in 'user' mode, the authentication is 
done before you can even see the shares and even if you have multiple 
shares mapped from the server they must all be as the same user.  There 
is also a 'share' mode where you authenticate separately per connection.


 From everything I have heard, a windows domain controller would be more 
work than it is worth for this size of project, as I am looking for 
something machine-scale, not enterprise scale.


You might look at webmin, since it has an option to maintain unix and 
samba passwords at the same time and it can also keep multiple machines 
in sync.  The other complication is that if you also want to share files 
via NFS, the permissioning mechanism is entirely different.  NFS just 
looks at the uid/gid/modes like a local file, so you need to make the 
password files consistent across all the Linux boxes.  There is also the 
issue that users who have root access to their own workstation can 
pretend to be any user over NFS.  For a single-user Linux workstation 
scenario, it might make more sense to only provide samba shares and use 
cifs mounts instead of NFS.  NFS makes more sense between multiuser 
unix/linux boxes where only the administrator(s) have root access.


I hope this more clearly expresses my desires, even if only so that 
everyone can tell me to keep dreaming, because what I want doesn't 
exist--or in the open source tradition, quit dreaming and start coding. 
(Unfortunately I am still working on my first C++ lesson book.)


I don't think you need to code anything since there are already several 
options with varying degrees of complexity.   Centralizing 
authentication will help if you have many users and password changes. 
But that can be as simple as turning on domain controller emulation on 
samba on your office2 server and configuring everything else (windows 
and Linux) to use it.  Or it can be as complicated as running a separate 

Re: [CentOS] Trouble brewing in dmesg... any ideas?

[Ted Miller]

Rob Townley wrote:

dmesg > dmesg.log

or

cd /var/log/
ls -lat | more

i liked the old days when dmesg, /var/log/messages and other syslog 
stuff was displayed automatically on a tty console.  I tried a softlink 
from /var/log/messages to tty9, but didn't have any luck.  Would it 
require a tee or a mod to dmesg?


One of those rare occasions when I know the answer, so I'll share.  Had to 
go look at my /etc/syslog.conf file to remember the magic incantation:


# Log everything to tty12
# Added by TCM on 24 Nov 06
*.* /dev/tty12

This puts all that "stuff" on Ctrl-Alt-F12 "where it belongs".  Enjoy.

Ted Miller





On Wed, Jun 18, 2008 at 6:38 PM, Tim Nelson <[EMAIL PROTECTED] 
> wrote:


Unfortunately I can't see the top of the errors as there are too
many... :-( I'll throw a console on it and start logging. Is anyone
else seeing this sort of activity? I'm running the latest stock
kernel available using yum from the repos. I'm not using any
additional repos(rpmforge, epel, etc...) and I don't have any custom
compiled modules. This box is a fresh installation running bind,
apache, and mysqld.

Tim Nelson
Systems/Network Support
Rockbochs Inc.
(218)727-4332 x105

- Original Message -
From: "nate" <[EMAIL PROTECTED] >
To: centos@centos.org 
Sent: Wednesday, June 18, 2008 6:19:07 PM GMT -06:00 Guadalajara /
Mexico City / Monterrey
Subject: Re: [CentOS] Trouble brewing in dmesg... any ideas?

Tim Nelson wrote:

 > There are others with various app names besides vi including
httpd, named,
 > sftp-server, etc..  Is this an imminent hardware failure? Do I
have kernel
 > issues? I've checked the system with lm_sensors and temps are
perfectly
 > normal. Also, performance and operation seems to be fine. Even
with these
 > errors, my services are running without any hiccups. HELP! :-)
 >

Would need to see the full error but it sounds like a kernel oops. For
me at least the useful info would be at the top of the error which
wasn't
included in your email.

Worst case, configure your system with a serial console and capture the
error using a terminal emulator on another machine plugged into your
serial console.

nate

___
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 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] wtmp binary is 5 times bigger than its equivalent text file

[Centos]

On one of our servers wtmp is growing very fast.
The binary file is about 50M but when I change it to ascii with fwtmp command,
The ascii file is only 10M.

binary file is being 5 times bigger than text file ?
Any suggestion ?

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

Re: [CentOS] recommendations for copying large filesystems

[Erek Dyskant]

On Sat, 2008-06-21 at 09:33 -0400, Mag Gam wrote:
> I need to copy over 100TB of data from one server to another via
> network. What is the best option to do this? I am planning to use
> rsync but is there a better tool or better way of doing this?

At gigabit speeds, you're looking at over a week of transfer time: 1
gigabit = 125MB/sec = 800,000 seconds = 9.25 days, not counting protocol
overhead.  You could speed this up with link bonding, which from
previous threads sounds like something you're working on already.

If it's a oneoff transfer and you can afford downtime while you're
fiddling with hardware, you may consider directly attaching both sets of
storage to the same machine and doing a local copy.

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

Re: [CentOS] Network FS w/o user setup

[Ted Miller]

Thanks for the reply.  I think we are making progress, see 
comments/questions interspersed below.


Les Mikesell wrote:

Ted Miller wrote:

Johnny Hughes wrote:


Well ... you would need to Join the "Samba Server" to your "Windows 
Domain".  If that domain is ADS (Active Directory Services) then it 
is a different procedure than if it is a WinNT type Windows Domain.


This is getting well outside the range of complexity that I am looking 
for.  If I add more detail, maybe something more suitable to my 
situation will suggest itself to members of the list.


1. This is a very small network, only one primary file server 
(office2). A second file server (RAIDer1) has only one shared 
directory, so is not really an issue.


2. Users log in primarily from Linux boxes, but have to run virtual 
Windows machines for some software, and also log in from Windows laptops.


Virtual windows machines should be no different in terms of network 
connections, so you can ignore that distinction.


3. office2 is set up with logins and home directories for all users, 
and directories are permissioned such that users can run programs on 
office2 (if needed) and directory permissions work right.


Is samba running there? If so, you are mostly done.


Yes, at the moment I have Samba running, but apparently not properly 
configured.  I am also in the process of moving this machine from Centos 4 
to Centos 5, and am trying to do it better this time.  At the moment 
office2 is dual boot, still defaulting to C4.


Because all the users and permissions already exist on office2, I 
would like those existing permissions to be reflected when the file 
system is shared, just the same as when it is accessed locally.  To 
restate: my desire is that users, logins, and permissions be identical 
whether a user is logged into office2 or whether that user is using a 
network file share from another virtual or physical machine, running 
Linux or Windows.  I would think there would be a "market" for a 
network file system where sharing a directory tree involved no more 
than assigning a network share name to it.  If (and only if) you had 
access to the file locally, you now have access to it on the network.  
Very simple to administer, very simple to understand--one set of 
permissions (kept locally) works everywhere.


This mostly "just works" if you deal with a few complications that on a 
small scale can be worked around without too much trouble.  The first 
complication is that you need to maintain passwords separately for Linux 
and Windows because they are stored with different encryption.  If you 
aren't already using samba, you need to 'smbpasswd -a username' for each 
user and input the password (or go around and let them type it 
themselves).


Done at this point.

After this, a windows user mapping a samba-shared 
directory from your office2 machine will have the same access as the 
same user logged in locally.  There are the same issues with directories 
that users share with group permissions, but samba offers some extra 
options to force owner/group/permissions on newly created files that 
will help.


That is something I need to fix, because I do have some issues with group 
accessed files, where certain operations require me to log in as root and 
run a script that cleans up the file ownership, otherwise some users can no 
longer access the files.  Any pointers on where to find documentation on this?


Windows/samba connections are treated as single users with 
all access through that connection treated with the permissions of the 
matching linux login.  With samba in 'user' mode, the authentication is 
done before you can even see the shares and even if you have multiple 
shares mapped from the server they must all be as the same user.  There 
is also a 'share' mode where you authenticate separately per connection.


I have been using 'share' mode, but a little reading makes it sound like I 
should switch to 'user' mode to make my life easier.  I have been adding 
various user permission lines to each share.  Will they keep working if I 
just comment out those lines?


 From everything I have heard, a windows domain controller would be 
more work than it is worth for this size of project, as I am looking 
for something machine-scale, not enterprise scale.


You might look at webmin, since it has an option to maintain unix and 
samba passwords at the same time and it can also keep multiple machines 
in sync.


Does anyone maintain webmin for Centos?  I have most of the common repos 
hooked to yum, but webmin draws a blank.


The other complication is that if you also want to share files 
via NFS, the permissioning mechanism is entirely different.  NFS just 
looks at the uid/gid/modes like a local file, so you need to make the 
password files consistent across all the Linux boxes.


Does NFS work with windows?  I have wasted considerable time on Google 
trying to answer that question, and the only answer I find is that there 
are 

Re: ****Re: [CentOS] Network FS w/o user setup

[Craig White]

On Sat, 2008-06-21 at 21:48 -0400, Ted Miller wrote:

> I have been using 'share' mode, but a little reading makes it sound like I 
> should switch to 'user' mode to make my life easier.  I have been adding 
> various user permission lines to each share.  Will they keep working if I 
> just comment out those lines?

share mode is designed to mimic Windows 98 file sharing behavior - it's
use is not recommended.

As for will they keep working...crystal ball cloudy. You can simply keep
editing smb.conf because samba re-reads the smb.conf every minute I
think.

> >>  From everything I have heard, a windows domain controller would be 
> >> more work than it is worth for this size of project, as I am looking 
> >> for something machine-scale, not enterprise scale.
> > 
> > You might look at webmin, since it has an option to maintain unix and 
> > samba passwords at the same time and it can also keep multiple machines 
> > in sync.
> 
> Does anyone maintain webmin for Centos?  I have most of the common repos 
> hooked to yum, but webmin draws a blank.

download/install from www.webmin.com - also install perl-net-SSLeay for
SSL (from dag/rpmforge)

> > The other complication is that if you also want to share files 
> > via NFS, the permissioning mechanism is entirely different.  NFS just 
> > looks at the uid/gid/modes like a local file, so you need to make the 
> > password files consistent across all the Linux boxes.
> 
> Does NFS work with windows?  I have wasted considerable time on Google 
> trying to answer that question, and the only answer I find is that there 
> are commercial products that (for a per-seat fee) will connect windows to 
> NFS.  I read that NFS v.4 was supposed to "play better" with windows, but I 
> could not find any official comment, or windows drivers, or even any 
> recommendations of client only drivers.

SFU (Services for Unix) is a free download from Microsoft

> Any pointers to where I could learn the implications/pluses/minuses of 
> that?  It might be useful with my multiple machines (real and virtual) per 
> user.

You put barriers into place deciding what you are willing and not
willing to do but in reality, what you want is all of the power and
features of an enterprise system with none of the knowledge...good luck.

With LDAP, you could have multiple servers, integrated users/groups
between posix and Windows and even have them see the same Desktop, the
same $HOME directories regardless of whether they connected via Windows
or Linux.

Craig

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

Re: [CentOS] cannot unmount volume "xxx"

[李晖]

Thank you for your help. I will try this when I mount the hard disk next
time. Thanks again.

2008/6/22 nate <[EMAIL PROTECTED]>:

> ÀîêÍ wrote:
>
> > I googled this problem for many days and didn't find any way to solve
> > this problem. Maybe it is a bug. I hope someone can help me. Thank you.
>
>
> You may lose data doing this but at least for NFS volumes if something
> is stuck I umount the file systems with the "-l -f" options. I'm not
> sure if it'll work for NTFS but if you really want to force unmount
> try
>
> umount (path to mounted file system) -l -f
>
> You may want to check to see if any files are in use on that file system
> first, for that I'd use lsof
>
> lsof | grep (path to mounted file system)
>
> nate
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>



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

Re: [CentOS] Network FS w/o user setup

[Les Mikesell]

Ted Miller wrote:



Is samba running there? If so, you are mostly done.


Yes, at the moment I have Samba running, but apparently not properly 
configured.  I am also in the process of moving this machine from Centos 
4 to Centos 5, and am trying to do it better this time.  At the moment 
office2 is dual boot, still defaulting to C4.


I don't think there should be a big difference samba-wise.

If you aren't already using samba, you need to 'smbpasswd 
-a username' for each user and input the password (or go around and 
let them type it themselves).


Done at this point.

After this, a windows user mapping a samba-shared directory from your 
office2 machine will have the same access as the same user logged in 
locally.  There are the same issues with directories that users share 
with group permissions, but samba offers some extra options to force 
owner/group/permissions on newly created files that will help.


That is something I need to fix, because I do have some issues with 
group accessed files, where certain operations require me to log in as 
root and run a script that cleans up the file ownership, otherwise some 
users can no longer access the files.  Any pointers on where to find 
documentation on this?


Newly created files default to having the group ownership of the primary 
group of the user creating it, and the RH scheme is to give every user 
his own group.  You can do something like this in the samba share 
configuration:

valid users = @groupname
force group = groupname
force create mode = 0775
force directory mode = 0775

This will make new files have the right group and r/w (and executable, 
which you might not want).  If files are also created on the Linux side 
you need to 'chmod g+s' on the directory to make new files take the 
group of the directory.


You can find samba docs here: 
http://us1.samba.org/samba/docs/man/Samba-HOWTO-Collection/


I have been using 'share' mode, but a little reading makes it sound like 
I should switch to 'user' mode to make my life easier.  I have been 
adding various user permission lines to each share.  Will they keep 
working if I just comment out those lines?


Share vs. user doesn't make a difference in how things work after the 
connection is established - it controls when authentication happens. 
Share mode just lets you browse the share list before authenticating and 
you can connect to different shares with different credentials.


You might look at webmin, since it has an option to maintain unix and 
samba passwords at the same time and it can also keep multiple 
machines in sync.


Does anyone maintain webmin for Centos?  I have most of the common repos 
hooked to yum, but webmin draws a blank.


This is one of the reasons I usually install k12ltsp instead of the 
stock centos distribution (you don't lose anything, it just adds some 
extras and makes the updates yummable).  You probably can grab the RPM 
directly from the webmin site.


The other complication is that if you also want to share files via 
NFS, the permissioning mechanism is entirely different.  NFS just 
looks at the uid/gid/modes like a local file, so you need to make the 
password files consistent across all the Linux boxes.


Does NFS work with windows?  I have wasted considerable time on Google 
trying to answer that question, and the only answer I find is that there 
are commercial products that (for a per-seat fee) will connect windows 
to NFS.


On windows you would use 'services for unix'.  But I'd stick to samba.


> I read that NFS v.4 was supposed to "play better" with windows,
but I could not find any official comment, or windows drivers, or even 
any recommendations of client only drivers.


Don't know about that.

There is also the issue that users who have root access to their own 
workstation can pretend to be any user over NFS.


Not an issue in this situation, users do not have root access.


Do they have the same uid/gid, and group lists on their workstations as 
on the file server?


For a single-user Linux workstation scenario, it might make more sense 
to only provide samba shares and use cifs mounts instead of NFS.  NFS 
makes more sense between multiuser unix/linux boxes where only the 
administrator(s) have root access.


That is what I did under C4, but with considerable frustration, but 
maybe a simplified version of what I had (minus per-share permission 
listing in smb.conf) would get me most of what I want.


If it is one user per workstation, treating it just like the windows 
connections with samba should work.


I don't think you need to code anything since there are already 
several options with varying degrees of complexity.   Centralizing 
authentication will help if you have many users and password changes. 
But that can be as simple as turning on domain controller emulation on 
samba on your office2 server and configuring everything else (windows 
and Linux) to use it.


Any pointers to where I could learn the implications/pluses/minuses

[CentOS] 3ware 9650 issues

[Joshua Baker-LePain]

I've been having no end of issues with a 3ware 9650SE-24M8 in a server that's 
coming on a year old.  I've got 24 WDC WD5001ABYS drives (500GB) hooked to it, 
running as a single RAID6 w/ a hot spare.  These issues boil down to the card 
periodically throwing errors like the following:


sd 1:0:0:0: WARNING: (0x06:0x002C): Command (0x8a) timed out, resetting card.

Usually when this happens, it's followed by:

3w-9xxx: scsi1: AEN: INFO (0x04:0x005E): Cache synchronization 
completed:unit=0.


On the less pleasant occasions, it's followed by:

scsi1: ERROR: (0x06:0x0036): Response queue (large) empty failed during reset 
sequence.
3w-9xxx: scsi1: ERROR: (0x06:0x002B): Controller reset failed during scsi host 
reset.

sd 1:0:0:0: scsi: Device offlined - not ready after error recovery

This of course leads to a several hour downtime as the system has to be powered 
down (not just rebooted) and then the volume needs to be fscked. I've been back 
and forth with both the vendor and (via the vendor) 3ware with this.  The card 
has been replaced, as well as the whole system.  I'm running the latest 
firmware and drivers from 3ware.


Have other folks had good luck with this card?  What sorts of configs are you 
running?  I'm in the position of needing more storage, and I'm a bit gun shy on 
3ware at the moment...


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

Re: [CentOS] 3ware 9650 issues

[John R Pierce]

Joshua Baker-LePain wrote:
I've been having no end of issues with a 3ware 9650SE-24M8 in a server 
that's coming on a year old.  I've got 24 WDC WD5001ABYS drives 
(500GB) hooked to it, running as a single RAID6 w/ a hot spare.  These 
issues boil down to the card periodically throwing errors like the 
following:


Have other folks had good luck with this card?  What sorts of configs 
are you running?  I'm in the position of needing more storage, and I'm 
a bit gun shy on 3ware at the moment...





I have no experience with that raid card, most of our larger systems use 
external SAN storage, but I will say that, IMHO, is a very large 
raid-6.   we usually don't make single raid sets much large than 7-8 
drives, and for a very large storage system, will stripe multiple 
raid5/6 sets rather than have one huge one.

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

Re: [CentOS] 3ware 9650 issues

[Joshua Baker-LePain]

On Sat, 21 Jun 2008 at 9:12pm, John R Pierce wrote


Joshua Baker-LePain wrote:
I've been having no end of issues with a 3ware 9650SE-24M8 in a server 
that's coming on a year old.  I've got 24 WDC WD5001ABYS drives (500GB) 
hooked to it, running as a single RAID6 w/ a hot spare.  These issues boil 
down to the card periodically throwing errors like the following:


Have other folks had good luck with this card?  What sorts of configs are 
you running?  I'm in the position of needing more storage, and I'm a bit 
gun shy on 3ware at the moment...





I have no experience with that raid card, most of our larger systems use 
external SAN storage, but I will say that, IMHO, is a very large raid-6.   we 
usually don't make single raid sets much large than 7-8 drives, and for a 
very large storage system, will stripe multiple raid5/6 sets rather than have 
one huge one.


Would that I had such luxuries.  This is a university lab with needs for 
massive amounts of data and not much money with which to do it.


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

Re: [CentOS] 3ware 9650 issues

[Ruslan Sivak]

Joshua Baker-LePain wrote:

On Sat, 21 Jun 2008 at 9:12pm, John R Pierce wrote

I have no experience with that raid card, most of our larger systems 
use external SAN storage, but I will say that, IMHO, is a very large 
raid-6.   we usually don't make single raid sets much large than 7-8 
drives, and for a very large storage system, will stripe multiple 
raid5/6 sets rather than have one huge one.


Would that I had such luxuries.  This is a university lab with needs 
for massive amounts of data and not much money with which to do it.


Wouldn't striping a bunch of raid6 volumes give you about the same 
amount of space? 


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

Re: [CentOS] 3ware 9650 issues

[Joshua Baker-LePain]

On Sun, 22 Jun 2008 at 1:01am, Ruslan Sivak wrote


Joshua Baker-LePain wrote:

On Sat, 21 Jun 2008 at 9:12pm, John R Pierce wrote

I have no experience with that raid card, most of our larger systems use 
external SAN storage, but I will say that, IMHO, is a very large raid-6. 
we usually don't make single raid sets much large than 7-8 drives, and for 
a very large storage system, will stripe multiple raid5/6 sets rather than 
have one huge one.


Would that I had such luxuries.  This is a university lab with needs for 
massive amounts of data and not much money with which to do it.


Wouldn't striping a bunch of raid6 volumes give you about the same amount of 
space?


No.  We have 24 drives.  Use one for a hot spare -> leaves 23.

1 array:  23 drives, - 2 for parity -> capacity = 21 * drive capacity
2 arrays: array1 = 12 drives - 2 for parity -> 10 drives
  array2 = 11 drives - 2 for parity -> 9 drives
-> capcity = 19 * drive capcity
3 arrays: array1 = 8 drives - 2 for parity -> 6 drives
  array2 = 8 drives - 2 for parity -> 6 drives
  array3 = 7 drives - 2 for parity -> 5 drives
-> capcity = 17 * drive capacity

With 1TB drives, you're losing 2TB worth of volume space for each 
increased number of arrays.  That's a lot of space.


Unless I misunderstood you...

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

Re: [CentOS] 3ware 9650 issues

[nate]

Joshua Baker-LePain wrote:

> periodically throwing errors like the following:
>
> sd 1:0:0:0: WARNING: (0x06:0x002C): Command (0x8a) timed out, resetting
> card.

Wondering if you have scheduled automatic media scans of all of the
disks in the array? Perhaps you have a disk that is going bad
causing the issue.

Something else that could be related, I was told by someone who
had a Isilon storage system(fancy NAS box), who was having his
WD disk drives hang on him on occasion, when this occured he had
to physically remove the disk from the system and re plug it in.
It was a firmware issue, I don't recall which WD drives he had,
he eventually got a fixed firmware though. This was about a year
ago.

I have media scans run once a week for about 7 hours on my 2 disk
3Ware systems (8006-2 controllers). For a 24 disk system you'll
probably need to run it longer. (unless the newer controllers scan
in parallel, the 8000 series seems to be serial).

I ran a couple 9650 series cards not too long ago, I think they
were just two disk systems running RAID 1 (up to 8 disks, but only
used 2). I've been using 3ware cards for about 8 years now and
have not run into those types of errors you describe. Probably
ran about 350 cards over the years, most of them in the 8000
series.

nate

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

Re: [CentOS] 3ware 9650 issues

[Jeff]

On Sat, Jun 21, 2008 at 11:04 PM, Joshua Baker-LePain <[EMAIL PROTECTED]> wrote:
> I've been having no end of issues with a 3ware 9650SE-24M8 in a server
> that's coming on a year old.  I've got 24 WDC WD5001ABYS drives (500GB)
> hooked to it, running as a single RAID6 w/ a hot spare.  These issues boil
> down to the card periodically throwing errors like the following:
>
> sd 1:0:0:0: WARNING: (0x06:0x002C): Command (0x8a) timed out, resetting
> card.
>
> Usually when this happens, it's followed by:
>
> 3w-9xxx: scsi1: AEN: INFO (0x04:0x005E): Cache synchronization
> completed:unit=0.
>
> On the less pleasant occasions, it's followed by:
>
> scsi1: ERROR: (0x06:0x0036): Response queue (large) empty failed during
> reset sequence.
> 3w-9xxx: scsi1: ERROR: (0x06:0x002B): Controller reset failed during scsi
> host reset.
> sd 1:0:0:0: scsi: Device offlined - not ready after error recovery
>
> This of course leads to a several hour downtime as the system has to be
> powered down (not just rebooted) and then the volume needs to be fscked.
> I've been back and forth with both the vendor and (via the vendor) 3ware
> with this.  The card has been replaced, as well as the whole system.  I'm
> running the latest firmware and drivers from 3ware.
>
> Have other folks had good luck with this card?  What sorts of configs are
> you running?  I'm in the position of needing more storage, and I'm a bit gun
> shy on 3ware at the moment...

This may be completely irrelevant, but we have a 9550 card running
RAID 5 with a 'prominent non-Linux' operating system that suffers from
the same symptoms (and 4 others that have never done it). We've heard
from our vendor (and 3ware) that there are some upcoming firmware
releases (looks like August) that might help. A 3ware tech told me
that the controller reset happens when communication between the
driver and the firmware times out, which appears to be exactly what is
in your error message.

Meanwhile, we just cross our fingers and thank our lucky stars the the
server in question is in our local office and not one of our
non-tech-staffed remote offices. There are unsupported pre-release
firmware downloads available if you like to gamble. I have not had the
courage to install the beta firmware on our servers. I have not used
3ware with CentOS, but I don't think this is a CentOS issue.

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

[CentOS] is it possible to login to a switch via RS232 / serial port from the shell?

[Rudi Ahlers]

Hi all

I have a problem switch in our data centre, which is connected to a 
linux server via a serial cable. I know I can / could access the switch 
from my Windows PC back at the office, using hyperterm, but trying to 
access it using minicom just doesn't seem to work.


I have changed the comms in minincom to use /dev/tty0 & the baud rate to 
9600, yet I can't seem to connect with minicom. Does anyone know how to 
connect to a serial device from the console?


--

Kind Regards
Rudi Ahlers
CEO, SoftDux
Office: 087 805-9573
Fax No: 086 609 6128
Cell:   082 554 7532


Web:   http://www.SoftDux.com
Forum: http://Forum.SoftDux.com

Join SA WebHostingTalk today, on http://www.WebHostingTalk.co.za

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

[CentOS] problems installing samba

[Rajeev R. Veedu]

Dear all,

 

I am trying to install Samba 3.0.30  on a new Centos 5 machine and I am getting 
following error. Samba has been downloaded form setnet. Could some please let 
me know how to fix this?

 

rpmdb: Program version 4.3 doesn't match environment version

error: db4 error(-30974) from dbenv->open: DB_VERSION_MISMATCH: Database 
environment version mismatch

warning: cannot open Solve database in /usr/lib/rpmdb/x86_64-redhat-linux/CentO

rpmdb: Program version 4.3 doesn't match environment version

 

Rajeev R. Veedu


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

RE: [CentOS] 3ware 9650 issues

[Joseph L. Casale]

>Have other folks had good luck with this card?  What sorts of configs are you
>running?  I'm in the position of needing more storage, and I'm a bit gun shy on
>3ware at the moment...

Does that drive have a jumper to slow it down to 1.5Gb transfer rate?
Cheap controllers and drives just cant do it, I have had no end of issues
even with *all* my LSI controllers until I jumped all my sata drives
down.

As far as performance, it made no impact on my systems.

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