Lowyat.NET: Latest topics by kons

DHCP perks!

kons — Sat, 26 Jan 2008 09:48:53 +0800

From Microsoft TechNet
http://technet.microsoft.com/en-us/library/bb727003.aspx

CODE

If the DHCPRequest message was received as a broadcast, the DHCP server sends the DHCPAck message to the DHCP client using the offered IPv4 address as the destination IPv4 address and the client's MAC address as the destination MAC address.

From ISC DHCP and Cisco DHCP

CODE

The DHCP/BootP Relay Agent receives the DHCPACK, and will forward the DHCPACK broadcast on the local LAN.

Microsoft is expecting the DHCPAck to be in unicast but ISC and Cisco is sending it as broadcast.
No wonder sometimes Windows XP can never get an IP, but Linux seems to work flawlessly.

[addedon]January 26, 2008, 9:51 am[/addedon]Saw this in RFC2131:

CODE

If the client is in the REQUESTING state, we assume that the client is not
yet fully configured and send the DHCPACK as a broadcast if the broadcast
flag is set and unicast if the flag is cleared.

I assume Microsoft never read RFCs?

[addedon]January 26, 2008, 9:53 am[/addedon]So, the result =

CODE

Jan 26 09:49:57 Side2 dhcpd: DHCPREQUEST for 172.16.0.237 (172.16.0.1) from 00:0e:35:cf:19:8f (compaqnb-ce) via eth1
Jan 26 09:49:57 Side2 dhcpd: Sending event IP=172.16.0.237 MAC=00:0e:35:cf:19:8f ENDS=1800
Jan 26 09:49:57 Side2 dhcpd: DHCPACK on 172.16.0.237 to 00:0e:35:cf:19:8f (compaqnb-ce) via eth1
Jan 26 09:50:01 Side2 dhcpd: DHCPREQUEST for 172.16.0.237 (172.16.0.1) from 00:0e:35:cf:19:8f (compaqnb-ce) via eth1
Jan 26 09:50:01 Side2 dhcpd: Sending event IP=172.16.0.237 MAC=00:0e:35:cf:19:8f ENDS=1800
Jan 26 09:50:01 Side2 dhcpd: DHCPACK on 172.16.0.237 to 00:0e:35:cf:19:8f (compaqnb-ce) via eth1
Jan 26 09:50:09 Side2 dhcpd: DHCPREQUEST for 172.16.0.237 (172.16.0.1) from 00:0e:35:cf:19:8f (compaqnb-ce) via eth1
Jan 26 09:50:09 Side2 dhcpd: Sending event IP=172.16.0.237 MAC=00:0e:35:cf:19:8f ENDS=1800
Jan 26 09:50:09 Side2 dhcpd: DHCPACK on 172.16.0.237 to 00:0e:35:cf:19:8f (compaqnb-ce) via eth1
Jan 26 09:50:24 Side2 dhcpd: DHCPREQUEST for 172.16.0.237 (172.16.0.1) from 00:0e:35:cf:19:8f (compaqnb-ce) via eth1
Jan 26 09:50:24 Side2 dhcpd: Sending event IP=172.16.0.237 MAC=00:0e:35:cf:19:8f ENDS=1800
Jan 26 09:50:24 Side2 dhcpd: DHCPACK on 172.16.0.237 to 00:0e:35:cf:19:8f (compaqnb-ce) via eth1

Some clients can never get IP from my dhcp server... sigh.

Capping at iDC

kons — Mon, 04 Jun 2007 18:48:14 +0800

QUOTE

TM Net's Management recently has approved that all "shared bandwidth" customers are to be capped as per below:

1. Dedicated Rack - 2.0 Mbps (per rack)

2. Half-Rack - 1.0 Mbps (per rack)

3. Shared-Rack - 512 Kbps (per rack)

4. 4U - 512 Kbps (per rack)

5. 1U - 128 Kbps (per rack)

The capping of bandwidth limit, as per above, will commence 40 days from the date of this announcement. The new ruling is applicable to all new and existing "shared bandwidth" customers.

Therefore, it is recommended that customers who are using / require higher bandwidth to be upgraded to "premium" package. You can refer to your respective AE's for the rack + bandwidth pricing lists.

Just like streamyx speed!

power supply

kons — Wed, 16 May 2007 16:46:50 +0800

How much did you guys spend to fix that?
I got a quotation of almost 3k just to fix that.
Need to wait 3 weeks for TNB approval.
Wondering if it's overpriced.

Factories been installing them for the cost of approx 10k, but I don't think houses would need those big DB.

Heard from rumours that somebody installed theirs for only 600 bucks! Now that got me worried...

Anyone has experience in this?

Cisco VPN client & Windows XP SP2

kons — Wed, 18 Apr 2007 22:46:21 +0800

I'm trying to get Cisco VPN client running in the Windows XP, updated until the latest patch available.

But failed.

Gave me blue screen everytime it's negotiating security parameters..
Something like error 0x0000007F(00000008,....

Anybody has experience in fixing windows installation problem?

mymirror.asiaosc.org is down?

kons — Wed, 28 Feb 2007 08:39:32 +0800

Anyone has insider details?

Ah. Back to crappy download days.

Stamping

kons — Thu, 28 Dec 2006 18:06:17 +0800

Hi All,

Does anyone know where we can do the stamping in M'sia?
Is it in Income Tax office?
Thanks!

radius and 802.1x

kons — Sun, 10 Dec 2006 21:47:24 +0800

hey guys..

recently we were tinkering around with 802.1x authentication with couple of different radius servers.
some radius actually act as the main authentication and accounting servers, while some of it act as a proxy.
the reason we are putting it a proxy in middle, is because we want to achieve load balancing architecture in authentication.

some radius, while interpreting attribute 81, private tunnel id, they treated 0x00 tag value as no tag, and proceeded to remove the 0x00 tag in the attribute value.

some radius, they can only operate with tags in the field. They can't interpret the field if there is no tag inside it.

radius rfc does not really give any logical idea nowadays.
guess nobody is following the rfc much anyway.

we have a hard time trying to figure that out.
all radius replies return ok, but the user client just said "authentication failed".
argh! do you know how frustrating that can be.

so I think it's a good thing to post it in here so those deploying 802.1x solution anywhere in the world to keep a look out out for that.

should we treat 0x00 as nothing? or should we really put 0x00 in the tag?

they really should decide on that.

that's all guys, ciao!

Pix

kons — Wed, 08 Nov 2006 17:54:13 +0800

Hi guys,

Any of cisco certified expert in here? Especially in pix..
I have a question here though...

I'm pretty used to access-list, access-group in Pix 5.2 and above..
Now I've encountered an ancient Pix running version 4.x

It seems the normal commands are called conduit...
How do I bind those conduit thing into a device_id?
Don't see access_group as an option though...

Thanks!

P/S: Any of you has CCO login?

Cheers!

[Intro]Linux & Open Source Software

kons — Wed, 01 Nov 2006 04:30:01 +0800

First of all, we would like to welcome you to this Linux and Open Source Software subforum in LYN.

General LYN rules applies here as well. So please play by the rules.

We accept anything and everything OSS is in here.
Tips, tricks, guides and more.
Cheats, hints, hacks and more.
Guides, questions, problems and anything.
Any distro, any platform.
As long as it is related to OSS.

Any suggestions is welcome!

That's all for now.

Note: If technical help is needed on Linux/Unix variant, please specify the distro used and the version of the distro.

1000 points gromac

kons — Mon, 04 Sep 2006 10:33:19 +0800

[17:53:31] Assembly optimizations on if available.
[17:53:31] Entering M.D.
[17:53:42] Protein: p1487_DPPC_DOPC_CHOL
[17:53:42] Writing local files
[17:53:43] Extra SSE boost OK.
[17:53:43] Warning: long 1-4 interactions

Woohoo...
Anyway, fold on everyone!

Maybe it was a mistake..

SATA HDD

kons — Fri, 01 Sep 2006 12:45:18 +0800

Hi all tech support specialists..

How are you guys? Hope everything's fine.

I just flew in back from other side of the earth..
And discovered that my BIOS can't detect two of my SATA HDD anymore.

Both are WD HDD though, plugged into my good old Intel 945GNT mobo.

I tried to plug in other SATA HDD and it works fine, BIOS can detect and windows can detect, but the rest of the HDD are powered 24/7 but just this 2 are plugged into a PC that has been switched off.

It was working fine before I left M'sia... And both HDD were bought only last year.

Anybody encounter this strange strange problem before?
Is it that SATA must be powered on 24/7 in order to prevent it from failing?

Yikes...
Thanks all!

How stress is your linux box?

kons — Fri, 11 Aug 2006 17:44:45 +0800

CODE

#cat /proc/cpuinfo
processor � � � : 0
vendor_id � � � : GenuineIntel
cpu family � � �: 15
model � � � � � : 4
model name � � �: Intel(R) Xeon(TM) CPU 3.40GHz
stepping � � � �: 3
cpu MHz � � � � : 3391.839
cache size � � �: 0 KB
fdiv_bug � � � �: no
hlt_bug � � � � : no
f00f_bug � � � �: no
coma_bug � � � �: no
fpu � � � � � � : yes
fpu_exception � : yes
cpuid level � � : 5
wp � � � � � � �: yes
flags � � � � � : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm nx lm
bogomips � � � �: 6763.31

CODE

5:42pm �up 86 days, �3:52, �1 user, �load average: 3.33, 2.91, 2.56
1600 processes: 1586 sleeping, 8 running, 0 zombie, 6 stopped
CPU states: 95.4% user, �4.5% system, �0.0% nice, �0.0% idle
Mem: �2058376K av, 2041220K used, � 17156K free, � � � 0K shrd, �121364K buff
Swap: 2040244K av, � �6140K used, 2034104K free � � � � � � � � 1199736K cached

Just spamming.
Err, would I get a warning for this?

ping problems..

kons — Wed, 12 Apr 2006 16:54:47 +0800

Hi guys,

How are you? Hope everything's good.
I'm not sure whether this is the right section for this question, but I'll give it a try anyway.

I'm trying to connect a few equipments that are having public IPs.
A modem, a PC with two NICs, and an AP.
Sounds easy, right?

The PC has two interfaces.
eth0 = 64.49.235.x range. (public)
eth1 = 192..168.0.x range (private)

The modem is having the public IP in the range of
64.49.235.x

And the AP is having an IP in the range of
64.49.235.x

I have connected the modem to eth0, (works great) and the AP to eth1 (works fine).

I can ping the AP if I'm using Netgear and Cisco AP..
But I can't ping the AP if I'm using 3Com ones..

If I connect the APs to eth having the same range of IP, I can ping all of them, no problem.

I've even tried
route add -host IP dev eth1 but it won't work.
Yikes.
Is 3Com worse than Netgear?
any advice?
Thanks.
Muaks.

rpmbuild

kons — Wed, 29 Mar 2006 12:04:11 +0800

I need to build a rpm for the kernel version that is different from the currently running kernel version...

uname -r gives me 2.4.21-27.0.4.EL
Put the kernel in /usr/src/linux-2.4.21-27.EL
I have kernel-source-2.4.21-27.EL installed.

I tried rpmbuild --define 'kernel 2.4.21-27.EL' but it wouldn't work.
It gave the error, kernel source tree not found.

What's wrong?

Thanks in advances.
Muaks!

HTML pages. Layout Problem.

kons — Tue, 21 Mar 2006 14:47:30 +0800

Hi all..

I'm facing some problems here...
I've done a few pages using just pure <table> and <img> tags in HTML.
No <div> at all.
I control all the dimensions using "height" and "width" property, no CSS at all.
But the layout runs riot in Firefox while it displays perfectly in IE...
Just a few millimetres but it's awful enough.

Any of you guys have tips to solve this problem?
Thanks a lot though.
Muaks.

Argh. Linux HDD woes..

kons — Mon, 19 Dec 2005 08:37:49 +0800

Hi guys,
rumours has it that there's no disk surface scan utility for Linux.
So the best tool lying around in the barn is fsck already?

I have this failing HDD which I think has developed bad sector syndrome.

No matter how many fsck's I've injected into it, the cloning software just can't get pass the bad sector block.

Err, can we use Windows Disk Surface Scan utility to scan the Linux Partition? Would it cause the Linux ext3 to turn belly up?

I've been using Symantec Ghost.. any other super cloning software that can support GRUB and ext2/3 to recommend?

Thanks guys.
Muaks!

Fedora Core 4

kons — Fri, 26 Aug 2005 09:12:58 +0800

Hail to all Fedore Core 4 users out there.

Something is wrong with the kudzu in FC4!
When I type kudzu in the prompt, it will ask me for the root password.
I'm sure that I typed in the correct root password.
Yet it says "Error creating netlink socket (Permission denied)".

I've tried resetting the password twice, and it will won't bulge!

I've search through bugzilla, and redhat/fedora forums.
Somebody else reported the same thing too but no one was able to solve the problem.

No updates is available for kudzu yet.

Any ideas, guys?

Thanks in advance.

Error on Tinker..

kons — Tue, 02 Aug 2005 18:55:45 +0800

[04:05:31] Finished a frame (400)
[04:05:31] TINKER is Exiting following Normal Termination
[04:05:31]
[04:05:31] Finished Work Unit:
[04:05:32] Header on frame 85 differs from expected header
[04:05:32] Got: 197046D-8C-42E7B903-213ADC4D: 0085
[04:05:32] Expected: 197046D-8C-42E7B903-213ADC4D: 0084
[04:05:32] Error: ARC file integrity could not be confirmed. Exiting
[04:05:36]
[04:05:36] Folding@home Core Shutdown: BAD_FRAME_CHECKSUM
[04:05:37] CoreStatus = 70 (112)
[04:05:37] + The core could not validate the current work unit for processing.
[04:05:37] Deleting current work unit & continuing...

This problem has been occurring randomly on one of the folding rigs.
Not all the time, but like once in 4 or 5 times?

I just want to know.. is this normal?

Went to folding forum and saw Vijay Pande himself is not sure of the problem too. Gee.

Bash error?

kons — Mon, 18 Jul 2005 15:36:36 +0800

Hello all open source experts out there..
Mind to enlighten me on what is this?

CODE

login as: root
Sent username "root"
root@192.168.1.23's password:
Last login: Mon Jul 18 15:40:18 2005 from 192.168.1.188
-bash: grep: command not found
-bash: grep: command not found
-bash: id: command not found
-bash: [: =: unary operator expected
-bash: id: command not found
-bash: id: command not found
-bash: id: command not found
-bash: [: too many arguments
[root@sip ~]# cd /usr/sbin
[root@sip sbin]# ls
-bash: ls: command not found
[root@sip sbin]# locate ls
-bash: locate: command not found
[root@sip sbin]# dir
-bash: dir: command not found
[root@sip sbin]# cd /usr/bin
[root@sip bin]# ls
-bash: ls: command not found
[root@sip bin]#

Thanks in advance.
I'm googling for it too.
I'll keep you guys posted as well.

filtered ports

kons — Fri, 13 May 2005 17:26:22 +0800

nmap gave me this..

707/tcp filtered unknown
1023/tcp open unknown
1072/tcp filtered unknown
5554/tcp filtered unknown
9996/tcp filtered unknown
12345/tcp filtered NetBus
12346/tcp filtered NetBus
20034/tcp filtered unknown
31337/tcp filtered Elite

netstat -lnp doesn't show me any process is listening on those ports.

what's happening to this poor box?