January, 06

January, 07

January, 08

January, 09

January, 11

It's a stupid question, but I had to ask this to a friend today in the morning, and I explain you why. During this week I went outside the office, have my lunch  and after some walk I found a place which serves "Jacket Potatoes". I picked my fillings: Bacon and Cheddar. For my surprise I got my potato with bacon, and a yellow cheese, very similar to our muzzarela in Brazil. I though that I had asked it wrong, or the may be the guy didn't understand well what I asked.

Some days after, I bought a sandwich at Tesco... Hamburger and cheddar, and again there was the yellow cheese.  This time it was not possible to be my mistake, it was written in the pack!

After a long talk about these events I knew that Orange Fanta is not orange, it's yellow! (and not good, don't taste!).

Beer is much more easier... Here you can have Guinness, paying the price of a "national" beer, it's great!

The cooker of my apartment looks like a spaceship panel, and I had no time to read the instructions yet and learn how "to fly" the machine, That's why I'm buying  just fast food.

If you have some tip about how to eat better in London, they are all welcome=)

Cya!

I do believe that have some notes written in some place about this trip will be much more funny to me than to who reads this posts, I really hope you like it, but if  don't please don't look for me to complaint about.

After 4 days, the major feeling is that I'm trapped in Brazil. May be explored would be the right word. And I'm not talking about the work, I'm talking about our government. This city has amazing things that I will not live to see in Brazil. The subway (tube) is really amazing, its possible to reach anywhere in a few minutes. The subway cars are very clean and the punctuality (notice the cliche) is Britain!

It's hard to explain how funny is the feeling of  to cross a street, look to the cars and have the impression that they are driving by themselves. Or look at the parked cars and have the impression that the passenger  is seated, waiting for the driver. I though that the right side wheel was a issue which would affect just the driver, but it's not! By the way, I'll get dead crossing these streets, because I always do look, the wrong way.  And the worst part is, when I get used to look to the right side, when I come back to Brazil, then I'll be surely get hit, since I'm sure no Brazilian drivers will stop, and wait for me cross the street =)

People from Curitiba, you are sympathetic! At least when compared to Londoner people. Take this as complement or insult, it's up to you).

London also has its "Curinthia" (Okay, I have to explain a little bit this part for non-Brazilian people). We have a soccer team in Brazil named "Corinthians", but the name is too complicate for their cheering.  So they say the name the best way they can) . The "Curinthia's brother team is the Tottenham,  says the legend that they are not able to win in the in the elimination phase of the Champions League.

Last Sunday I went to Greenwhich park (GMT-ZERO hehehehe). It is huge, but I didn't walk so long, I was kind of worried, thinking that I could have lost my cellphone. In the end I just forgot it at home.

I bought a digital cam, it should arrive next Friday, then I can post some photos!

Cya!

This is the scenary:

• Area 0: R1 and R2 (Area backbone)
• Area 1: R2 and R3 (Area configured as virtual link)
• Area 2: R3 and R4 (Area connected to ISP and nssa)
• Area 3: R1 and R5 (Area totally stub)

The lab is fully functional, 4 blocks /24 (192.168.0.0/22) where added to loopbacks in the Area 1 and another 4 in the Area 2 (192.168.4.0/22) for summarization routing tests purposes. Follow below some notes that could be useful:

1. Virtual Link is always configured between ABR routers in the area that will be utilized as transit.
2. Areas with Virtual Link can not be stub or not so stub.
3. NSSA areas do propagate summary routes and default route.
4. Totally stub areas just receive default routes.

You can download and enjoy the project here: GNS3 LAB - OSPF Multi Area . If you have any question, please let me know, send me a message!

By the way, I've used the Cisco IOS: c2600-i-mz.123-22.bin

I hope you liked it!

This is the scenario:

• Routers running RIP: R1, R2, R4 and R5
• Routers running OSPF (Area 0): R1, R3 and R5
• R1 has both routing protocols and it is the responsible to redistribute they.
• R4 and R5 are interconnected through RIP too, and the default behavior of R5 is try to reach R4's networks through OSPF routes (longer path), because OSPF has a better administrative distance than RIP.

The lab is fully functional, with routing redistribution working. It's possible to check the cust increment for each redistributed route received on R4 and R5 (OSPF just increment the cust when usint external type 1). But the biggest trick to observe in this lab is the administrative distance manipulaton on R5, that is what fix the suboptimal routing decision.

You can download and enjoy the project here: GNS3 LAB - Route redistibution between RIP and OSPF .
By the way I've used the Cisco IOS c2600-i-mz.123-22.bin

I hope you liked it!

The problem seemed simple, figure out which process was binded on this port, check if it was an stuck instance of the same service previously ran or a legitimate service and then take appropriate action.

Usually the netstat command has a parameter to inform the process associated to the port, it's -p on Linux, -b on windows and so on. But, what about Solaris? Damn! Solaris's netstat hasn't this feature.

I found away various ideas to solve the problem, but all they using one of two approaches, lsof or pfiles. The lsof isn't built-in on Solaris, so only left me pfiles.

The below script does the dirty job, isn't a pretty thing to grep a bunch of proc files, but it works (if you have access to do that).

#!/bin/bash
# Get the process which listens on port
# $1 is the port we are looking for

if [ $# -lt 1 ]
then
echo "Please provide a port number parameter for this script"
echo "e.g. $0 22"
exit
fi

echo "Greping for your port, please be patient (CTRL+C breaks) ... "

for i in `ls /proc`
do
pfiles $i | grep AF_INET | grep $1
if [ $? -eq 0 ]
then
echo Is owned by pid $i
fi
done

I hope it helps you too.

It's easy to deploy, easy to use and Cisco recommends to use in the Guide to Harden Cisco IOS Devices, so why do not use it?

It relays on SSH and AAA, so both features have to be enabled on the device:

Router(config)#hostname MRXROUTER
MRXROUTER(config)#crypto key generate rsa general-keys modulus 512

% The key modulus size is 512 bits % Generating 512 bit RSA keys, keys will be non-exportable...[OK]
MRXROUTER(config)#
MRXROUTER(config)#aaa new-model
MRXROUTER(config)#aaa authentication login default local
MRXROUTER(config)#aaa authorization exec default local

In order to use scp to manage configuration we must have an user account with enough privileges to access it:

MRXROUTER(config)#
MRXROUTER(config)#username admin privilege 15 secret 0 topsecret

Finally, we can turn the scp server on:

MRXROUTER(config)#ip scp server enable

On the client side we can use an utility such as pscp from PuTTY suite (windows) or the native scp client on Linux, to interact with our SCP server – the router!

C:\>pscp.exe PuTTY Secure Copy client Release 0.59
Usage: pscp [options] [user@]host:source target
pscp [options] source [/source] [user@]host:target
pscp [options] -ls [user@]host:filespec

For example, we can download the startup-config and put it on a directory (a backup?):

C:\>pscp.exe admin@192.168.0.42:nvram:startup-config C:\MyConfigs\MRXROUTER.cfg
admin@192.168.0.42's password: MRXROUTER.cfg                    | 0 kB |   0.6 kB/s | ETA: 00:00:00 | 100%

C:\>

It's already time to left tftp rest in peace

References
Cisco.com: Cisco Guide to Harden Cisco IOS Devices
Cisco.com: Cisco Secure Copy (SCP) Feature Guide – 12.2T
 PuTTY: A Free Telnet/SSH Client

In linux the directories that should be included by ldconfig to create the links/cache dynamic libraries are listed in /etc/ld.so.conf. Build a new library and install it into a directory already listed in this file is part of cotidian, and a simple "ldconfig" it's enough to rebuild the links/cache. But on OpenBSD the path to be searched is the ldconfig's argument, and if you don't inform this path... well, you probably already got it.

Don't panic!

/usr/lib will remain linked so everything you really need, still working, which doesn't include X, bash, among other things.

Moreover ldconfig is properly ran when the O.S. boots, so a simple reboot would solve the problem. But if you think, (like me) this isn't the wise way to solve the problem you can do it by yourself:

# . /etc/rc.conf
# ldconfig /usr/local/lib /usr/X11R6/lib $shlib_dirs
# exit

and be happy !