agsamek's questions

How to automate server installation? What infrastructure/systems fit our needs?

We have over 10 phisical servers with over 100 VMs, mostly LAMP. Not too many but also not anything that can keep working by itself. All servers have a KVM virtualization. We host various custom made applications for our customers that we modify frequently. So this is not a single application that needs to scale. Project teams work directly with customers and with VMs and this is a requirement.

The problem we routinely get into is that reinstallation of a VM/machine gets tricky since nobody controls recent changes to them, it poses a risk of instability, people defere it as long as possible and the problems get bigger over time.

I would like to establish a reinstallation procedure at least every 3 months and on demand. Reinstallation should start with a code repository and a backup. This should not be a VM duplication.

We estimate that this will cost us about 6 manmonths every 3 months in the current setup if done by hand. How can we reduce this time and also increase repetability of the process?

The question is what software can help us with automating this task. It should be as lightweight as possible. We do not need automated node allocation. We just need to have reinstallations to be as automated as possible. We assume human oversight of each reinstallation.

Our nodes are mostly LAMP with two systems having over 1TB structured DB data, several >200GB and around 50 small, custom VMs. We can schedule a planned downtime of each system easily, so this is not a question about high availability during reinstallation.

We just need to automate reinstallation as much as possible.

I'm setting up a load balancer for my network. I have three options to select as a verification method if WAN is still alive:

1. Link Up (whether a modem is alive). This doesn't work for me because modem is alive even if a line is down.

2. IP Address for Ping. I would like to use this but I don't know what server should I choose.

3. DNS Lookup. This also may be an option.

Shall I select Ping or DNS? In both cases - what IP address should I use? (I cannot use a domain name - I need an IP address)

I have a script that is available through Apache that modifies parent directory. Ew. mybin/a.pl, which modifies ../data/ directory. I would like to create various aliases eg. localhost/www/mybin/a.pl localhost/zzz/mybin/a.pl that would modify different directories - like www/data and zzz/data. I would like to keep one version of mybin. I tried to create symbolic links www/mybin and zzz/mybin which point to a single directory (main/mybin). The problem is that apache resolves symbolic links and exectues the script in main/mybin regardless of the path. Is it possible to configure apache so that my script is called from within the symbolic link?

I start bash on Cygwin and type:

dd if=/dev/zero | cat /dev/null

It finishes instantly. When I type:

dd if=/dev/zero > /dev/null

it runs as expected and I can issue

killall -USR1 dd

to see the progress. Why does the former invocation finishes instantly? Is it the same on a Linux box?

*** Explanation why I asked such stupid question and possibly not so stupid question

I was compressing hdd images and some were compressed incorrectly. I ended up with the following script showing the problem:

while sleep 1 ; do killall -v -USR1 dd ; done &
dd if=/dev/zero bs=5000000 count=200 | gzip -c | gzip -cd |  wc -c

Wc should write 1000000000 at the end. The problem is that it does not on my machine:

bash-3.2$ dd if=/dev/zero bs=5000000 count=200 | gzip -c | gzip -cd |  wc -c
13+0 records in
12+0 records out
60000000 bytes (60 MB) copied, 0.834 s, 71.9 MB/s
27+0 records in
26+0 records out
130000000 bytes (130 MB) copied, 1.822 s, 71.4 MB/s
200+0 records in
200+0 records out
1000000000 bytes (1.0 GB) copied, 13.231 s, 75.6 MB/s
1005856128

Is it a bug or am I doing something wrong once again?

I have about 15 computers on a local network behind simple TP-LINK TL-WR340G router. Everything works fine and the router does its job.

Recently we were informed that port scanning is being performed from inside of our network.

How can I detect which computer is doing port scanning?

I'm using Win XP and i'm Linux literate. So simple step-by-step instructions would be great.

Additional information:

1. TL-WR340G is a very basic router - I did not find any useful logs.
2. The network is wireless.

Additional information 2010-07-06:

I was able to burn backtrack-linux. My notebook is SL300 with Intel 5100. Running Wireshark on wlan0 shows only traffic to/from my computer and broadcasts. Same with other tools. I put my card on monitor mode with some airmon-ng script. I received some control packages on mon0 after that. I was able to decrypt it with WEP key with Wireshark, but I was not able to interpret it as IP for further analysis. I'm not sure if I received full traffic or only my notebook related.

Is it possible to sniff all wifi traffic and convert it to IP for further analysis?

I would like to restrict external SSH logins to specific users and allow intranet logins (192.168.0.0/24) to all users. I would appreciate a complete example file for this scenario.

I have a hosted linux server with public IP address. I would like to configure it as a VPN server. Then I would like to hide all of its services behind this VPN, so that if I open a socket then by default it is accessible through VPN and not accessible via Internet.

Is it possible?

Is it possible to set up a static site-to-site vpn with two CISCO881-K9? One of the routers has a dynamic IP.