Category Archives: Hardware

NTP Spoof Attacks

Apparently 2 out of 3 of our GPS clocks were being used in a spoof/DDOS attack.

The basic premise is that time requests are sent to the service with a spoofed IP for the response. Add in 100/1000/10000 clocks send results to a server/service that did not request them results in a denial of service attack.

Both clocks have been removed from the pool, scheduled for February 6th. Hickory.kulish.com has port 123 UDP closed at this time. The maintainer of Dickory.kulish.com has been notified.

Partitioning in Windows 7

Needed to remove junk from a USB thumb drive. Google search yielded a nice find at http://www.bleepingcomputer.com/forums/t/219693/unable-to-format-usb-stick-to-full-capacity/

1) Type “DISKPART” from the Command Prompt (accessible by clicking on Start and then typing “cmd” into the open field); you will then see the following prompt: DISKPART>

2) Type “LIST DISK” to see what number your USB drive is listed as.

3) Type “SELECT DISK 2″ (if your USB is disk 2; replace # with your disk #); Diskpart will confirm that “Disk 2 is now the select disk.”

Debian Jumbo Frames

2 NAS servers both with 802.3ad bonded gigE nics based on the Realtek 8169 chip.
The highest MTU I could set was 7000 even though the D-Link DGS-1210-24 Rev. A switch can support up to 10k.

Below is just a single sample, but all tests stayed within 57x Mbits for MTU=1500 and 77xMbits for MTU=7000.

The important bits.
iperf was used for this testing.

MTU 1500:
[ 3] 0.0-10.0 sec 687 MBytes 576 Mbits/sec


MTU 7000:
[ 3] 0.0-10.0 sec 926 MBytes 777 Mbits/sec

Barracuda Green ST2000DL003

http://www.seagate.com/files/staticfiles/support/docs/manual/desktop/Barracuda%20Green/100649225c.pdf

Reformatted a bit to fit on page.

I have 6 of these (2TB) in each NAS Raid5 setup plus motherboard, AMD cpus, RAM etc.
This page is for researching a replacement PSU.

2.8.1 Power consumption
Power requirements for the drives are listed in Table 2 on page 15. Typical power measurements are based on an average of drives tested, under nominal conditions, using 5.0V and 12.0V input voltage at 25°C ambient temperature.

• Spinup power Spinup power is measured from the time of power-on to the time that the drive spindle reaches operating speed.

Debian Squeeze 802.3ad

Debian Lenny and Squeeze
2x Realtek 8169
1x Reaktek 8169, 1x nForce
D-Link DGS-1210-24 Rev. A

Testing:
iptraf
cat /proc/net/bonding/bond0