udpをport forward

udpをssh越しにはport forward出来ないということで
ncコマンドでudp-> tcp --| ssh |-- tcp -> udpという風に橋渡しさせるという方法。

[ port forward ]
ssh server_address -L 4800:localhost:4800

[ local server ]
nc -l -u 3939 < /var/tmp/squid_fifo | nc localhost 4800 > /var/tmp/squid_fifo

[ remote server ]
nc -l 4800 < /var/tmp/squid_fifo | nc -u 127.0.0.1 3939 > /var/tmp/squid_fifo

*上記はFreeBSDでの例です。
---
参考サイト:　http://zarb.org/~gc/html/udp-in-ssh-tunneling.html

pipe of death

::1 - - [04/Jul/2006:11:00:47 +0900] "GET / HTTP/1.0" 404 198 "-" "Apache/2.2.2 (Unix) (internal dummy connection)"

| This function connects to the server, then immediately closes the
| connection.
| This permits the MPM to skip the poll when there is only one listening
| socket, because it provides a alternate way to unblock an accept()
| when the pod is used.

| The pipe of death is used to tell all child processes that it is time
| to die gracefully.

qmail/queue/lock/trigger

久しぶりに、また作り忘れてしまった。trigger。

---
kinoko# pwd
/var/qmail/queue
kinoko# cd lock
kinoko# ls
sendmutex tcpto
kinoko# mkfifo trigger
kinoko# chown qmails:qmail /var/qmail/queue/lock/trigger
kinoko# chmod 622 trigger
kinoko# ls -l /var/qmail/queue/lock/
total 4
-rw------- 1 qmails qmail 0 Feb 9 20:06 sendmutex
-rw-r--r-- 1 qmailr qmail 1024 Jun 24 00:37 tcpto
prw--wx-wx 1 qmails qmail 0 Jun 20 11:28 trigger

kernel: NETDEV WATCHDOG: eth0: transmit timed ou

kernel: NETDEV WATCHDOG: eth0: transmit timed out
kernel: e1000: eth0: e1000
-----

TCP Segmentation Offload (TSO)
http://www.sys.cs.tuat.ac.jp/~shina/memo/nic.html

# ethtool -i eth0
driver: e1000
version: 5.6.10.1-k2
firmware-version: N/A
bus-info: 0000:02:01.0

# ethtool -k eth0
Offload parameters for eth0:
rx-checksumming: on
tx-checksumming: on
scatter-gather: on
tcp segmentation offload: on
----
# ethtool -K eth0 tso off
# ethtool -k eth0
Offload parameters for eth0:
rx-checksumming: on
tx-checksumming: on
scatter-gather: on
tcp segmentation offload: off

下記、どこのサイトで拾ってきたかわすれちゃった…。
----
１．modules.confの[alias eth? e1000]の下に以下を追加
# vi /etc/modules.conf

options e1000 RxIntDelay=0,0

２．再起動
# reboot

３．再起動後、下記のコマンドを実行
# ethtool -K eth0 tso off
# ethtool -K eth1 tso off

４．上記コマンドは再起動すると無効になる為、rc.localに以下を追加

# vi /etc/rc.local

if [ -x /usr/sbin/ethtool ]; then
/usr/sbin/ethtool -K eth0 tso off
/usr/sbin/ethtool -K eth1 tso off
fi

----

ssh key

ssh-keygen -b 1024 -t dsa

root password変更

http://lantech.geekvenue.net/chucktips/jason/chuck/1002317496/index_html から

-----------
[Boot in single user mode]

As the operating system is starting, it will display the following message:

Hit [Enter] to boot immediately, or any other key for command prompt.

Booting [kernel] in 10 seconds...

You should now press the space bar, and you will see the following message:

Type '?' for a list of commands, or 'help' for more detailed help.

ok

Now type boot -s and press the enter key to start FreeBSD in single user mode. After the system boots, you should see the statement:

Enter full pathname of shell or RETURN for /bin/sh:

Press the enter key and you will have a # prompt.

[Mount the filesystems]

At the command prompt, issue the mount command. This command will mount all the filesystems listed in your /etc/fstab file.

# mount -t ufs -a

[Change the root password]

Issue the passwd command and you will be prompted to enter a new password for the root account.

# passwd

New password:_
Retype new password:_
passwd: updating the database...
passwd: done

# exit

There is no need to reboot the operating system to go to multi-user mode. The exit command will cause the system to return to multi-user mode without a reboot.

cisco: active? standby?

login
en
show standby brief