Configuring Remote access at PIX

Cisco Easy VPN

Application Overview

The Cisco Easy VPN Remote feature and the Cisco Easy VPN Server feature offer flexibility, scalability, and ease of use for site-to-site and remote-access VPNs.

 
 

EzVPN client has two operation mode:

1. Client mode (PAT mode) : the client end performs PAT for all its traffic for its host inside.
   
2. Network extension mode : the client end makes its inside nodes routable across outside network. PAT don't apply to VPN traffic. Inside nodes get IP address statically or assigned by DHCP .
   

    
    

    
    

   I will demonstrate how to achieve Remote IPSec VPN access in PIX.
   

   IKE Phase1
   

   crypto isakmp enable OUTSIDE ======== invoke ISAKMP on outside interface
   

   crypto isakmp policy 1 ======== IKE Phase proposal
   

   authentication pre-share
   

   encryption des
   

   hash md5
   

   group 7
   

   lifetime 86400
   

    
    

   Define address pool from which clients obtain IP address.
   

   ip local pool POOL 10.0.0.1-10.0.0.3
   

   Create a user
   

   username joey password cx15108
   

    
    

   IKE Phase2 ( negotiate IPSec parameters)
   

   crypto ipsec transform-set IPSEC_SET esp-des esp-md5-hmac
   

   crypto dynamic-map DYN 1 set transform-set IPSEC_SET
   

   crypto map MAP 1 ipsec-isakmp dynamic DYN
   

   crypto map MAP interface OUTSIDE
   

    
    

    
    

    
    

   tunnel-group TUN type ipsec-ra
   

   tunnel-group TUN general-attributes
   

   address-pool POOL
   

   tunnel-group TUN ipsec-attributes
   

   pre-shared-key *
   

    
    

    
    

   A dialogue window pops up
   

   
   

    
    

   Verify the result
   

   IKE phase 1 CA :
   

   
   

   
   IPSec CA :
   

   
   

    
    

   VPN client routing table shows that the default path .
   

   
   

    
    

Install Gentoo

Install Gentoo from CD

I used the latest Gentoo iso that I had laying around, the 2006.0 x86 LiveCD. I opened the laptop, plugged it in, turned it on, and put the CD in the DVD/CD-RW. About 20 seconds later, the boot: prompt came up.

boot: gentoo doscsi

The doscsi flag was more just me being paranoid, knowing that the T60p uses Serial ATA and libata adds those as scsi devices. You can probably do without it, I never bothered as I only went through this process once. When the liveCD booted and X started up, I clicked on the graphical install wizard, and went with mostly all the default (or recommended) options. I always resync, recompile, and reconfigure after this step anyways. I chose an all GRP install (networkless) since that allows me to get past this step the quickest and used the liveCD kernel (no sources). One thing that I noticed right away about the wired ethernet (e1000 driver) is that if I don't setup the eth0 device in the install wizard, the device never gets bound. I've also read in other places some minor issues with the Intel Corporation 82573L Gigabit Ethernet Controller like it getting dropped or the device not getting bound after the e1000 module gets loaded. Other than configuring eth0 and saving, I went straight through the install then a reboot sans LiveCD.

 
 

 
 

Booting off the harddrive

I want to get to this point of the installation as quick as possible, since if things break here, then you've got serious issues. Otherwise, you're mostly smooth sailing with the exception of the few iffy drivers (ipw3945, fingerprint reader, ATI FireGL V5200).

Since the liveCD kernel doesn't come with sources, I need to emerge them. I wanted the most recent sources so I did a sync first:

# emerge --sync

Then I choose gentoo-sources (with the ibm stuff), and also chose this time to update genkernel and portage:

# emerge portage genkernel # USE="ibm" emerge gentoo-sources

The next thing, configure my kernel. The liveCD kernel is genkernel'ed, so I'll just stick with that (although with udev in and coldplug out, there's no reason to use genkernel anymore):

# genkernel --menuconfig all

Here is my kernel .config file, I mainly just turned the majority of the random drivers on or made them into modules. I also turned on all the Bluetooth subsystem and the drivers (bluetooth), kismet, SYN Cookies and netfilter/iptables since I use that stuff. Another thing you may want to consider turning off is the Direct Rendering Manager, as I read the ati drivers won't jive with it. I also used the kernel ALSA drivers for sound. Turn on ALSA support, and turn on Intel High Definition Audio if you want to use the kernel drivers, otherwise just turn on ALSA support (and you'll need to emerge alsa later). After the kernel and modules are finished compiling, update your boot loader (I used grub, and here is my grub.conf). Remember that your drive will be /dev/sda because of SATA.

UPDATE: I'm using the 2.6.19 kernel now and the newest ipw3945 drivers no longer requires you to remove the kernel ieee802.11 stuff. I also updated the kernel config file for the 2.6.19-r5 kernel. The SATA drivers are now built into the kernel so make sure to turn them on as well as Inotify (should be on by default) so that udev (replaces coldplug) can automatically keep the /dev nodes up to date.

 
 

 
 

Installing drivers and software

At this point, you're going to need a network. If you configured eth0 (e1000 wired ethernet), net.eth0 should be started automatically. Before you start emerging stuff, make sure you setup your USE flags. Here is my make.conf file, and you can see my USE flags.

After that, I wanted to get off the wire and get on my wireless, so I did the following:

# emerge ipw3945 ipw3945d wireless-tools

I choose to use wireless-tools instead of wpa_supplicant, since I already have the /etc/conf.d/wireless file from my old laptop. Here's a link to gentoo's wireless How-To, it walks you through both iwconfig and wpa_supplicant. I then created a net.eth1 and added it using rc-update:

# cd /etc/init.d # ln -s net.lo net.eth1 # rc-update add net.eth1 default

Then I reboot just for safe measure, remembering to remove the ethernet cable from the laptop. Coldplug automatically found my wireless so I didn't need to add that to the modules.autoload.d list. The ipw3945 driver is still a little flakey. Currently, I have issues from time to time when I bring down eth1 (802.11 wireless) and bring it back up (usually on a different network). The 802.11 a/b/g LED flickers really fast and it hangs on * Configuring wireless network for eth1 for a little while before failing. /var/log/messages reports: ADDRCONF(NETDEV_UP): eth1: link is not ready, which doesn't seem that unusual (it reports that even if bringing up eth1 doesn't fail). I have a dodgy work around for this when I don't feel like restarting the machine: shutdown ipw3945d, get rid of the ipw3945 and the ieee80211 modules, and reinsert them. (Note: These directions don't quite work for the newest versions of the gentoo portage tree and the ipw3945 driver, see the UPDATE at the bottom)

# ipw3945d --kill # rmmod -f ipw3945 # rmmod ieee80211_crypt_wep # rmmod ieee80211 # rmmod ieee80211_crypt # modprobe ipw3945 # /etc/init.d/net.eth1 start

We need to first shutdown ipw3945d properly using the --kill flag. Then we need to force remove the ipw3945 module, this is probably unadvisable, but it's being used by kernel processes (nice=-5) and this is the only way to get rid of both the processes and the module. You can then remove the other 802.11 modules and reinsert ipw3945. This will bring up ipw3945d as well as the other 802.11 modules and I'm able to start net.eth1 without any problems.

UPDATE: I recently discovered the purpose of the little switch next to the IR port is for (I usually don't read manuals). In case you didn't know already, this turns on/off wireless. It also happens that when I turn it off, /etc/init.d/net.eth1 (my 802.11 interface) stops automatically, and when I turn it on, it starts automatically. This completely alleviated the issue of the device hanging on startup. I'm uncertain what packages/modules I installed for that to work, but if that requires a module, it should be thinkpad and/or ibm-acpi. So now, I never issue a start/stop to my wireless device anymore, and instead, just use this handy little switch. UPDATE: The script above doesn't really work anymore in the more recent versions of the ipw3945 driver. The ipw3945d daemon is now started using an rc script. It's still a little finicky when starting up and connecting to an AP, but the worst case is to stop the daemon, make sure the ipw3945 module is unloaded, then restart the daemon using the rc script. This doesn't work 100%, and it kinda sucks. UPDATE: But it seems the most recent ipw3945 daemon fixes this problem. So this was what I ended up having to do: Firstly, remove net.eth1, ntp-client, ntpd, sshd, and whatever network services you have automatically startup using rc-update. The only thing you need is to make sure /etc/conf.d/net (all the wireless config got moved to this file, so you no longer need to fiddle with /etc/conf.d/wireless) is setup correctly and that ipw3945d is started. The daemon seems to take care of bringing wireless up when one of my APs I have configured in /etc/conf.d/net is in range. As for the network services, I added this to the end of my /etc/conf.d/net: postup () { /etc/init.d/ntp-client restart /etc/init.d/ntpd restart /etc/init.d/sshd restart return 0 } So now whenever I plug my ethernet cable into the jack or if I get in range of my AP, my network services automatically get started. You will need to make sure udev has the persistent-net rules setup for your ipw3945 device, otherwise it won't come up automatically. I have this in my /etc/udev/rules.d/70-persistent-net.rules file: # PCI device 0x8086:0x4227 (ipw3945) SUBSYSTEM=="net", DRIVERS=="?*", ATTRS{address}=="[your MAC address]", NAME="eth1"

I wanted my laptop to have all the things my old laptop had, so I just took the world and emerged everything all in one go. Here is my world file as well as an equery list. I took the /var/lib/portage/world file from my old laptop, removed the Thinkpad T42p specific drivers (like ipw2200) from the list, and renamed it old_world. My emerge command looked like this:

# emerge --update `grep -v '^ ' old_world | sed -e s/'\-[0-9].*'/""/g` # emerge --update world

This removes all the random non-package lines (-v '^ ') and removes the version info at the end of each package (s/'\-[0-9].*'/""/g). The second emerge command makes sure the stuff that wasn't part of my old_world gets updated as well. I didn't run into any problems except for a few packages that got pruned out of the dep tree (just emerge --unmerge them) and the change in the name that slocate uses. You can add a -vp to check for the former.

Now that I have nearly everything installed and my new T60p is a spitting image of my old T42p, I'll bring up a few random things:

• I prefer SSH.org's implementation over openssh's. That's as simple as unmerging openssh and emerging net-misc/ssh.
  
• I emerged all the bluez and gnome-bluetooth stuff, although I'm not sure if the bluez stuff is needed
  
• If you didn't compile ALSA drivers into the kernel, emerge alsa-driver and stuff, then rc-update add alsasound default
  
• You may also want to emerge thinkpad and tpctl (and most likely acpitool so that hotplugging works for thinkpad specific stuff)
  
• I emerged ati-drivers, but the version that I got (8.21.7-r1) isn't one that works with this graphics card
  

UPDATE: As of a few months ago, SSH.com's license for versions after 3.9.2 no longer fall under the 'free for non-commercial use' agreement. Therefore, gentoo has masked it and we'll all have to use openssh now. Recently another ebuild was added into portage that may be use for thinkpad users. The tpb package allows you to bind commands to miscellaneous buttons on your thinkpad. You may want to emerge that as well.

 
 

 
 

Fingerprint Reader

This was one thing that I really wanted to get working. There's this fingerprint reader siting just below the keyboard and it would suck if I wasn't able to use it. This took a bit of trial and error, but you want to start here: http://toe.ch/~tsa/ibm-fingerprint/. There are 3 ebuilds here that you want from there and they're located in the overlay URL under sys-auth: sys-auth/bioapi, sys-auth/tfm-fingerprint, and sys-auth/pam_bioapi. You can read up on the Gentoo 3rd party ebuilds How-To if you want, but basically you want to stick these in your /usr/local/portage/sys-auth/ directory. Once you have everything it should look like this:

/usr/local/portage/sys-auth/bioapi/ /usr/local/portage/sys-auth/bioapi/ChangeLog /usr/local/portage/sys-auth/bioapi/Copyright /usr/local/portage/sys-auth/bioapi/Manifest /usr/local/portage/sys-auth/bioapi/bioapi-1.2.2.ebuild /usr/local/portage/sys-auth/bioapi/files /usr/local/portage/sys-auth/bioapi/files/20bioapi /usr/local/portage/sys-auth/bioapi/files/digest-bioapi-1.2.2 /usr/local/portage/sys-auth/pam_bioapi/ /usr/local/portage/sys-auth/pam_bioapi/pam_bioapi-0.2.1.ebuild /usr/local/portage/sys-auth/pam_bioapi/Manifest /usr/local/portage/sys-auth/pam_bioapi/files /usr/local/portage/sys-auth/pam_bioapi/files/pam_bioapi.c-0.2.1.patch /usr/local/portage/sys-auth/pam_bioapi/files/digest-pam_bioapi-0.2.1 /usr/local/portage/sys-auth/pam_bioapi/files/digest-pam-bioapi-0.2.1 /usr/local/portage/sys-auth/tfm-fingerprint/ /usr/local/portage/sys-auth/tfm-fingerprint/files /usr/local/portage/sys-auth/tfm-fingerprint/files/tfmessbsp.cfg /usr/local/portage/sys-auth/tfm-fingerprint/files/digest-tfm-fingerprint-1.0 /usr/local/portage/sys-auth/tfm-fingerprint/tfm-fingerprint-1.0.ebuild /usr/local/portage/sys-auth/tfm-fingerprint/Manifest

Before you emerge these 3 ebuilds, the bioapi package NEEDS the qt USE flag. There are 2 versions of the program that creates the fingerprint file, a graphical version (QSample) and a command line version (Sample, but (as of 6/11/2006) the command line version doesn't allow you to enroll a fingerprint. In order to get the graphical version, you need qt. So this is what I did:

# USE="qt" emerge sys-auth/bioapi sys-auth/pam_bioapi sys-auth/tfm-fingerprint

You can pretty much follow the directions on the website from here on. Everything works as advertised. You may need to play around with /etc/pam.d/xscreensaver a bit in order to get that working to your satisfaction. I created a /etc/pam.d/bioapi:

auth required pam_env.so auth sufficient pam_unix.so likeauth nullok auth sufficient pam_bioapi.so {5550454b-2054-464d-2f45-535320425350} /etc/bioapi/pam/ auth required pam_deny.so account required pam_unix.so session required pam_limits.so session required pam_unix.so

and added an include in the xscreensaver pam.d file. Consult the pam.d man page for more options.

UPDATE: There's a new ebuild in the portage tree that's a different implementation of the fingerprint reader than this one called sys-auth/thinkfinger. It also provides a pam module that you can include in services you want to be able to authenticate with a fingerprint. This one doesn't require the qt package and having both the bioapi and thinkfinger installed at the same time doesn't seem to break anything.

 
 

 
 

Other stuff

X11

I mentioned earlier that the ati-driver that's part of the most recent portage tree doesn't work with the ATI Mobility FireGL V5200 card in the T60p. I didn't bother with a more recent version since the current work around requires the ChipId 0x71c5 in the xorg.conf file to trick the driver into thinking you've got a Radeon X1300, and I'm willing to wait for a real driver to be released. Currently, I just have what Xorg -configure outputted, which uses vesa as its driver. The only annoying thing about it is I can't get overlays to work.

UPDATE: I found some linux proprietary ati drivers, version 8.25.18-1, that supported the FireGL V5200. After I installed it, I removed the ChipID option in my xorg.conf and changed the driver to fglrx. Now I get GL accelleration, but I still can't get XVideo to work, even when I turn the option on. UPDATE: The new ati driver was much slower than the vesa driver for some reason. So I switched back to the vesa driver. Oddly, glxgears now runs at 880fps compared to 200 with the ati driver. UPDATE: XVideo is working now. The most recent (as of July) gentoo portage tree has Xorg X11 7.0-r1, that put together with ATI's most recent FireGL drivers (Thanks much to Lance Warner) 8.26.18 from ATI's website, I have X Overlay now. I ran a couple of other tests (like glxgears) and GL doesn't seem to be working all that much faster than with the vesa driver, but I don't use hardware GL very much; I was mostly concerned with the X overlay. For the specifics, the most recent emerge --sync required me to unmerge xorg-x11 (among a few other things like mozilla), and the emerge --update world put the newest X11 libs as well as xorg-x11 7.0-r1. Afterwards, I ran the ati-driver-installer-8.26.18-x86.run file that I got from ATI's website, making sure that the older drivers were removed. One problem with this is that it blows away whatever Gentoo's portage installs, but I guess that can't be helped. It's something you just have to remember the next time you see an error when using portage. UPDATE: The last --update world performed (November) upgraded X11 to xorg-x11 7.1.1 and the proprietary ati driver version 8.26.18 no longer works. But fortunately the newest driver available on the ati website (8.29.6) works with the newer version of xorg-x11. X Overlay still works, seems to be a little faster than before, but the DRM is broken and I get no opengl. Maybe I'll mess with this some more when I get bored. UPDATE: After the most recent kernel upgrade, I decided to try the ebuild instead of downloading from the ati website. I unmasked the most recent ebuild for ati-drivers (8.35.5) and emerged it. Everything works fine. DRM/DRI works, opengl works fine, Xvideo works fine, and on top of that, all the ATI control panels now work.

 
 

Intel Pro/1000 Gigabit Ethernet

I didn't have any problems with this for a while, but the other day I restarted after compiling the kernel while experimenting with the ati drivers and the wired ethernet adapter is nowhere to be found. I looked through dmesg and got this:

e1000: 0000:02:00.0: e1000_probe: The EEPROM Checksum Is Not Valid e1000: probe of 0000:02:00.0 failed with error -5

The Intel support says to check the driver. So I check the driver by recompiling my kernel with e1000 as a module. When I modprobe, I get the same message. The Intel support then says to run diagnostics. I didn't know of any so I decided to take things into my own hands (WARNING, THIS COULD BE DANGEROUS). I opened up /usr/src/linux/drivers/net/e1000/e1000_main.c and commented out 2 lines in the block of code that checks the EEPROM:

/* make sure the EEPROM is good */   if (e1000_validate_eeprom_checksum(&adapter->hw) < 0) { DPRINTK(PROBE, ERR, "The EEPROM Checksum Is Not Valid\n"); /* err = -EIO; goto err_eeprom; */ }

This way, I'll still get the error message in dmesg, but the driver loads successfully, and gigabit ethernet works again. I know that this is dodgy, but I was impatient as there were other things that I wanted to deal with.

UPDATE: Thanks to udev, you can have network come up automatically when you plug in an ethernet cable. You need to make sure you have this in your /etc/udev/rules.d/70-persistent-net.rules file: # PCI device 0x8086:0x109a (e1000) SUBSYSTEM=="net", DRIVERS=="?*", ATTRS{address}=="[your MAC address]", NAME="eth0"

 
 

Bluetooth

With the Bluetooth subsystem compiled into the kernel as well as the bluez packages, everything seems to work. When I turn on the bluetooth device (Fn + F5), /etc/init.d/bluetooth automatically starts and I can scan devices. When I turn off the bluetooth device, /etc/init.d/bluetooth automatically shuts down. I haven't had the chance to talk to another device or transfer any files yet.

 
 

ACPI

I disabled all the CPU frequency scaling and suspend/hibernate, so I'm not sure if that stuff works correctly. The ibm acpi controls are definitely working (light, beep, display, bay, etc...), but I'm not sure if it's tied to the hotplug agent correctly. I don't have another Ultrabay device to test this on.

 
 

USB

The USB seems to be working without any tweaking. I was able to connect an iPod up to the T60p, mount it, and copy files to/from it. I was also able to connect to multiple different digital cameras and mount them as scsi devices.

 
 

IrdA

Not tested yet.

 
 

PCMCIA

Not tested yet.

 
 

Modem

Not tested yet.

 
 

WAN Wireless

Not tested yet.

UPDATE: I did however notice a few kernel upgrades ago that the Sierra Wireless USB driver is in the kernel now, under Device Drivers->USB Support->USB Serial Converter. I compiled it into the kernel but haven't tested it because I don't have an account with Verizon. Also, thomasgersdorf.com has a detailed walkthrough for getting this working.

Countries should help each others?

Many people think that countries have a natural obligation to help each others, while other argue that the aid money is misspend by the governments that receive it , so the international aid money should not be spent to poor countries where the governments are corrupt. What's your opinion ?　　　

With the process of economic globalization, many countries are seeking to strengthen mutual collaboration in the fields of industry , finance , economic , education , agriculture and trade. In order to gain mutual benefits , many countries spare no effect to help undeveloped countries by means of technology , medical , education and finical aid . it's a accelerating agent to global economic development.

In my points of view , aid from other countries or international organizations could help them develop their domestic economy. If they could efficiently utilize it , aid money would have tremendous effect. Due to the backwardness of technology and economy , people living in those areas or region such as Asia , Africa , south America suffer a great ideal from poverty , famine and drought .

International aids give a good medical care and help promote hygienic condition in the poor countries . every years , a great number of people died of malaria , cholera and smallpox , which almost had been eradicated in developed countries many years ago. With the help of the world health organization , these deadly disease , which have devoured millions of lives every years , will be eliminated soon and people will survive from them .

After all , aids to undeveloped countries or developing countries could in turn benefit the donators. This is a win-win strategy , which create an effective , high profitable producing environment , firms in developed countries have to fact several problems in domestic market such as high producing cost due to the limitation of nature and human resource , so it is hard for them to successfully seize more proportion of the market through reducing the cost of products . Internet creates a global platform which excerts a profound influence on the traditional business model . the notion of global manufacturing factory has become realistic with the assistance of INTERNET. So that , firms setup their factories in developing countries with low tax , low cost . meanwhile , it also help ease the tension of unemployment in some developing countries and increase their national revenue. So many factors have driven the process of flattening the world .

Nevertheless , corruption usually grows with poverty . more and more people are showing great concerns about the utilization of the international aids. A rigorous regulation should be enact to supervise how the money are being used ? also , the countries received the aids should cherish such great opportunity .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Feminism

'virtue' too often involves the renunciation of pleasure in the name of some higher purpose , a purpose that involves power for men and sacrifice for women . pleasure is describe as a shallow and frivolous in a world of high-minded , serious purpose . but pleasure does not exclude serious pursuit and intension .it is the only real reason for staying alive.

Recently there has arisen a heated debate over whether zoos are necessary ? people retain diverse views on this hot issue , before presenting my view , I intend to discuss both sides of the argument.

Advocates of zoos hold the beliefs that the benefits of the zoos to any community are educational , recreational . first of all , it is an effective /practical way to preserve those rare animals and endangered species in zoos . it is a fact that the natural environment is no longer a suitable habitat for some species to survive and some species are on the brink of extinction due to the deterioration of environment and hunting. If we don't take some actions to protect them , the ecological balance will be broken . finally , it will exert a dangerous influence on human beings . zoos help increase the population of endangered species . moreover , zoos also serve as an ideal scientific center where scientists can conduct researches which could give us more information about these species so that we can come up with a practical method to preserve them . in addition , besides protecting endangered species , zoos play an important role in improving public knowledge about animals . last of all , zoos are a place of entertainment and recreation , there're also a tourist attraction which boost the local economy .

 

Some other people , however , are strongly against zoos for several reasons . primarily , it is morally wrong that take an animal from its home , confining it in unnatural settings and use it for entertainment . the nature of zoo completely reflects human's arrogance towards other species and foster a disregard for all life . furthermore , worst of all , adult and children visiting the zoos are given the subliminal message that it is all right to use animals for our own ends and it is natural that we possess them . it is unethical that to ignore the freedom and the feelings of the animals . human being have already taken a lot from the nature and the animals . we should be such greed and indulgent , the animals and us should share the basic right living under the same sky .

 

In the final analysis , both sides have sound foundations . if asked to make a choice , I will no hesitate to assent to the latter. I believe animal belong to the their natural habitat , they deserve the freedom . it is a breach of their natural right to take them by force into captivity for our own purpose . so zoos should be discouraged .

I’m under the press of time

The graph illustrates the proportion of populations aged 65 and over in three countries : Japan( bitch) , USA, Sweden . from the above graph , we can see that the populations in these countries generally have an increasing trend.

The Sweden's percentage and the usa's percentage have the same increasing rate during 1940 to 1980 . During the following two decades THE percentage of Sweden rises from approximate 7.5 to about 14 and the percentage of USA jumps from 9 to 15. Meanwhile , the percentage of JAPAN drops to about 3 at 1960 and levels off. The Sweden's percentage passes the USA's percentage in 1998 , and then it keeps the same increasing rate ,finally reach a high point –20 . there's nothing changed in THE USA's percentage , but the Japan's percentage steadily grows . during 2020 to 2040 , three of all them increase at an amazing/extremely/shocking/soaring/ speed , especially Japan passes the rest of them and reached 27. Followed by Sweden with about 25 , usa ranks the last below 25.

According to the graph , we can find out an interesting situation that the rank of the percentages of the three countries at the beginning is just reversed at the end.

 

 

Nothing , but just want to supervise myself!

Ok. This time is a table.

The tables shows Australians company profit before tax during 1986 to 1990 , it lists four industries : manufacturing , mining, retail trade, other. As we can see from the table, it is noticeable that manufacturing had kept the top one during that period, mining ranked second , followed by retail trade . in general , all of them had a increasing trend except other.

The four industries had different trend in gaining profit during that period. Manufacturing's profit had a great/ rise between 1986 and 1988 from 4.650 to 6.616 million. It still kept the trend and jumped to 8.617 million during 1988 to 1998 . however , the trend didn't keep on going , it reversed from a great increase to a sudden drop during 1989 to 1990 , its profit fell to 7.810. as far as mining industry is concerned , its profit dropped from 3.840 to 3.540 during 1996 to 1989 and had a rise at the end of later 1980s . as to retail trade , even though its profit is the smallest among them , but its had kept a stable increasing rate during 1980 to 1998 , the profit jumped from 699 to 1900 million . the same as manufacturing , other industries had a sharp increase during 1980 to 1985 from 1000 to 5000 and then dramatically fell to 500 million .

We can conclude from the above table . over the four-year period , manufacturing , mining , retail trade had made a great achievement in gaining profit . but other industry experienced a dramatic fall in the year 1988-1999.

 

Keywords : at the end of later 1980s , in later 1998s , As to , as far as sth is concerned , over the four-year period .

Description of a pie graph

There is a pie graph that shows the proportion of the meat market. The total meat sold in the market has been classified into 6 types as follows: chicken , fish , lamp , beef , port , other.

From the pie graph we can find out that chicken is the most popular meat, which took up 40 percent proportion . then next is port with 20 , followed by beef , constituting 18 , and finally come lamp , fish , other at 15 , 5 , 2 respectively . it can concluded from the pie graph that chicken is the most commonly bought meat while others is the least bought meat.

Can you believe that the above composition is born from my hands ? it really really reminds me of my elementary school days. Right now A undergraduate is practicing how to write an article which he was supposed to do 10 years ago . what does this stand for ? you're just a pupil in the field which just enter. Of course . repeating truth is nonsense . there're some classical sentences which are helpful . the way to describe the percentage . Never fight with a man who buys his ink by barrel.

There comes another composition : two charts .

There're two charts illustrating the results of a survey on adult education . the bar charts shows/illustrates/depicts/reveals why people decided to further their education , while the pie graph illustrates how people surveyed think about the cost of adult education should be shared.

According to the bar charts , it is apparently that highest percentage of reason is "interest in subject" , which takes up approximately 40 and the lowest percentage is for " to meet people" , constituting 4 ,the reason "to be able to change jobs " is just 3 higher . the reason " to gain qualifications" ranks second , representing 38 . the third is the reason " helpful for current job" . it is interesting to notice that the reason" to improve prospects of promotion "and the reason of "enjoy learning " have the same proportion.

The pie charts reveal that there're three choices when the cost issue is concerned : individual , employer , taxpayer. The large part has been taken up by "individual", followed by "employer" . the rest is "taxpayer" . by this time , we can reasonably arrive at the conclusion that interest is the leading reason that why adult choose to further their education and people are willing to pay the cost of the study by themselves. Let's go further , interest is the finally instinctive promoting agent .

 

What I have done today . joey

No. today , an idea just came out from my mind. I want to put what I have done everyday onto my blog. So that I can reflect on my life. Record down every single step I finish. By the way , the blog is just dedicated to myself. Anyone if , you couldn't help running into here. Please don't mock me .