酷!學園

Current envoirnmemt:
-current domain controller is Winnt4.0 server (PDC)
-nt4 workstation & win2000 workstation domain


After Migration:
-a new server installed win2003srv
-orginal server (nt server) will stop service
-win2003srv be domain controller with AD
-workstation are winnt & win2000pro


Now,,i had built a nt server & 2003 server for testing,the plan is:
-install 2003srv be member server
-join orginal nt domain
-install AD,promote 2003srv to be Domain controller
-demote orginal nt server
-workstation join AD


But...it seems no work...
-during the configuration of 2003srv AD,only can  join existed AD or create a new AD
-but,my orginal domain is a NT envoirnment,no a AD....


Question:
-is possible to migrate form NT domain to AD envoirnment?or how to do the migration?
-existed domain workstation necessary to join the AD again?


Thanks for all suggestion and helpful.

主機是一台LAMP,人一多就扒(or可能其他原因)
所以想做個負荷測試,看看那是不是request多就掛...
thanks.

這兩天,每天4:05am都出現一些message:

Kernel:Directory sread (sector 0x20) faild
Kernel:attempt to access beyond end of device
Kernel:02:00:rw=0,want=16,limit=4

是否i-node or data block出問題?
有人遇到嗎?

Remark:
-kernel 2.4.6-10
-沒有raid
-每天4:05左右出現message(自15Oct2003)
-這接近的cron job是 cron.daily(4:02am,做logrotate等等)
-出現這message前沒有其他error message

有部Windows2000 member server裝了IIS and ftp server enabled. (上了domain)
但login時一定要userID + domain ID,format: xxxx/xxxx.
The xxxx before forward slash (/) is their domain, after is username.

請問iis可否設當login一定是用domain account,令ftp login時可以不再需要輸入domain ID.

Thanks.

問個有點白爛的問題...
Description:
-login to opwebmail,把整個inbox download了
-不知做了甚麼,整個inbox的mail全部沒有了

Action:
-把之前download的inbox(file)放回/var/spool/mail/$user
-chown,chmod /var/spool/mail/$user

Result:
-重新login入openwebmail,依然沒有mail...
-check /var/spool/mail/$user,mailbox變回0KB....

Question:
-download了出來的inbox(file)可以import回openwebmail嗎?
-如果不能,如何能取回inbox(file)裡的file attachment?

Thanks.

請問有沒有軟件可以模擬網絡系統?
原有網絡:(附圖1)
1)有150台windows workstation
2)整個網絡很簡單的用lv2 switch連接
3)沒有core switch or router

將來:
1)再加100台,total 250
2.1)我建議加一個core switch增加整體效能,做網絡重心 (附圖3)
2.2)但間vendor就只打算在原有的switch加fiber module (附圖2)
2.3)管理層就打算用vendor的solution,因為平嗎...(說深一層就是落tendor是以價錢為主,效能為幅,那些vendor理你死活)
3)沒data,自討苦食.有沒有模擬網絡系統可以有些實數來証實?
(我對那規劃舊有網絡那vendor已很不滿,為錢理你網絡的效能死活)


請各位給些意見.
p.s:我認為windows的網絡40~50workstation好了,再多..multicast多,collision detection多,那效能就不好了.再加上用lv2的switch做core..真的有點....
可否給些預測,加了lv3 core switch及沒加,對未來的效能及影響有多大?

知識尚淺,請多位多多指教.Thanks.


附圖1:
這是舊有的網絡,由某間xxxx公司做的:



附圖2:
這是加了新網絡,間xxxx公司的solution:



附圖3:
這是加了新網絡,我的規劃:

兩個file的其中一段code:
select.html
<TR><TD width="150">Period: </TD><TD width="350"><INPUT TYPE=text NAME="period" SIZE=40></TD></TR>

<TR><TD width="150">Country: </TD><TD width="350"><INPUT TYPE=text NAME="country" SIZE=40></TD></TR>



result.php

$query_string = "select * from $tablename where country='$country' and period='$period'";



select.html有個table做search (period & country),variable會pass入result.php查詢mysql的db.
我想做的是:當user在select.html查詢時,如國period or country其中一個沒有鍵入資料,那result.php處理時就把當成"所有"
e.g. country=hongkong  and period=*

問題1:如何令select.html空白欄default值就當 "所有"
問題2:如何在mysql裡查詢所有,where country="$country' and period='*' 行不通..

p.s我是個新手,多多指教.Thanks.

zone alarm聽就聽多了,但前幾天幫人裝時才發現也不是很易用...
他有兩張NIC,一connect internet,一connect LAN.之後按了"LOCK"後,連LAN的connection也給lock了...
請問可否要它只lock其中一張NIC?

寄去News.list.request@tlug.sinica.edu.tw的,出事了..

This is the Postfix program at host tlug.sinica.edu.tw.

I'm sorry to have to inform you that the message returned
below could not be delivered to one or more destinations.

For further assistance, please send mail to  

If you do so, please include this problem report. You can
delete your own text from the message returned below.

              The Postfix program

: Command died with status 1:
   "/home/ycheng/zope-service/zope251/lib/python/Products/MailBoxer/smtp2zope.py
   http://liy.slat.org/mailing/News/manage_mailboxer 200000". Command output:
   Traceback (most recent call last):   File
   "/home/ycheng/zope-service/zope251/lib/python/Products/MailBoxer/smtp2zope.py",
   line 27, in ?     from quopri import decodestring as qdes ImportError:
   cannot import name decodestring

 
Final-Recipient: rfc822; news.list@tlug.sinica.edu.tw
Action: failed
Status: 4.0.0
Diagnostic-Code: X-Postfix; Command died with status 1:
   "/home/ycheng/zope-service/zope251/lib/python/Products/MailBoxer/smtp2zope.py
   http://liy.slat.org/mailing/News/manage_mailboxer 200000". Command output:
   Traceback (most recent call last):   File
   "/home/ycheng/zope-service/zope251/lib/python/Products/MailBoxer/smtp2zope.py",
   line 27, in ?     from quopri import decodestring as qdes ImportError:
   cannot import name decodestring

終於聽到mp3了...
之前也沒真正用x-window做desktop,看大家玩得這甚開心,也安裝來玩到
安裝時只是monitor的refresh rate要小心些...
但入x-windows後就要惡戰了...
用的是VIA AC87 onboard audio,rh8.0原來的根本就不能用...甚麼device no found...
之後去VIA的網站下載source來安裝,complie時又話kernel version不對(它只有for 7.3的..)
之後就去下載ALSA的source來安裝...甚麼...complie時又忘了出了甚麼error.
最後用OSS的driver才順利聽到很響亮的一聲~~~i like it~~~
那個xmms又要個xmms-mpg123的plup in(幸好在這裡看過,不然又要浪費一堆時間..)
之後順利裝了realplyer8
現在還在苦戰xcin...rpm又在一大堆lib的相依...
之後應該還有一大堆問題要應付吧...
還不太習慣x-window的控制...

有部webmail server,當user由outlook or 其他MTU傳送一封是BIG5編碼的題目BIG5編碼的附件的e-mail到這webmail server時.這webmail server看到的是BIG5編碼的題目(正常),但附件卻變成亂了的GB碼(=?gb2312?b?t3O8V6TlpnKk5aXzLnR4dA==?=)...
請問是那裡出了問題??
Thanks your helpful.

近日又在整vlan,這次有些新進展,但還遇到些問題,想各位能指點指點~
先來一幅簡單的架構圖吧,看招:請移玉步,大了些請認內

解說:
-10.117.144.68是server(Vlan3),10.117.145.45(Vlan4) and 10.117.145.53(Vlan6)是3個不同的vlan.
-10.117.144.68<--->10.117.145.45 Vlan3可連接Vlan4
-10.117.144.68<--->10.117.145.53  Vlan3可連接Vlan6
-10.117.145.52<-X->10.117.145.53 Vlan4不可連接Vlan6

Switch上的設定:
#member of vlan3:(10.117.144.68)
Unit1 Port18 PVid

#member of vlan4:(10.117.145.45)
Unit1 poer20 Port
Unit1 Port18 802.1q

#member of vlan6:(10.117.145.53)
Unit1 poer17 Port
Unit1 Port18 802.1q

#10.117.144.68的設定:
#modprobe 8021q
#ifconfig eth0 0.0.0.0 up
#vconfig add eth0 4
#vconfig add eth0 6
#ifconfig eth0.4 10.117.144.68 broadcast 10.117.145.255 netmask 255.255.254.0 up
#ifconfig eth0.6 10.117.144.69 broadcast 10.117.145.255 netmask 255.255.254.0 up

#兩部workstation不用設定.

測試:
[root@10.117.145.45]#ping 10.117.144.68
echo-request,echo-reply正常.

[root@10.117.145.53]#ping 10.117.144.69
沒有任信息....

[root@10.117.144.68]#tcpdump -i eth0.6
10.117.145.53 > 10.117.144.68 icmp:echo request
10.117.145.53 > 10.117.144.68 icmp:echo request
10.117.145.53 > 10.117.144.68 icmp:echo request

[root@10.117.144.68]#tcpdump -i eth0.4
arp who has 10.117.145.53 tell 10.117.144.69
arp who has 10.117.145.53 tell 10.117.144.69
arp who has 10.117.145.53 tell 10.117.144.69

[root@10.117.144.68]#route -n
Destination    Gateway Genmask Iface
10.117.144.0 0.0.0.0 255.255.254.0 eth0.4
10.117.144.0 0.0.0.0 255.255.254.0 eth0.6
127.0.0.0 0.0.0.0 255.0.0.0 lo


依tcpdump的結果看來:
10.117.145.53 ping 10.117.144.69 由eth0.6這vlan interface進來,但卻試由eth0.4回覆,由於不同vlan,所以得不到回應,以至client機一直在等待,連host unreachable的回音也沒有.


問題:
所以想請問應該在那裡下刀?我個人估計可能是那linux server的routing問題,但卻想不出如何作.

註:
-沒有router or L3 switch,只有linux做gateway,希望能以linux無限的創造力去完成~~
-switch是3com super stack 3300 lv2 switch
-測試NIC是eepro100

參考:
802.1Q VLAN implementation for Linux
http://www.candelatech.com/~greear/vlan.html

LINUX VLAN + Cisco HOWTO(請問這篇是介紹把linux接上原有的vlan,還是拿linux來做router去連接其他vlan?看了幾次還不太明...)
http://www.candelatech.com/~greear/vlan/cisco_howto.html

3com network administrator gulid

Thanks.有勞各位了.

有user report話他有個 .bmp file不能delete,我也試用administrator login入去,真的是access deny,就連take ownership也是access deny
出現這些error:
cannot delete z.bmp
Make sure the disk is not full ow write protect and that the files is not current in use.

我看過attribute沒甚麼特別,也確定沒人在用那file,我已把整個folder設為不給任何人存取兩天了.
我懷疑會不會是在nt 本身jam住了.
請問有甚麼好方法?(不用reboot就可以了)

我好想要呀~~~好想看呀~~~(聽的看來沒機會吧...)

為何一個讀computer science,考左MCP的人連安裝windowsNT,開account也不懂...為何要我由頭教過佢....到底那些xyz公司如何請人...>_<
我快吐血了