在线等!急!在linux下安装9i建立数据库的时候总是蹦出来"invalid entry size(expected 429924352 but got 429924342"?

[shasha_lover_527]

如题!各位大侠快点来啊!

[shasha_lover_527]

没人理我?

[huiwsey]

你在什么环境中安装的？到安装到哪一步出的问题，我安装过10i,没出过这个问题，我的环境是虚拟机，red hat Enterprise

[shasha_lover_527]

red hat Linux 4.0

[wariob]

你那问题解决了没有！我也遇到这个问题啦！
我的是WIN SERVER2000+orale9i（正版），全新的，从2000到XP都试了！为什么创建数据库时候会出现这种错误

还有在安装是会出现写入 D:\oracle\ora92\sqlj\lib\translator.jar 错误之类的情况 有时候装一个oracle 要丢失几十个文件，现在装这次只出现三次 是最成功的一次。装完了创建数据库
就出现 “invalid entry size ”

要是解决了我把我的300分全送！

[wariob]

今天终于在一家英文网站上找了答案 说是一个BUG，手工建吧！
Caveats
Some of the issues that I encountered are reported to
have been corrected in Oracle10g, although I haven’t yet
experimented with the DBCA in 10g to verify the fixes.
After the cloning process completes, it may be
necessary to UNLOCK and RESET users’ passwords.
You can use OEM to verify and unlock any users.
Also, the listener and TNSNAMES file must be
configured via the Net Configuration Assistant.
During an initial attempt at creating a template, I
received the following

error: Invalid Entry Size (expected
3130007552 but got –1164959744 bytes).

The process
stopped at the UNDOTBS01 file creation, which was
created on the development platform with a size of
2.91GB.
After checking with Oracle, it turned out to be a bug
(2536772) in 9.2. On MetaLink, note 209516.1 lists a
workaround, which is to reduce the datafile size to less
than 2GB if possible.
According to reports, the bug has been fixed in 10g.
It’s not possible to fix this issue in 9i DBCA because of
Java’s limitation with respect to its Zip classes.
As an option, if you’re considering using the
template feature to clone databases, you may want to
create your primary database with multiple .dbf files
for tablespaces under the 2GB size, prior to building
any templates.
And finally, I encountered another small problem
with DBCA. When using DBCA to create a template
from a running DB, the database is locked. After the
template creation process is completed, the DB remains
locked. There’s no notification given prior to or after
the process completes. According to Oracle, this appears
to be expected behavior. A bug was filed on this, and
development engineers said that it was expected behavior
and wouldn’t be fixed. However, they did suggest that at
some point they might put in a confirmation message
before the shutdown of the database occurs. So, the only
change in the behavior expected to be added is that there
will be a message to let the user know that the database
will be shut down.