2005年01月18日

n AIX Advanced Interactive Executive, a flavor of UNIX from IBM

Corporation.

Aix 高级交互执行系统,

n Analyze A DDL command that collects or drops statistics on tables,

indexes, or clusters in a database. Also used to validate structures.

分析:一个用来收集或删除表、索引、簇信息的DDL语句,也可以用来验证结构

 

n Archive Log A copy of the redo log file created by the archiver process.

An archive log is used to recover the database from a previous backup.

归档日志 归档进程建立的重做日志的拷贝

一个归档日志用来恢复当前备份

n ARCHIVELOG Mode A database can be run in either ARCHIVELOG

mode or NOARCHIVELOG mode. In ARCHIVELOG mode, the database

uses an additional background process (ARCH) to create a copy of filled

redo log file at one or more pre-determined archive destination(s).

归档模式 数据库可以运行在归档和非归档模式下面,在归档模式下,数据量用一个附件的后台进程来建立重做日志的备份

n Archiver Process (ARCH, or ARCn) Oracle background process

responsible for managing archival of the redo log files. This process reads

the filled redo log file and copies it to one or more archive destinations.

归档进程

 

n Asynchronous I/O A specialized form of I/O that improves throughput.

Synchronous I/O takes place when processes wait for the operation to

complete. Asynchronous I/O operations run in background and processes

requesting it not to wait for its completion. Oracle takes full advantage of

asynchronous I/O on some platforms with certain types of data storage

implementations. Asynchronous I/O is also known as non-blocking I/O.

异步模式

 

n Auto RAID See RAID, Auto.

自动RAID

n Autotrace SQL*Plus command that can be used to automatically obtain

the execution plan and statistics for the SQL statement upon its execution.

自动跟踪

 

n Background Processes Processes performing specialized tasks on behalf

of all sessions. For example, the Database Writer (DBWRor DBWn) is

responsible for writing changed blocks from the buffer cache to the database

files. The log writer (LGWR) is responsible for writing blocks from the redo

log buffers to the redo log files. The archiver process (ARCH or ARCn) copies

completed redo logs to one or more predetermined location(s). With Oracle8

and up, a checkpoint process (CKPT) performs the checkpoints, reducing the

workload of the LGWR process. Other processes (such as SMON and PMON)

perform specific housekeeping functions such as recovery operations, clean

up of terminated sessions, and so on. Assorted background processes are

enabled only if certain Oracle options are enabled.

 

绑定

n Bind As applicable to the SQL statement execution, during the bind

phase, the values for any bind variables used in the statement, are resolved.

See also Parse, Define, Execute, Fetch.

n Bit Mapped Index See Index, Bit Mapped.

n Blocks The smallest unit of storage in an Oracle environment, ranging from

2K–32K. Most systems use 8K block size for transactional systems and 16K

block size or higher for data warehousing/decision-support systems. Larger

block sizes may be available with the 64-bit implementation of Oracle.

n Buffer A generic term that represents memory areas to hold information.

Such memory areas are defined in the initialization parameter file for the database.

n Buffer Pools Term used to define various memory areas, in the database

buffer cache. These are keep pool, recycle pool, and default pool. The

former two are defined using initialization parameters and are available in

Oracle8 and up.

n Cache An attribute on a table that puts the blocks in the MRU end of the

LRU list for the buffer cache.

n Cache, Database Buffer A portion of the system global area in the

memory that contains copies of data blocks, index blocks, rollback segment

blocks, and so on, from database files. It is configured primarily to reduce

device I/O by allowing multiple sessions to access frequently or recently

accessed data in memory.

n Cache, Dictionary A portion of the shared pool area within the system

global area that contains the information from the data dictionary. This

information is primarily used by recursive SQL.

n Cache Hit Ratio Percentage about accesses to the buffered information

in Oracle memory structures. We don’t believe in it for gauging the

performance of the database.

n Cache, Library A portion of the shared pool area within the system global

area that contains SQL statements and information about them. Oracle

searches this area to determine whether an identical SQL statement already

exists. If so, it will avoid hard parsing of the SQL statement.

n Checkpoint A procedure when the checkpoint process (CKPT) or the

log writer (LGWR) process writes synchronization information to all the

file headers.

检测点

n Checkpoint Process (CKPT) From Oracle8 and up, this background

process helps complete the checkpoint procedure reducing the workload

on the log writer process (LGWR). From Oracle8i, the CKPT process also

keeps a heartbeat of the instance/database with the control file every three

seconds. This helps determine the starting point of a recovery, if applicable

and when needed.

检测进程

n Contention A struggle between one or more processes to access the same

resource

 at approximately the same time.

竞争

n Correlated Sub-Query See Sub-Query.

子查询

 

 

n Cost Based Optimizer See Optimizer, Cost Based.

基于消耗的优化器,(相对的基于规则的优化器)

 

n CTD Compulsive tuning disorder—a state of mind and body that refuses to

stop tuning and cannot overcome the urge to “tweak just a little bit more.”

n Cursor An identifier that allows naming a SQL statement, access

information in the private SQL area of the statement, and, to some extent,

controls its processing. A cursor is said to be open when the SQL statement

is executed and the result is stored in the private SQL area.

n Database See Oracle Database.

n Database Block See Blocks.

n Database Buffer Cache See Cache, Database Buffer.

n Database Files Files that contain the data, index, rollback, and temporary

segments that comprise an Oracle database.

n Database Writer (DBWR, DBWn) Oracle background process responsible

for managing the database buffer cache and the dictionary cache. This process

writes the dirty, or changed, blocks from the buffer cache to the data files.

n Default Pool Actually, it is the database buffer cache and not a

specifically allocated area. With the introduction of other pools in the

database buffer cache, from Oracle8, the term default poolis used to

denote no special pool assignments for a given segment.

n Define As applicable to the SQL statement execution, during the define

phase, the user and server processes exchange data type information about

involved columns. See also Parse, Bind, Execute, Fetch.

n Enqueue A mechanism to track processes waiting to acquire lock that is

held by other processes. It also keeps track of the order in which processes

request the lock.

n Event See Wait Event.

n Execute As applicable to the SQL statement execution, during the execute

phase, the server process reads data blocks from the data files into the

memory (for update, delete, and insert operations) as needed. The data is

manipulated in this phase, as execution plan is put to work. See also Parse,

Define, Bind, Fetch.

n Extent A set of contiguous blocks within a datafile assigned to a specific

segment within that tablespace.

n Fetch For select statements, during the SQL statement execution, this

phase signifies reading the relevant data blocks into the database buffer

cache and applying the execution plan. Results are returned to the user

process. See also Parse, Define, Bind, Execute.

n Filesystem A block storage device that is mounted and used to stored

Oracle files.

n Fragmentation The breaking up of contiguous space within data files and

memory structures into smaller pieces. Also applies to unutilized space

within a database block, and to migrated or chained rows.

n Function Based Indexes See Index, Function Based.

n Hash Join See Join Methods.

n Hint User-chosen directive for the cost-based optimizer to follow certain

access paths while executing the SQL statement. The notation for a hint is

/*+ HINT */.

n Histogram A bar graph that shows the frequency of the data values. Oracle

supports constructing histograms for data values to aid the optimizer in

building optimal execution plans.

n I/O Input from or output to one or more storage device.

n Index An object that supports faster retrieval of data from a table, it contains

the information on which the index is created and pointers (ROWIDs) to the

data in the table. An index is stored in a B*- treedata structure.

n Index, Bitmapped Unlike a regular index where each data value in an

index has a corresponding ROWID that points to it, a bitmapped index

contains data values with a corresponding bitmap for a range of ROWIDs

to indicate which data value occurs in which row.

n Index, Function Based Introduced in Oracle8i, these indexes are built

on columns with one or more functions, such as upper, lower, round,and

364 Oracle Performance Tuning 101

so on. This allows the query to use this index rather than performing a full

table scan due the presence of a function on the said indexed columns.

n Init.ora Generic name for the Oracle initialization file that contains

various parameters with user-defined values.

n Instance See Oracle Instance.

n ISM Intimate sharable memory. Only on Sun Solaris platform, a specialized

method used to lock shared memory structures to improve their performance.

n Java Pool Available in Oracle8i and up, this memory structure in the

system global area is reserved for Java and its associated objects.

n Join The process of combining data from two or more tables (or views)

to produce the desired output.

n Join Methods A generic term for different algorithms the optimizer builds

to combine data from multiple tables or views. There are the following

types of join methods:

n Nested Loop A row operation that sends processed rows from one

step to the next step before completing processing of all rows in the

first step.

n Sort Merge A set operation that completes each step before rows

are sent to the next step.

n Hash Joins Usually involves full table scans. Oracle constructs a hash

table in the memory for the smaller table and probes it with rows from

the larger table.

n Keep Pool Available from Oracle8, it can be defined to contain data

from smaller tables or indexes. Data in this pool remain in memory,

eliminating the time (and I/O) required to re-read often-used data

blocks from disk.

n Latch A mechanism to protect a memory structure from concurrent

access by multiple processes. It is a specialized form of serializing access

to shared memory structures used by Oracle.

n Large Pool Available in Oracle8 and up, this memory structure in system

global area is reserved for special operations used by RMAN, parallel

query, and multi-threaded server implementations. It facilitates better

management and reduction of fragmentation of the shared pool area.

Appendix A: Glossary 365

n Library Cache See Cache, Library.

n Locally Managed Tablespace Available in Oracle8i and up, a tablespace

can be configured to self-managethe available space within it—hence,

“locally managed.” The space management tasks do not access the data

dictionary and the extents are all of the same size.

n Lock A mechanism used by Oracle to protect resources that a particular

process needs for a certain period. Other processes requiring the same

resource wait in line to acquire the lock on the resource, if the resource is

currently locked by some other process. Oracle uses various types of locks

and locking methods.

n Log Writer (LGWR) This Oracle background process responsible for

managing the redo log buffer reads blocks from the redo log buffers and

writes to the redo log files.

n LRU Least recently used.

n Metalink Name for the online support services available on the Internet

from Oracle; the URL for the Web site is http://metalink.oracle.com. You

must have support agreement with Oracle and your CSI number to register

to access this Web site.

n Mirroring A process of writing the same data to two members of the

same storage volume simultaneously, it provides protection for the data in

the event one of the members of the storage device becomes unavailable.

n MRU Most recently used.

n Multiversion Read Consistency Read consistency is the act of providing all

users with a consistent view of the data that was asked for. The multiversion

aspect provides that consistent view across multiple user sessions. In very

simplistic terms, it is a scenario where each user sees his or her own copy of

the data.

n Nested Loops See Join Methods.

n Offline A tablespace or database file that is unavailable for access.

n Online A tablespace or database file that is available for access.

n Open Cursor See Cursors.

n Optimizer An optimization procedure that usually chooses the most

efficient way to execute a SQL statement.

n Optimizer, Cost-Based A process to choose the most efficient way to

execute a SQL statement based on various statistical information available

on the objects. The optimizer computes the cost of each possible access

path and chooses the one with the smallest cost.

n Optimizer, Rule-Based A process to choose the most efficient way to

execute a SQL statement based on a set of pre-defined rules and ranks of

available access paths.

n Oracle Database A set of files—data files, redo log files, and control files.

n Oracle Instance A set of shared memory segments and background

processes that attach to it.

n Oracle Performance Management (OPM) Step-by-step process of

iteratively investigating, determining, and implementing tuning solutions

using a proven methodology.

n OTN Oracle Technology Network. This is another online resource on

the Internet from Oracle and is free to use; the URL for the web site is

http://technet.oracle.com. You will find online documentation and

various Oracle software to download.

n Paging An operating system process that writes memory pages to

predefined swap areas based on usage and demand for such memory areas

in physical memory. See also Swap Space.

n Parallel Query A query that is divided into multiple tasks, performed by

two more parallel query slaves.

n Parallel Query Slaves Processes that perform work on behalf of a

server process.

n Parse The step in the execution of a SQL statement where the server

process checks the syntax of the statement, performs object resolution,

and checks security. Upon successful completion of this procedure a parse

tree and an execution plan for the SQL statement is built. See also Define,

Bind, Execute, Fetch.

n Parse, Hard When Oracle builds the parse tree and an execution plan

for the SQL statement for the first time. However, there are other situations

when it will need to rebuild it, causing additional and undesired hard parses.

n Parse, Soft When, Oracle finds a matching SQL statement in the memory

with its already built parse tree and execution plan.

n Partitioning The process of dividing data into multiple smaller chunks for

ease of management, availability, performance, and security.

366 Oracle Performance Tuning 101

n PDML Parallel data manipulation language.

n RAID Stands for redundant array of independentor inexpensive disks,

depending on who you ask. However, it is the technology for expanding the

capacity of the I/O system and providing the capability or data redundancy.

There are various types of RAID configuration to choose from each offering

differing capabilities.

n RAID, Auto (Auto RAID) Hewlett Packard’s automatic RAID technology

that configures the I/O system based on the nature and type of the I/O

performed on the disk blocks within the RAID array.

n RAID 0 A very simple type of RAID providing good read/write

performance, it provides striping but no recoverability.

n RAID 1 Provides mirroring and full data redundancy. It is most often

called “mirrored disk” and requires twice the amount of disk storage as

compared to RAID 0.

n RAID 0+1 Stripe first, then mirror what is striped.” This RAID level

combines RAID 0 and RAID 1, and also provides good write and read

performance with data redundancy without the parity calculation overhead.

n RAID 1+0 Mirror first, then stripe what is mirrored.” This level of RAID

has the same functionality as RAID 0+1, but is better suited for high

availability requirements.

n RAID 2 This level of RAID incorporates striping. The redundancy and

data protection is provided through parity, and the parity overhead will

affect write performance. This is very rarely used, as other RAID levels

have since replaced it.

n RAID 3 Similar to RAID 2, it uses parity; however, the parity is stored on just

one dedicated disk. It is best suited for data mart or data warehouse applications

that support a few users but require sequential bulk I/O access patterns.

n RAID 4 Slight variation of RAID 3, in which parity calculations differ. It is

rarely used.

n RAID 5 The data redundancy is provided via parity calculation but the parity

is stored along with the data. Thus, the parity information is distributed across

the number of disk drives in a storage device. It offers good performance for

read operations; however, write intensive applications will encounter slow

performance due to parity calculation and distribution.

n RAID 6 A very rarely used level of RAID, the parity calculation uses a

much more complex algorithm. It also stores two sets of parity for each

Appendix A: Glossary 367

368 Oracle Performance Tuning 101

block of data and thus makes writes even slower than RAID 5; however,

it facilitates quicker recoverability in case of disk failures.

n RAID 7 A better implementation of RAID 3, and it offers better

asynchronous capability for reads and writes.

n RAID S EMC Corporation’s implementation of RAID 5/3.

n Raw Devices A character storage device that is an unmounted file system

but can store Oracle files.

n Recycle Pool Available from Oracle8, it can be defined to contain data

from larger objects. When not assigned to the recycle pool, these objects

may cause other objects to be aged out prematurely from the default pool,

thus causing additional physical I/O.

n Redo Log Buffer Buffer containing database changes that have not yet

been written to the redo log files. The contents of the redo log buffer are

periodically flushed to the online redo log files. They are always flushed

before completion of a commit, a checkpoint, or before a set of dirty data

blocks need to be written to disk.

n Redo Log Files Files that contain transaction journals that assist in the

recovery of a database in an event of a system, media, or database failure.

n Reserved Area An area in the shared pool, reserved to store large SQL

objects, including PL/SQL packages, procedures, function, and so on.

n RMAN Available from Oracle8 and up, the Recovery Manager (RMAN) is

a free tool set from Oracle to manage database backup and recovery tasks.

n Rollback Segment Segments that store original (or before-image) copies of

database blocks that have been changed by transactions. Rollback segments

contain the information that is utilized to undo a transaction, when you issue

a ROLLBACK command. These before-images in rollback segments will be

overwritten, as transactions reuse the space within a rollback segment.

n Rule Based Optimizer See Optimizer, Rule Based.

n SCN System Commit Number. Oracle uses this to represent the state

(incarnation) of the database at any given time. This number increments on

database changes caused by structural modifications to the database objects

or committed DML operations. See also System Change Number.

n SCSI Small computer serial interface.

n Segment An object in an Oracle database comprised of one or more

extents. Examples are tables, partitions, indexes, rollback segments,

temporary segments, and clusters.

n Semaphores A signaling mechanism used to send messages to other

processes. Oracle uses semaphoreson UNIX platforms to synchronize

concurrent processes that need access to shared resources.

n Server Process ”Shadow” processes that act as arbitrators by processing

SQL on behalf of user sessions, a server process can perform processing

for a single session only (known as a dedicated server) or for multiple

sessions (in the multi-threaded server configuration where the server

processes are shared).

n Shared Memory A memory structure that multiple processes can use to

interchange data and other information. Oracle processes access shared

memory segments and use inter-process communication (IPC) to exchange

information with one another.

n Shared Pool Area A portion of the system global area (SGA) that contains

library cache, dictionary cache, control structures, and, sometimes, private

session information (if multi-threaded servers are enabled).

n Sort A process to arrange data in a specific order.

n Sort-Merge See Join Methods.

n STATSPACK Available at no additional cost with Oracle8i, the STATSPACK

tool set offers a better facility than utlbstat/utlestat scripts to gather

performance related statistical information. This tool set makes it easy to

create and save performance information for future use and to perform

historical trend analysis.

n Striping A process of dividing data into pieces and distributing it across

multiple disks, these disks support a single logical volume. This process

facilitates increase in I/O bandwidth, improving read and write operations.

n Sub-Query A select statement (child) contained within another select

statement (parent). A sub-query is called a correlated sub-query when

the child statement is executed each time for every row returned by the

parent statement.

n Swap Space A special area on a storage device that is reserved for the

OS to write all the memory pages of an entire process when the system

experiences memory starvation. See also Paging.

n System Change Number Used with the sequence number (stored in the

block header), it determines the version of the data block to apply changes

to, for either a roll forward or rollback operation. See also SCN.

n System Global Area (SGA) A collection of shared memory areas that can

be accessed by multiple user sessions. The main components of an Oracle

SGA are shared pool area, database buffer cache, and redo log buffers.

n Table A collection of one or more rows and columns containing data.

n Table, Temporary See Temporary Table.

n Tablespace A logical structure that houses segments in a database, it is

usually configured to store segments of one type. A tablespace can consist

of one or more database files, but any given database file can belong to

only one tablespace.

n Tablespace, Locally Managed See Locally Managed Tablespace.

 本地管理表,不再用数据字典来管理

n Tablespace, Temporary Type See Temporary Type Tablespace.

表空间,临时表空间

 

n Temporary Segments These segments store sort data generated by joins,

group by, order by, summary functions, or create index commands.

n Temporary Table Available from Oracle8i, it is a global temporary table

临时表

that can be used simultaneously by multiple sessions as scratch space.

The rows in the table are visible to only to the session that inserted it,

and can be retained either until the end of the transaction or the end of

the session. There is no redo generated for any modifications to the data

in the temporary tables. Remember that this table will be created in the

temporary tablespace assigned to the user.

n Temporary Type Tablespace From Oracle 7.3, a tablespace can be

configured as purely temporary in nature. It improves the performance

of sort operations. Instead of creating temporary segments to be dropped

after the completion of sort a process, a temporary type tablespace keeps

the temporary segment for reuse by subsequent sort operations. This

minimizes space management operations during disk sorts, thus improving

sort performance.

n Tkprof Transient kernel profile. A free tool from Oracle, it analyzes trace

files to provide readable output that is meaningful to the DBA. There are

various options available to control what information is displayed and the

order in which it is displayed.

n Two-Pronged Approach A method to determine the current bottlenecks

in your system, which utilizes OS monitoring techniques and the Oracle

wait interface.

n Wait Event As of Oracle8i, there are about 200 wait events. A named

section of the Oracle kernel code, this concept was introduced in Oracle

7.0.12 and enhanced in each Oracle release thereafter. It is used to define

certain events or steps during processing that can be categorized as idleor

non-idlewaits. Non-idle waits are when Oracle processes are waiting for

resources to perform assigned work, while idle waits are when Oracle is

waiting for more work.

n Wait Interface A collection of wait event information from V$SYSTEM_

EVENT, V$SESSION_EVENT, V$SESSION_WAIT, and trace files generated

by the 10046 event number.

Appendix A: Glossary 371

 

2005年01月02日

     新年目标:每次都是那么动不动的就给自己立个目标什么的,按说这样给自己一个约束,也给自己一个奋斗的目标,这样不至于象今年一样到了年底写总结都不知道要说什么,能写什么,想到这里就想起“优秀是一种习惯”,关键在于坚持,技术方面也是如此,建立自己的知识库,代码库,积累很重要,整理也很重要,多不能代表什么,少而精才是王道,昨天今天新年都在出差中,机房中,不像空下来,却又找不到忙碌的事,脑袋一片空白进入深度抑制的状态,学而不思则罔,思而不学则殆,回顾和复习很重要,继续优秀是一种习惯,不会的事情,不想做的事情,先学会模仿,模仿久了就也成了自己的习惯,写字不是自己的强项,因为看的少,记得少,看书只是一带而过,只注重情节不注重用词,记忆力的东西要有待提高,说道半天,新年的目标

    新年的目标:平常的工作是少不掉了的,学习业务,还是往其它方面充电,还是转型做技术管理方面的DD,DBA很多都是做PLSQL开发的人转型的,可以考虑,目标仍旧是数据库,还是JAVA的重点突破,学习要积累,不是会新的框架就好,设计模式,框架,WEB services,

EJB好多的DD,静心去做一件事,安心去做一件事:)

   几个月都没来更新这里了,书很少的看,网却拼命的上,天天都迷失在网络里面,看着mop嘻嘻哈哈,看着XICI滥情横生,自己的时间在滴嗒滴嗒中流逝,往事以罔然,年年催人老,岁月不饶人,脑中的语言很匮乏,聊天也只是短语的拼凑,连在一起好像洋洋洒洒的破铜烂铁,哈哈

2004年11月10日

 自行车公司门口被偷  -_- 郁闷。。。。















 

news.longhoo.net  2004-7-26 10:44:33  推荐本稿  短信订阅








  我的自行车又丢了。加这个“又”字,是有着以五部自行车为铺垫的切肤之痛的。因此,当第六部车在我进餐厅消费一个小时之后,发现“又”不翼而飞了,我有着异乎寻常的淡定和冷静。拿出手机,提着小偷扔下的粗如儿子手腕的大锁,一边无谓地报警一边向家的方向走去。破财免灾吧。


  我一直对无处不在的偷车贼感到不可理喻。只要你将一辆新车,特别是如捷安特一类的“高档车”或电动自行车车,敢胆放在无人值守处,保险它会在某个时间段内不见踪影。我的一辆车,是停在某军校的大门口哨兵的眼皮底下,心想这该没有问题吧,等到我回来时,我不禁愕然:居然又不见了。小偷的这种专业技巧和心理素质,恐怕没有铁道游击队的水准玩不来。只是,对于这样肉眼看不见却又密如蝗虫一般的偷车贼,警察不是轻易就能钓出来吗?难道真的老鼠天生能够闻到猫的味道吗?


  在我所生活的这个城市,如果哪一家没有丢过自行车,恐怕是要悬赏才能找得出来的。据说,自行车失窃不独这个城市才有,在全国也具有相当的普遍性,现在所使用的插卡式的牌照好像就是南方某个城市先使用的。因此,当这种在“理论上”非常“有效”的牌照发明出来之后,各城市引进速度之快令人感慨。我生活的南京自然不会落后:别管你的旧牌照什么时候买的,统统在限定的期限内,换。只要自行车能够保得住,换牌照就换吧,钱也不多也就区区几块,尽管就是两个“神奇”的塑料小牌牌,认了。可当我接二连三地丢车之后,没有人因为牌照不管用而向我进行起码的解释,自认倒霉吧。可你想想,一个几百万人口的城市都换牌照的话,那也是一笔可观的数字,如果用这笔钱养一个抓贼别动队,不信小偷还敢这样猖狂。已经有城市认为没必要而将牌照取消,还受到其它城市的非议和抵制:没牌照怎么防盗?可他们也不想想一年找回来几辆賍车呢?


  自行车失窃风还催生了一个“病态”的行业:锁业。这几年为了防范自行车失窃,各种各样的防盗锁层出不穷,恰如各种各样的保健品和“特效药”。你如果肾虚,有大力丸、鳖精什么的。你如果个子不高,有增高药。如果你不聪明,有“聪明素”,你如果酒量小,有解酒药。如果有人觉得不够丰挺,就有人说“没什么大不了”的,吃了**,从此男人就不能“一手掌握了”。凡正什么病治不了,什么病难治,有什么难言之瘾,保证有各种各样的“对症”药来。病急乱求医,可现在的有些“医生”简直就是披着白大褂的狼,等着你喂呢,那才叫被人骗了还帮人数钱,数完钱还被人骂傻B。因此,当各式标榜防盗的“概念锁”出来之后,价格自然不菲,当代心一横决然地武装到牙齿后,可最后也难保被小偷一网打尽,真是赔了夫人还连带丫环的,跟谁说去?最近听说有人发明了全球定位防盗装置,可谓高科技到顶了,可到底是不是如纳米、核酸一类的东东,听着好听,管不管用就天知道了。


  自行车被偷,于是便想到保险,可是人家保险公司说这个险种包陪不赚,不做了,你有什么脾气?去年南京市所有的保险公司突然推出出租车和私家车500万以下绝对免赔的招,理由是太容易出事了,不划算。当然最后因为车主意见太大被保监会叫停了,但是几家公司坐下来商量一下就能够推出这个断然措施,想想也真是够意思。平时吃香的喝辣的以为都是天上掉下来的呀,觉得不划算就不做了,大爷嘴脸一目了然。自行车投保无门你奈谁何?把这笔保险费交给“偷车协会”恐怕所有的问题都解决了。


  小小自车行,却是和市民百姓的生活息息相关。因此,当自行车的安全题成为一个哥德巴赫难题的时候,百姓的安全感可能多少要打点折扣。因此在我看来,如果哪个城市不能解决这个问题,就没有理由申报什么最佳“人居城市”。我们有个爱国卫生大检查,某个城市要想申报全国卫生城市,在迎接检查的时候,必然洒扫庭除不留死角,一片风巻残云窗明几净。如果能有个“自行车防盗全委会”,不解决偷车问题就对领导一票否决,或许这个难题很快就能解决了,因为“办法总比困难多”。


  不过,这个设想只是乌托邦,没有影子的事。哎,车丢了跟谁急呢?和看不见却又如影随形的小偷先生生闷气吗?

2004年10月27日

lsattr -El mem0
lsattr -El sys0 -a realmem

 

BootInfo -r

2004年10月24日
教你8个治病小方
(10/21 17:47)
  热花生油治蜂蜇不慎被蜂蜇后,可将花生油放入锅内烧热,放凉后以不烫手为宜,直接涂抹到患处,很快就可达到止痛消肿的效果。如果效果不明显,可以继续用一段时间。

  治老年眩晕方白果 银杏 30克,有呕吐者加干姜6克,一起研成细末后分4份,每日早晚各服1份,温开水送服。一般服4次—8次即可见效。

  大枣黄芪粥治低血压取大枣10枚,黄芪16克,糯米50克。先煮黄芪去渣,用汤汁与大枣、糯米同煮成粥,每晚一次,连用两个月。

  胡萝卜汤治腹泻将胡萝卜制成泥,再加水成汤,500克胡萝卜制成汤1000毫升,所含热能较低,又呈碱性,含有果胶,有使大便成形和吸附细菌及毒素的作用,腹泻初期以胡萝卜汤治疗,效果不错。

  治风湿筋骨痛方柳枝、槐枝、桑枝各七条 切成寸段 、艾杆适量。将备好的药剂加水煎成汤,每晚睡觉前,将汤药倒入盆内,先熏蒸,后泡洗 洗后注意不着凉 ,一连熏蒸泡洗七天,即有良效。

  佛手治胃痛佛手20克,先加水,煎汤后去渣取汁,再与粳米100克煮粥,粥成时加红糖适量 也可不加 。每日分2次食用,治慢性胃炎所致胀气疼痛等症。

  树皮治痔疮椿树皮、大枣各50克,加水适量,煎汤拌白糖冲服,睡前一次服下。

  红糖白酒治痛经将白酒约50克装入一容器中,加热至沸腾时放入20克红糖,边煮边搅拌,待白酒再次沸腾后即可。痛经时趁热服下,一日一次,连服3天即有明显效果。寿柯

ATA是广为使用的IDE和EIDE设备的相关标准。ATA是AT Attachment的缩写,意思是AT计算机上的附加设备(还记得IBM PC/AT吗?)。ATA可以使用户方便地在PC机上连接硬盘。

ATA标准从1994年至今共经历了7代标准,现在简述如下:

ATA-1(1994):是最早的IDE标准。

ATA-2(1996):是EIDE的标准,支持PIO3,4和DMA 1,2传输方式,最大数据传输速度为16MB/s。

ATA-3(1997):引入了SMART和安全特性,没有制定新的传输标准。

ATA-4(1998):著名的“UDMA33”标准。引入了新的ATA命令和协议,最大数据传输速度为33MB/s。

ATA-5(2000):增加了一些新的ATA命令。最大数据传输速度为66MB/s。

ATA-6(2000):UDMA100。

ATA-7(2002):UDMA133。

UDMA (Ultra Direct Memory Access)-极端直接内存访问,硬盘技术的一种,它可以让硬盘直接与内存沟通而无须CPU过多的参与。

※也就是说ATA是一个标准,而UDMA是一种技术

2004年10月23日

Enhancements in PL/SQL Developer 6.0.3
======================================
* Fixed connection delays and ORA-12154 errors when using Oracle Names 9.0
  or 9.2
* Using the Large Data Editor could lead to instability
* Some applications could not be started when PL/SQL Developer is running
* Memory overhead in object list reduced (Export User Objects,
  Compare User Objects, and so on)
* Code Contents parse error fixed when using the words “left” or “right” in
  a condition
* SQL Window “Next SQL” button did no longer clear the editor after the
  last SQL
* Text Importer replaced only 1 occurrence of # in an SQL function
* The View function for program units would bring an already opened,
  editable window to the front and make it read-only
* Creating/Editing users with special characters would fail
* Fixed error while downloading updates & News details
* Preferences are now synchronized between different instances on the same PC
* Fixed problem with copy/paste multi line data in grid
* PL/SQL Beautifier would raise error for table.column expression with a
  space between table and .column
* XML editor would replace header with <?xml version=”1.0″?> after making a
  modification
* Large Data Editor could add 0-byte to BLOB data
* Code Assistant did not come up for lower..upper expression without spaces
* Data Generator did not generate master/detail/detail data correct (only
  test was correct)
* Text Importer would not always handle quotes correct
* Large Data Editor could enable word-wrap in other editors
* Preference for maximum size of recall statement would not always be saved
* Improved the resizing behavior of the Plan Window
* Safe Compilation preference did not work correctly if owner preceded the
  object name (create or replace package scott.employee …)
* Clear function for SQL, Command, Test and Report Window did not also clear
  the “Saved” indicator
* Left-clicking on a Code Contents item will now always navigate to the
  start of that item in the editor
* Preferences will be stored in Application Data Path if the default
  directory is read-only
* Not all font and color preferences were applied immediately in preference
  dialog
* Date format preference could be changed externally
* Changing the Session Mode preference could cause errors if PL/SQL Developer
  is not restarted
* Find dialog would be displayed at wrong position for multi monitor systems

Enhancements in PL/SQL Developer 6.0.2
======================================
* Partitioned tables were not detected correctly on Oracle8i
* Partitions tab page was visible when viewing a non-partitioned table
* Code Assistant did not describe materialized views
* Code Assistant did not describe record types declared in another
  package
* Code Assistant could remove spaces typed after a comma
* File -> Close All could lead to invisible confirmation dialog
* Font Bold & Italic preferences did not work for browser and grid
* Key Configuration preference would not handle all keys properly
* Compare User Objects could incorrectly change column properties when
  column order for source and target table is different
* Large Data Editor had problems with XML data with lines > 65000
  characters
* Preference “Use DBA views if available” could cause ORA-00942 after
  switching from a DBA user to a normal user
* Compilation errors were not displayed when “create or replace” and
  object name are on different lines
* Could no longer connect with just / for externally identified users
* Deleting a column selection in the editor could also delete adjacent
  highlighted text
* Dropping a java source from the Object Browser would cause ORA-29501
* PL/SQL Beautifier did not format all ANSI join types correctly
* On Oracle7 Explain Plan and Test function did not replace PL/SQL
  variables by bind variables
* Text Importer would only import 99 records when pasting from clipboard
* Data Generator did not handle a master with 2 details correct
* UxTheme related errors fixed
* Browser Folders would not save connection property
* TSV Exporter would skip first field if it was empty
* Large Data Editor now closes when pressing Escape
* Compiling a body with a missing or invalid specification with “Safe
  compilation” preference would give compilation errors withtemporary
  object name
* Hyperlink navigation did not work correctly for compound custom
  keywords
* Hyperlink navigation did not immediately update Code Contents
* Generating a Test Script for a non-packaged program unit with
  parameters with user-defined types would not create local variables
  with correct types
* Dropping an object in the Compile Invalid Objects tool would not
  update the compilation error list
* Fixed a problem starting specific applications when PL/SQL Developer
  is running
* Data Generator number precision displayed for fields with a scale of
  0 would not be correct
* Explain Plan and Test function now also replaces constants by bind
  variables in select statement
* Pressing a cell button in an empty Single Record grid in a SQL Window
  would cause an error message
* Export to Excel could display cells as #### due to column width
* SQL Window now preserves result message after navigating in result
  grid when “Show dictionary info” preference is set
* Tables folder in Tablespace folder in Object Browser did not list
  partitioned tables
* Compare User Objects memory usage optimized
* Compare User Objects list included tables from the Oracle10g
  recycle bin
* Fixed splitter problem and alignment of editor in XML Editor
* Oracle Logon preference “Fixed Users” is now stored encrypted
* Excel interface would not handle Japanese characters properly
* Fixed “Invalid pointer operation” errors

Enhancements in PL/SQL Developer 6.0.1
======================================
* Support for Oracle10g EZCONNECT added
* Performance of “Copy to Excel” function improved
* DDL for synonyms now uses “create or replace” syntax for Oracle 9.2
  and later
* Pressing an “Auto refresh timer” button now immediately executes
  the Window
* Shortcuts Plug-In enabled again
* Scrollbar in Explain Plan function now scrolls all contents
* Compare User Objects would try to rename columns when column order is
  different
* Code Hierarchy did not have Explain Plan and Test popup menu items
  for select statements
* Executing a Report would copy the SQL to the clipboard
* Report would not ask for password when “locked for others”
* PL/SQL Beautifier preview did not show effect of
  “Wrap zero level AND/OR” option for Control Structures
* Proxy settings added for Updates & News
* Renaming an object in the SQL Window could lead to incorrect
  “Object has been deleted” message
* Beautifier DML Rule “AND/OR under where clause” was always enabled
* Code Contents could sometimes be blank or out-of-date
* Fixed various access violations
* Explain Plan function did not replace record field PL/SQL variables
  in SQL text by bind variables on Oracle 9.0 and earlier
* Safe compilation could fail if the package body contains references
  to objects with the same name in another package
* Disabling the “Include Storage” option during Export User Objects
  would still include partition storage information
* Text Importer now correctly autodetects text with pipelines
* Large Data Editor now supports undo, redo, find and select all
* Text Importer data preview would not highlight multi line fields
  correctly

New features in PL/SQL Developer 6.0.0
======================================
* Program Window enhancements:
  > Code Hierarchy pane added
  > Code Contents extended with code structure and hot tracking
  > Compiles in background if session mode is “multi session”
  > Highlight if, loop, case, and blocks
  > Safe compilation preference added
  > Save before compilation preference added
  > Oracle10g PL/SQL Compiler warnings supported
* Support for partitioned tables added
* Support for materialized views added
* Support for compressed indexes added
* Support for reversed indexes added
* Support for queues added
* Support for queue tables added
* Support for NLS Length Semantics added
* Support for LOB and nested table column storage added
* Support for column renaming added (Oracle9.2 and later)
* Editor enhancements:
  > Hyperlink navigation now works in all SQL and PL/SQL editors
  > Highlight parentheses
  > Table and View drag & drop into editor can now launch the Query
    Builder
  > Find function can now wrap to start/end of document if the search
    text is not found
  > Append function added to append text to the current clipboard text
  > Sort function added to sort a selection
  > Guideline function added
  > SpecialCopy syntax extended with #define compress
  > Multi-line strings properly highlighted in SQL and PL/SQL editors
  > Oracle10g keywords highlighted in SQL and PL/SQL editors
* Code Assistant enhancements:
  > Support added for packaged record types and rowsubtypes
  > Support added for cursor variables
  > Support added for select statements on inline views
  > Home / End key will go to top / bottom of the list
  > Wraps to top / bottom from last / first line
* Authorization added
* To-Do Items added
* Data Generator tool added
* New preference dialog with preference sets
* External Tools generalized as External/Session Tools
* New Large Data Editor for CLOB, BLOB, LONG, LONG RAW and VARCHAR2
* Find Database Objects improvements:
  > Search for object name, owner, status, creation date range, and
    modification date range
  > Criteria can be saved and loaded
* Export User Objects improvements:
  > Include Storage option added
  > View button added to view the resulting script
* Compare User Objects improvements:
  > Option added to include storage in generated DDL for new objects
  > Table column renaming supported (Oracle9.2 and later)
* Export Tables improvements:
  > Include Storage option added for SQL and PL/SQL Developer format
  > Include Privileges option added for SQL and PL/SQL Developer format
* HTML Manual enhancements:
  > Multi-tabbed search results
  > Roll up / roll down buttons
  > Search text history
* Object Browser enhancements:
  > User defined folders added
  > Folder hierarchy
  > Folder color
  > Connection specific folders
  > Add/remove objects through right-click in Object Browser
  > Super/Subtype displayed for type objects
  > Column renaming added (Oracle9.2 and later)
  > Dropping a user that owns one or more objects is now possible after
    an explicit confirmation
  > Object popup menu now shows the debug information status
  > Recycle bin added for dropped tables on Oracle10g with Purge and
    Restore functions
* AutoRefresh option added for Session Window, SQL Window,
  Report Window, Test Window, and Command Window
* Added “Change Window to..” to popup menu for SQL Window, Program
  Window, Test Window, Command Window and Report Window
* SQL Window enhancements:
  > New number layout preference: “Formatted with thousand separator”
  > New “Null Value Color” preference
  > Object Browser is immediately updated after creating / dropping /
    altering and object
  > Compiling a view will save & restore instead of triggers
  > Query can now be saved as “Standard Query” for a Table, View or
    Materialized View
  > Direct XMLTYPE query support on Oracle Net 9.2 and later
  > Saving a view source now provides the view name as default filename
  > Clear, Clear Result and Select All added to popup menu of Single
    Record view
  > ReadOnly substitution variables added
  > Hidden substitution variables added
* Test Window enhancements:
  > New debugger preference “Never add debug info for objects”
  > Current debug line is now centered
  > Watches are saved and loaded with the Test Script
* Plan Window enhancements:
  > Copy function prefixes each item with its name
  > Oracle10g columns added
* Command Window enhancements:
  > Set prompt command added
  > Exit command can now exit PL/SQL Developer
* New PL/SQL Beautifier options:
  > Parameter declaration: List at left margin
  > DML: Split at zero-level AND/OR
  > DML: AND/OR after expression
  > DML: AND/OR under where clause
  > Control structures: Split at zero-level AND/OR
  > Control structures: AND/OR after expression
* PL/SQL Beautifier enhancements:
  > Support for Merge statement added
  > Support for Join syntax added
  > Constants are aligned when “Align declarations” is enabled
* Recall Statement enhancements:
  > Buffer size and location Preferences
  > Export to CSV file
  > Find function
* Oracle10g Instant Client support added
* All combo boxes now have AutoComplete functionality
* Taskbar icon now only displays connection information
* File > E-mail item added
* Keyboard search in property windows
* Popup menu added to the desktop with some primary functions
* Rollup button added to Window List and Template Window
* Shift-click in Window List closes the selected window
* New command-line parameters:
  > commandfile: execute a command file
  > viewobject: view database object
  > editobject: edit database object
  > noplugins: don’t load Plug-Ins

 

2004年10月22日

其实加引号是PL/SQL的规范,数据库会严格按照“”中的名称建表,如果没有“”,会按照ORACLE默认的设置建表(DBA STUDIO里面),默认是全部大写,这样,在ORACLE数据库里的字段就如“Column_1”,如果你把引号去掉,ORACLE自动默认为全部大写,即“COLUMN_1”,所以不必担心,这段SQL在PL/SQL中执行的时候是没有任何问题的,如果不加引号,在PL/SQL会自动识别为大写。
 这个问题是生成脚本格式的问题,因此,我们可以尝试在DBMS配置文件中修改相应的格式设置来解决这个问题。

选择Database->Edit current database,选择Sql->Format,有一项CaseSensitivityUsingQuote,它的comment为“Determines if the case sensitivity for identifiers is managed using double quotes”,表示是否适用双引号来规定标识符的大小写,可以看到右边的values默认值为“YES”,改为“No”,点击【确定】按钮。在弹出的确认对话框中点击【确认】。

2004年10月21日

图标clsid=windowsmedia 文件名.{00020c01-0000-0000-c000-000000000046}
无关联文件 文件名.{00021401-0000-0000-c000-000000000046}
BMP文件 文件名.{d3e34b21-9d75-101a-8c3d-00aa001a1652} HTML文件 文件名.{25336920-03f9-11cf-8fd0-00aa00686f13}
媒体剪辑文件 文件名.{00022601-0000-0000-c000-000000000046}
打印机 文件名.{2227a280-3aea-1069-A2de-08002b30309d}
控制面板 文件名.{21ec2020-3aea-1069-A2dd-08002b30309d}
网上邻居 文件名.{208d2c60-3aea-1069-A2d7-08002b30309d}
拨号网络 文件名.{992cffa0-F557-101a-88ec-00dd010ccc48}
计划任务 文件名.{D6277990-4c6a-11cf-8d87-00aa0060f5bf}
回收站 文件名.{645ff040-5081-101b-9f08-00aa002f954e}
Web文件夹 文件名.{Bdeadf00-C265-11d0-Bced-00a0c90ab50f}
历史记录 文件名.{Ff393560-C2a7-11cf-Bff4-444553540000}
收藏夹 文件名.{1a9ba3a0-143a-11cf-8350-444553540000} 预订文件夹 文件名.{F5175861-2688-11d0-9c5e-00aa00a45957}
ActiveX高速缓存文件夹
文件名.{88c6c381-2e85-11d0-94de-444553540000}

desktop.ini中
[.ShellClassInfo]下

将文件夹改为含有图标的任意文件
iconfile=c:\windows\system\shell32.dll
iconindex=2(任意数字,66为.dll文件)

隐藏所有文件 uiclsid={7bd29e00-76c1-11cf-9dd0-00a0c9034933}
再将文件夹改为只读

补充:

excel.{00020810-0000-0000-C000-000000000046}

word.{00020900-0000-0000-C000-000000000046}

media.{00022603-0000-0000-C000-000000000046}

CAB.{0CD7A5C0-9F37-11CE-AE65-08002B2E1262}

计划任务.{148BD520-A2AB-11CE-B11F-00AA00530503}

搜索-计算机{1f4de370-d627-11d1-ba4f-00a0c91eedba}

网上邻居.{208D2C60-3AEA-1069-A2D7-08002B30309D}

我的电脑.{20D04FE0-3AEA-1069-A2D8-08002B30309D}

控制面板.{21EC2020-3AEA-1069-A2DD-08002B30309D}

打印机.{2227A280-3AEA-1069-A2DE-08002B30309D}

html.{25336920-03f9-11cf-8fd0-00aa00686f13}

mht.{3050F3D9-98B5-11CF-BB82-00AA00BDCE0B}

mshta.{3050f4d8-98B5-11CF-BB82-00AA00BDCE0B}

我的文档.{450D8FBA-AD25-11D0-98A8-0800361B1103}

XML.{48123bc4-99d9-11d1-a6b3-00c04fd91555}

回收站(满).{5ef4af3a-f726-11d0-b8a2-00c04fc309a4}

回收站.{645FF040-5081-101B-9F08-00AA002F954E}

ftp_folder.{63da6ec0-2e98-11cf-8d82-444553540000}

网络和拨号连接.{7007ACC7-3202-11D1-AAD2-00805FC1270E}

写字板文档.{73FDDC80-AEA9-101A-98A7-00AA00374959}

Temporary Offline Files Cleaner.{750fdf0f-2a26-11d1-a3ea-080036587f03}

用户和密码.{7A9D77BD-5403-11d2-8785-2E0420524153}

Internet 临时文件.{7BD29E00-76C1-11CF-9DD0-00A0C9034933}

已下载的程序文件的清除程序.{8369AB20-56C9-11D0-94E8-00AA0059CE02}

公文包.{85BBD920-42A0-1069-A2E4-08002B30309D}

ActiveX 高速缓存文件夹.{88C6C381-2E85-11D0-94DE-444553540000}

mail.{9E56BE60-C50F-11CF-9A2C-00A0C90A90CE}

历史记录.{FF393560-C2A7-11CF-BFF4-444553540000}

目录.{fe1290f0-cfbd-11cf-a330-00aa00c16e65}

Internet Explorer.{FBF23B42-E3F0-101B-8488-00AA003E56F8}

Snapshot File.{FACB5ED2-7F99-11D0-ADE2-00A0C90DC8D9}

预订文件夹.{F5175861-2688-11d0-9C5E-00AA00A45957}

MyDocs Drop Target.{ECF03A32-103D-11d2-854D-006008059367}

Policy Package.{ecabaebd-7f19-11d2-978E-0000f8757e2a}

搜索结果.{e17d4fc0-5564-11d1-83f2-00a0c90dc849}

添加网上邻居.{D4480A50-BA28-11d1-8E75-00C04FA31A86}

Paint.{D3E34B21-9D75-101A-8C3D-00AA001A1652}

管理工具.{D20EA4E1-3957-11d2-A40B-0C5020524153}

字体.{D20EA4E1-3957-11d2-A40B-0C5020524152}

Web Folders.{BDEADF00-C265-11d0-BCED-00A0C90AB50F}

DocFind Command.{B005E690-678D-11d1-B758-00A0C90564FE}

脱机文件夹.{AFDB1F70-2A4C-11d2-9039-00C04F8EEB3E}

打印机.{2227A280-3AEA-1069-A2DE-08002B30309D}