Re: Locking

From: HansF <News.Hans_at_telus.net>
Date: Fri, 21 Oct 2005 20:50:35 GMT
Message-Id: <pan.2005.10.21.20.49.15.119359_at_telus.net>


On Fri, 21 Oct 2005 22:21:22 +0800, Pat interested us by writing:

> 
> "HansF" <News.Hans_at_telus.net> wrote in message
> news:pan.2005.10.20.18.15.50.70906_at_telus.net...

>> On Thu, 20 Oct 2005 23:50:58 +0800, Pat interested us by writing:
>>
>> >
>> > in oracle form when user locks a row. some other process which needs
>> > the same row will need to wait until the first user commit/rollback.
>>
>> Depends ENTIRELY on what locks are in place and what lock state the other
>> process needs.
>>
>> For example, even if a process updates a row, other processes will
>> normally be able to read the previous copy (or before begin of
>> transaction copy) of the row without any hesitation. Writers do not block
>> readers and readers do not block writers - under 'normal' circumstances
>> and within 'normal' limits.
>>
>>
> Thanks. instead both processes need to update the same row.

In that case, the second to the row will wait on the first to commit or rollback.

Sounds like a design issue or flaw if this is a concern.

-- 
Hans Forbrich                           
Canada-wide Oracle training and consulting
mailto: Fuzzy.GreyBeard_at_gmail.com   
*** I no longer assist with top-posted newsgroup queries ***
Received on Fri Oct 21 2005 - 22:50:35 CEST

Original text of this message