Re: Indexing TNSNAMES?

Actually, I can't imagine that reading a few-hundred line text file would be a big performance hit anyway, particularly if it tends to stay in at least the OS's buffer. Ever hear of the reference to indexing it?

As far as the bequeth, I've think part of my problem was thinking beq and IPC were the same thing. I just found another article on Metalink about setting up IPC connections, and so am getting ready to retry my tests.

Thanks.

• Ed

On Tue, 13 Nov 2001 14:58:28 +0000, Connor McDonald <connor_mcdonald_at_yahoo.com> wrote:

>Ed Stevens wrote:
>>
>> A few days ago I posted a question here (see "Puzzling Results of
>> Using Bequeaqth") that didn't draw any responses, so I posted it to
>> MetaLink. (I generally find this NG provides faster response and
>> better advice than MetaLink). Their response was to point me to a
>> white paper -- CR 212951 --that didn't even approach the subject of my
>> question, but did have a couple of statements I found interesting:
>>
>> "When a client application initiates a transaction to retrieve data
>> from a server, the entire Tnsnames.ora file is read."
>>
>> "Instead of reading the entire file and scanning for the relevant
>> entry, it is better to implement an indexing method."
>>
>> First, while I had never been told otherwise, I always assumed
>> tnsnames was read until the relevant entry was found, rather than
>> reading the entire file into a buffer then scanning. Right or wrong?
>>
>> Second, and much more interesting . . . I've never heard of indexing
>> tnsnames. What is this all about?
>>
>> BTW, I'm still searching for an answer to my question in "puzzling
>> performance results with bequeath connection" if anyone wants to take
>> a crack at it.
>>
>> Thanks.
>>
>> - Ed Stevens
>> --
>> Ed Stevens
>> (Opinions expressed do not necessarily represent those of my employer.)
>
>Dunno about your bequeath problem (sorry), but I remeber having a
>tnsnames.ora problem once when we had erroneously duplicated some
>entries. If memory serves, it took the last one (a test I've just
>re-done on 9i with the same result).
>
>So presumably, it reads the whole lot
>
>hth
>connor
>--
>==============================
>Connor McDonald
>
>http://www.oracledba.co.uk
>
>"Some days you're the pigeon, some days you're the statue..."

--
Ed Stevens
(Opinions expressed do not necessarily represent those of my employer.)