Hi Victoria,
>> echoplusAT> *I can't run a 16 bit program the file setup.exe is
damage. Try to
>> echoplusAT> reinstall* (!?!?!?)
>>
>> echoplusAT> I've never seen this error ... anyone can help me in order to
solve this
>> echoplusAT> prob?
>
> May be you have a corrupted installation packa
Hi,
I've a problem in order to install Mysql server 3.23.47 on NT 4 box
when I try to run the file setup.exe, I get an error msg like this:
*I can't run a 16 bit program the file setup.exe is damage. Try to
reinstall* (!?!?!?)
I've never seen this error ... anyone can help me in order t
I'd rather not fire up Viz C just to build MySQL 3.23.39 NON-DEBUG.
Is there much of a performance difference between DEBUG & NON-DEBUG ?
Also with the Windows binary builds how does one NOT use MySQL MAX, is that a runtime
option or do you need a different build entirely ?
Is there much of
Do the names of indexes have any significance to MySQL internals or is the name merely
a convenience for the DB operator ?
For now I try to match the index names to the column names used but I'm wondering if
the name matters at all.
thanks,
- Sam.
-
Do You
t
detailed info on where and how to change your indexes, syntax
etc.
S A schrieb am Dienstag, 24. Juli 2001, 01:08:31:
> I have standard search functionality on my site where a user can see NN items per
>screen of a much longer list of search results. I've found that MySQL retur
PS : I forgot to add that as a further optimization I cache the list of IDs from the
first query in the 2 part approach& I reuse that list as users advance through it.
-
Do You Yahoo!?
Make international calls for as low as $.04/minute with Yahoo! Messenger
I have standard search functionality on my site where a user can see NN items per
screen of a much longer list of search results. I've found that MySQL returns small
result sets out of a list of hits faster when queried in 2 parts than queried once. Am
I doing something wrong ?
I have a fairl
Is there some way to take the following 2 queries & make them into one ?
I want to be able to in one query get the # of overall rows and do a LIMITed result
set to get the first NN rows.
Is it possible to to do this in one query ? Presumable MySQL has to figure out what
the overall # is to re
Hello Benjamin.
I forgot to mention that there could be items from multiple posters in each folder so
the query you sent worked on those cases just like the old query. However looking at
your nicely formatted example made me realize what I needed to do.
By concatenating the poster_name in fron
When I do the following query I get seemingly arbitrary results for most of the
columns.
SELECT f.folder_name, COUNT(g.post_id), n.poster_name, p.file_name
FROM users u, folders f, users_to_files g, posts p, posters n
WHERE f.user_id = u.user_id AND u.user_id = 1 AND g.folder_id=f.folder_id
How much if at all slower should a BETA-LOG build of MySQL be than one that doesn't
log ?
We have on BSD 3.23.26-beta-log which is MUCH slower than 3.23.32 on Linux.
- Sam.
-
Do You Yahoo!?
Yahoo! Auctions $2 Million Sweepstakes - Got something to sell?
We need FULLTEXT searching on 2 character length words.
If we recompile MySQL to do FULLTEXT indexes on 2 char length words is it going to
work OK or will such short words undo the advantages of FULLTEXT ?
Also will the data files be useable by other standard MySQL binary builds or will we
h
I have 3 variations of a fairly simple aggregate query done on a medium sized table of
2 million rows.
Doing an ORDER BY makes the query go from less than a second to often taking 30
seconds.
Is this a bug ?
We're running on Free BSD 4.2 with Dual PIII 700mhz & 1GB RAM.
- Sam.
A. The o
I have a query that works fine until my date range goes one day over some kind of data
size threshold & then it mysteriously goes from 0.34 seconds to 30 seconds. My DBA
pointed out that my SQL syntax was loose & that Oracle would've flagged it as an error.
I did a GROUP BY based on an numeric
My DBA & I can't find a .cnf file to start customizing our Free BSD MySQL setup.
Could someone please post one up to start from ?
thanks,
- Sam.
-
Do You Yahoo!?
Yahoo! Auctions $2 Million Sweepstakes - Got something to sell?
I have some complex multi-step queries that take over 60 seconds on Free BSD 4.2. They
seemed much faster on Linux.
Should FreeBSD 4.2 be just as fast as Linux on comparable hardware ?
When I simulate concurrent users the query times go up oddly.
1 user 66 seconds
2 users 157 seconds
3 user
Can you use UTF-8 with MySQL ?
Is UTF-8 supported now, just compatible or plain won't work ?
I've read conflicting discussion on this topic.
thanks,
- Sam.
-
Do You Yahoo!?
Yahoo! Mail Personal Address - Get email at your own domain with Yahoo! Mail.
I find the fuzzy matches from MATCH and AGAINST to be too unpredictable.
Does using a FULL TEXT index also speed up simple LIKE searches ?
-
Do You Yahoo!?
Yahoo! Mail Personal Address - Get email at your own domain with Yahoo! Mail.
Turns out the Linux table_cache was 457 to Solaris 64.
Could that have been all the speed difference ?
- Sam.
Solaris
mysql> show variables;
+-+-
--+
| Variable_name | Value
|
+-
We've got our MySQL DB on comparable Linux & Solaris hardware but Linux beats Solaris
by about 6 times.
Anyone have Solaris tuning tips ?
-
Do You Yahoo!?
Yahoo! Mail Personal Address - Get email at your own domain with Yahoo! Mail.
Can MySQL help with generation of ALPHANUMERIC unique identifiers ?
ID's that use the full ASCII char range create more possible combinations with the
same # of characters which can be important when squeezing into cookie storage space.
4 character numeric ID has 10,000 combinations.
4 charac
How do I do a single query to group a table by one column & then get the row of say
the highest of each group ?
Data is like so :
Album, City, Plays, Date
Britney, NYC, 103, 1-10
Britney, SF, 101, 1-10
Eminem, NYC, 100, 1-11
Eminem, SF, 102, 1-11
So I want to GROUP BY Album, then get
How do I do a single query to group a table by one column & then get the row of say
the highest of each group ?
Data is like so :
Album, City, Plays, Date
Britney, NYC, 103, 1-10
Britney, SF, 101, 1-10
Eminem, NYC, 100, 1-11
Eminem, SF, 102, 1-11
So I want to GROUP BY Album, then get the
I will respond in more detail later but just quickly now :
> Hm. Sorry, I don't understand your argument. By the for loop, your
>query does the same, doesn't it (selecting the rank of one album in
>each metro is the same as selecting for all metros the rank of one album)?
The subtle differ
> Huh? You got it 7 times faster by only extracting that query from the
> for loop and changing it to a group by? Or did you do anything else?
Sorry there were some other unnecessary queries that were mistakenly in the
measurement..
Here is the missing EXPLAIN.
mysql> EXPLAIN SELECT same_
This is the latest approach we're using now. It turns out that the query to get Total
Plays for an Album per City can be taken out of the City for loop & done only once as
a GROUP BY.
The worst case time for MySQL is now 16 seconds compared to Oracle's 8 seconds. On
most cases MySQL is now f
> Could you please post the result of "EXPLAIN" for both queries and "SHOW INDEX FROM
>album_by_city".
I haven't used EXPLAIN before.
There are indexes on each column.
> And, for how much cities does this gets executed?
There are about 100 cities or regions that are used for ranking.
> Ad
I'm moving over a music play database from Oracle to MySQL. So far MySQL is faster for
all reports except this one where Oracle takes 8 seconds & MySQL will take 108 with my
current method.
I have a big table of times that an album is played like so :
album_id INT,
total_plays INT,
date DAT
I want to SELECT on one term & have the SELECT return as quickly as possible on the
first TRUE instance.
Ideally I'd like it to after the first matching case stop looking for other matches (
for speed ) and return 1. If there's no matches at all I want it to return 0.
This works OK but return
29 matches
Mail list logo
