Re: Possible to delay index writes until server is less busy?
Write to a memory table first then do a hotcopy on a scheduled basis. - Original Message - From: Mathias [EMAIL PROTECTED] To: mysql@lists.mysql.com Sent: Thursday, June 30, 2005 9:10 AM Subject: Possible to delay index writes until server is less busy? We've been benchmarking a database that in real-life will have a huge write load (max peak load 1 inserts/second) to the same table (MyISAM). We will need about 4 indexes for that table. However, from our benchmark tests, it is clear that writing indexes takes too many resources and impedes the speed of inserting new records. To overcome this, we are thinking of: 1 - using several smaller tables (instead of one big one) by creating and writing to a new table every x hours, 2 - wait with writing the indexes until a new table has been created where the next inserts will be (i.e, not write indexes until the table has been closed) The biggest problem now is if the indexes are created when the server is very busy. If there was a way of telling MySQL to delay creating the indexes when it is busy, then a big obstacle would be out of the way. Is this possible? We could not find anything in the MySQL documentation concerning this. Any suggestions would be greatly appreciated. Kind regards, Mathias -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: What is best open-source platform alternative to overcome innodb 2Gb memory limit on Linux? FreeBSD?
I agree The New AMD's (Can't say just opertron) but the 246 and 248 CPU's are moving data between cpu and Ram at 6 gig per second versus 2gig for the Xeon's peak right now. The New opterons communicate with ram better than any other CPU on the market and with the right MySql setup that is a huge benefit. My News site platform is going to be moving from Xeon to AMD for that very reason. Our software is written to avoid harddrive calls at all cost to keep our page load super fast. I would add to his suggestion a RAID 0-1 setup that would double your drive output speed. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.prnewsnow.com Free content for your website 469 228 2183 -Original Message- From: David Griffiths [mailto:[EMAIL PROTECTED] Sent: Wednesday, June 15, 2005 11:17 AM To: Brady Brown Cc: mysql@lists.mysql.com Subject: Re: What is best open-source platform alternative to overcome innodb 2Gb memory limit on Linux? FreeBSD? Why not go AMD-64? Dual Opteron, with 8/16/32 gig of RAM? Get a 3ware SATA drive, and run Gentoo for AMD-64. You can increase your innodb buffer pool to use almost all that space. You can make your buffer pool as large as the physical RAM in your machine can support. No 2.5 gig per process, 4-gig limit on addressable memory (without the address-extensions). Your hardware is holding you back more than your operating system. David Brady Brown wrote: Hi, I am currently running a large database (around 20Gb) on a 32bit x86 Linux platform. Many of my larger data-crunching queries are disk-bound due to the limitation described in the innodb configuration documentation: *Warning:* On 32-bit GNU/Linux x86, you must be careful not to set memory usage too high. |glibc| may allow the process heap to grow over thread stacks, which crashes your server. It is a risk if the value of the following expression is close to or exceeds 2GB: Being a responsible citizen, I have my innodb_buffer_pool_size set below 2Gb. But the time has come to scale the application, so I need an alternative solution that will allow me to set innodb_buffer_pool_size as high as my heart desires (or at least well beyond 2Gb). Do any of you have battle-tested recommendations? How about FreeBSD? From what I can gather, it is a good idea to build MySQL on FreeBSD linked with the Linux Thread Library. Would doing so re- invoke the 2Gb limit? I look foward to your collective responses. Thanks! Brady -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Dual Opteron, linux kernels, 64 bit, mysql 4.1, InnoDB
Why not Raid 3 and take advantage of disk write and read performance. Raid 3 isn't commonly used because it has CPU overhead. But at the same time Apache causes CPU overhead waiting for the data from the drives. I am buying this exact same server With 32 Gig of ram. Frankly the slowest thing in my current Raid 5 server is still waiting for the disks to read. That is what prompted me to think bigger processors, more ram and faster motherboard to compensate for using a Raid 3 to overcome the slowest hardware in my server. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Jeremiah Gowdy [mailto:[EMAIL PROTECTED] Sent: Monday, May 09, 2005 6:37 PM To: Dathan Pattishall; Richard Dale; mysql@lists.mysql.com Subject: Re: Dual Opteron, linux kernels, 64 bit, mysql 4.1, InnoDB I use Redhat Advanced Server v4 (2.6 kernel) on my four dual opteron systems. I've had no real performance issues with the I/O scheduler, but that's because I run 8GB of ram with a huge key cache. I recommend taking the box to 8GB of ram, it's worth it. Definately use RAID 10. - Original Message - From: Dathan Pattishall [EMAIL PROTECTED] To: Richard Dale [EMAIL PROTECTED]; mysql@lists.mysql.com Sent: Monday, May 09, 2005 4:15 PM Subject: RE: Dual Opteron, linux kernels, 64 bit, mysql 4.1, InnoDB -Original Message- From: Richard Dale [mailto:[EMAIL PROTECTED] Sent: Sunday, May 08, 2005 9:37 PM To: mysql@lists.mysql.com Subject: Dual Opteron, linux kernels, 64 bit, mysql 4.1, InnoDB A new server is about to arrive here and will have have 8x15K RPM spindles, dual Opteron processors and 4GB of RAM, and will have around 100GB of database (primarily stock market prices) - the SCSI controller will also have battery-backed RAM too. InnoDB will be used exclusively. I've searched the list and seen varying reports of which Linux kernels work best etc. I'd be intersted to know the following: a) Which 64 bit Linux distributions are good for the task? Suse 8.1 2.4.21-215-smp #1 SMP Tue Apr 27 16:05:19 UTC 2004 x86_64 unknown b) Which 64 bit Linux distributions are bad for the task? 2.6 the IO sceduler is still messed up. RedHat AS / Suse 9.x are messed up as well c) Any comments on kernels, particularly with regard to 64 bit support and schedulers? Any problems with the latest kernels (2.6.11 2.6.12-rcX)? d) Any recommendations for RAID volume setup Use RAID-10 split the disks evenly across each channel e) Any MySQL optimisations for multiple spindles, onboard caching, stripe sizes, RAID5 vs RAID10. Don't use RAID5, use Reiser FS if your using SUSE f) Any MySQL reliability settings to take into account the battery-backed RAM on the RAID controller? I'm happy to collate the responses into a summary too. I'm aware of the following discussions which describes a reasonably grunty Dual AMD system with a similar configuration to mine: http://meta.wikimedia.org/wiki/Hardware_ordered_April_2005 http://meta.wikimedia.org/wiki/Hardware_order_May_2004 Best regards, Richard Dale. Norgate Investor Services - Premium quality Stock, Futures and Foreign Exchange Data for markets in Australia, Asia, Canada, Europe, UK USA - www.premiumdata.net -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Can I use LIKE this way?
{
$dead_beat_ads = "Delete from fsearch_temp where fsearchKeyword='$str' and
fsearchTime='$time' and fsearchIp='$fbip' and fsearchHost LIKE '%$dbad%'
"; $dead_beat_result =
$dbi-query($dead_beat_ads);}
What I am doing is
eliminate "Run of site" advertisers from our network LIKE EBAY.com I
would rather give free crawler results than to allow this spam advertising
anymore. So I before I develope our XML output of search results to our
partners I am going to remove them but since their tracking can change All I
want to do is search for them by host Like ebay.com and I will pull
this out in real time thus all the $time requirements. This will mean that
no matter which of our partners ebay buys advertising from they won't be
displayed through our network. Or if someone buys a commission junciton ad
we will pull them out by the host even though the click url may be a qcksrv
ad.
So the question is
can that % be butted up against the variable or should I put a space in
there. While reading the LIKE on Mysql .com it talks about 2nd
position with a space but I didn't know if that meant for the use of
%_$dbad_% or if % $dbad % would have it
looking for second position stuff.
ThanksDonny LairsonPresident29
GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
RE: Can I use LIKE this way?
Yeah, the problem is we are trying to do something that is not done by any
of the other search engines. I am working on a Quick Clean to clean out
Search spammers on the fly. We have Upstream partners (Like Overture) who
like to Fill us with their Run of Site ads when even they don't have
paid advertisers. While this is good for the bottom line we find its
annoying as hell to those who use our search.
Its our hope that folks will use our feed for their search because of our
work to keep it clean. There are tons of sites out there producing feeds.
Because I couldn't do it the way I wanted to I found a new way by cleaning
the URL's from each of our partners before we add them to ram for our output
sorting. By cleaning the urls I was able to delete by using:
{
//print $dbad;
//print br /;
$dead_beat_ads = Delete from fsearch_temp where fsearchHost='$dbad';
$dead_beat_result = $dbi-query($dead_beat_ads);
}
This will give a greater degree of certainty to who we are deleting on the
fly. I have already tested this and we are actually running it live on our
Outdoor network and will implement this on http://www.sharedfeed.com for it
opening day release next week.
By avoiding talking to the hard drive cleaning up content on the fly is a
snap.
Thanks
Donny Lairson
President
29 GunMuse Lane
P.O. box 166
Lakewood NM 88254
http://www.gunmuse.com
469 228 2183
-Original Message-
From: Rhino [mailto:[EMAIL PROTECTED]
Sent: Sunday, May 01, 2005 4:44 PM
To: [EMAIL PROTECTED]; Mysql
Subject: Re: Can I use LIKE this way?
I think you want the wildcard characters, % and _, to be within the
variable, not within the variable name. For example, if you are looking for
all advertisers whose name begins with 'A' followed by zero or more unknown
characters, you would set your variable equal to this pattern [I don't know
PHP so if this isn't how a variable declaration would look in PHP, adjust it
so it does]:
myVar = 'A%'
Then, you would execute your query, plugging in the variable in the
appropriate place [again, I don't know PHP so maybe you need a $myVar
instead of :myVar or at least something along those lines]:
select *
from mytable
where advertiser = :myVar
In your question, you seem to want to put the wildcard character(s) in the
variable name like this:
select *
from mytable
where advertiser = % :myVar
While this might conceivably work in some programming languages, it's not
the way it is normally done in any language I know.
Rhino
- Original Message -
From: [EMAIL PROTECTED]
To: Mysql
Sent: Sunday, May 01, 2005 4:27 PM
Subject: Can I use LIKE this way?
{
$dead_beat_ads = Delete from fsearch_temp where
fsearchKeyword='$str' and fsearchTime='$time' and fsearchIp='$fbip' and
fsearchHost LIKE '%$dbad%' ;
$dead_beat_result = $dbi-query($dead_beat_ads);
}
What I am doing is eliminate Run of site advertisers from our network
LIKE EBAY.com I would rather give free crawler results than to allow this
spam advertising anymore. So I before I develope our XML output of search
results to our partners I am going to remove them but since their tracking
can change All I want to do is search for them by host Like ebay.com and I
will pull this out in real time thus all the $time requirements. This will
mean that no matter which of our partners ebay buys advertising from they
won't be displayed through our network. Or if someone buys a commission
junciton ad we will pull them out by the host even though the click url may
be a qcksrv ad.
So the question is can that % be butted up against the variable or
should I put a space in there. While reading the LIKE on Mysql .com it
talks about 2nd position with a space but I didn't know if that meant for
the use of %_$dbad_% or if % $dbad % would have it looking for second
position stuff.
Thanks
Donny Lairson
President
29 GunMuse Lane
P.O. box 166
Lakewood NM 88254
http://www.gunmuse.com
469 228 2183
No virus found in this incoming message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.11.0 - Release Date: 29/04/2005
Get a Random Row on a HUGE db
I am wanting to
display a random page from my site, But I have over 12,000 articles right now
and we add over 150 per day. What I wound up doing was a Virtual DOS
attack on my own server because the 40 mb db was being loaded to many
times.
I have tons of memory
and a Dell Dual Xeon 2.8 gig.
Can someone think up
a better way of doing this? I wish Mysql would just bring me back 1 valid
random row It could be used in so many ways it should just be a part of
MySql anyway.
?phpini_set("display_errors", '1');header("Pragma:
private");header("Cache-Control: post-check=0, pre-check=0", false);
header("Cache-Control: no-cache,
must-revalidate");require_once("firebase.conf.php");$dbi = new
DBI(DB_URL);$stmt = "Select * from firebase_content Order By rand() DESC
Limit 0, 1";$result = $dbi-query($stmt);while($row =
$result-fetchRow()){$title = $row-title;$cate =
$row-category;$get = "Select cat_url from firebase_categories
where cat_name='$cate'";$now = $dbi-query($get);$rows =
$now-fetchRow();$url = "">$link = $url .
$title;}header("Location: http://www.prnewsnow.com/$link");exit;/*
Sudo code that I am trying to create to relieve server stress.function
randomRow(table, column) {var maxRow = query("SELECT MAX($column) AS maxID
FROM $table");var randomID;var randomRow;do {randomID =
randRange(1, maxRow.maxID);randomRow = query("SELECT * FROM $table WHERE
$column = $randomID");} while (randomRow.recordCount == 0); return
randomRow;}*/?
ThanksDonny LairsonPresident29
GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
RE: Get a Random Row on a HUGE db
Thanks for that I implemented to my Random code. Same problem that select *
portion is just a nightmare. Remember I selecting 38mb of data when I do
that.
What I want to do is jump to a Valid random row. Now If I didn't delete
content often that would be easy grab the last autoincremented row_id and
get a random number between 1 and End Jump to that row to create the link.
Very fast. Zero load
So what I am trying is this.
$last_row =SELECT from firebase_content LAST_INSERT_ID();
$last_row_query = $dbi-query($last_row);
$last_row_result = $row-id;
But what I am seeing is this:
Object id #9
and not the number that is in the database.
What am I sending to this variable that is wrong?
[snip]
I am wanting to display a random page from my site, But I have over
12,000 articles right now and we add over 150 per day. What I wound up
doing was a Virtual DOS attack on my own server because the 40 mb db was
being loaded to many times.
I have tons of memory and a Dell Dual Xeon 2.8 gig.
Can someone think up a better way of doing this? I wish Mysql would
just bring me back 1 valid random row It could be used in so many ways
it should just be a part of MySql anyway.
?php
ini_set(display_errors, '1');
header(Pragma: private);
header(Cache-Control: post-check=0, pre-check=0, false);
header(Cache-Control: no-cache, must-revalidate);
require_once(firebase.conf.php);
$dbi = new DBI(DB_URL);
$stmt = Select * from firebase_content ORDER BY RAND(NOW()) LIMIT 1;
$result = $dbi-query($stmt);
while($row = $result-fetchRow())
{
$title = $row-title;
$cate = $row-category;
$get = Select cat_url from firebase_categories where
cat_name='$cate';
$now = $dbi-query($get);
$rows = $now-fetchRow();
$url = $rows-cat_url;
$link = $url . $title;
}
header(Location: http://www.prnewsnow.com/$link;);
exit;
/* Sudo code that I am trying to create to relieve server stress.
function randomRow(table, column) {
var maxRow = query(SELECT MAX($column) AS maxID FROM $table);
var randomID;
var randomRow;
do {
randomID = randRange(1, maxRow.maxID);
randomRow = query(SELECT * FROM $table WHERE $column = $randomID);
} while (randomRow.recordCount == 0); return randomRow;
}
*/
?
[/snip]
Try this ...
SELECT * FROM foo ORDER BY RAND(NOW()) LIMIT 1;
12000 rows is not huge at all, so this should be pretty quick
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Get a Random Row on a HUGE db
What I had to do was do this for my navigation db and not my content db. My
server can easily handle lots of calls to a 4mb table then tell it to fetch
the content once that has been achieved.
The reason I bringing this up is this seems to be a patched way of doing
this.
If I have 40,000 items in db that get updated and row_ids change for a
catalog and want to randomly display a product.
I should be able to ask Mysql for a Random valid row. It indexes with a
Primary so it knows what valid at that time. There is too much jumping
around just to say get random and be fair about it, so no one row comes up
every time or more often that others.
Gunmuse,
SELECT from firebase_content LAST_INSERT_ID()
In that cmd, 'from ...' ain't right.
I didn't understand either what's wrong with ORDER BY RAND() LIMIT 1.
Also check the Perl manual for how to retrieve a single value.
PB
-
[EMAIL PROTECTED] wrote:
Thanks for that I implemented to my Random code. Same problem that select
*
portion is just a nightmare. Remember I selecting 38mb of data when I do
that.
What I want to do is jump to a Valid random row. Now If I didn't delete
content often that would be easy grab the last autoincremented row_id and
get a random number between 1 and End Jump to that row to create the link.
Very fast. Zero load
So what I am trying is this.
$last_row =SELECT from firebase_content LAST_INSERT_ID();
$last_row_query = $dbi-query($last_row);
$last_row_result = $row-id;
But what I am seeing is this:
Object id #9
and not the number that is in the database.
What am I sending to this variable that is wrong?
[snip]
I am wanting to display a random page from my site, But I have over
12,000 articles right now and we add over 150 per day. What I wound up
doing was a Virtual DOS attack on my own server because the 40 mb db was
being loaded to many times.
I have tons of memory and a Dell Dual Xeon 2.8 gig.
Can someone think up a better way of doing this? I wish Mysql would
just bring me back 1 valid random row It could be used in so many ways
it should just be a part of MySql anyway.
?php
ini_set(display_errors, '1');
header(Pragma: private);
header(Cache-Control: post-check=0, pre-check=0, false);
header(Cache-Control: no-cache, must-revalidate);
require_once(firebase.conf.php);
$dbi = new DBI(DB_URL);
$stmt = Select * from firebase_content ORDER BY RAND(NOW()) LIMIT 1;
$result = $dbi-query($stmt);
while($row = $result-fetchRow())
{
$title = $row-title;
$cate = $row-category;
$get = Select cat_url from firebase_categories where
cat_name='$cate';
$now = $dbi-query($get);
$rows = $now-fetchRow();
$url = $rows-cat_url;
$link = $url . $title;
}
header(Location: http://www.prnewsnow.com/$link;);
exit;
/* Sudo code that I am trying to create to relieve server stress.
function randomRow(table, column) {
var maxRow = query(SELECT MAX($column) AS maxID FROM $table);
var randomID;
var randomRow;
do {
randomID = randRange(1, maxRow.maxID);
randomRow = query(SELECT * FROM $table WHERE $column = $randomID);
} while (randomRow.recordCount == 0); return randomRow;
}
*/
?
[/snip]
Try this ...
SELECT * FROM foo ORDER BY RAND(NOW()) LIMIT 1;
12000 rows is not huge at all, so this should be pretty quick
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.10.2 - Release Date: 4/21/2005
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Get a Random Row on a HUGE db
This difference between using a 40 mb table and 4mb table with the same
traffic was a 70 server load versus a .9 server load. So it was the amount
of data that I was selecting that was choking this feature.
-
[EMAIL PROTECTED] wrote:
Thanks for that I implemented to my Random code. Same problem that select
*
portion is just a nightmare. Remember I selecting 38mb of data when I do
that.
What I want to do is jump to a Valid random row. Now If I didn't delete
content often that would be easy grab the last autoincremented row_id and
get a random number between 1 and End Jump to that row to create the link.
Very fast. Zero load
So what I am trying is this.
$last_row =SELECT from firebase_content LAST_INSERT_ID();
$last_row_query = $dbi-query($last_row);
$last_row_result = $row-id;
But what I am seeing is this:
Object id #9
and not the number that is in the database.
What am I sending to this variable that is wrong?
[snip]
I am wanting to display a random page from my site, But I have over
12,000 articles right now and we add over 150 per day. What I wound up
doing was a Virtual DOS attack on my own server because the 40 mb db was
being loaded to many times.
I have tons of memory and a Dell Dual Xeon 2.8 gig.
Can someone think up a better way of doing this? I wish Mysql would
just bring me back 1 valid random row It could be used in so many ways
it should just be a part of MySql anyway.
?php
ini_set(display_errors, '1');
header(Pragma: private);
header(Cache-Control: post-check=0, pre-check=0, false);
header(Cache-Control: no-cache, must-revalidate);
require_once(firebase.conf.php);
$dbi = new DBI(DB_URL);
$stmt = Select * from firebase_content ORDER BY RAND(NOW()) LIMIT 1;
$result = $dbi-query($stmt);
while($row = $result-fetchRow())
{
$title = $row-title;
$cate = $row-category;
$get = Select cat_url from firebase_categories where
cat_name='$cate';
$now = $dbi-query($get);
$rows = $now-fetchRow();
$url = $rows-cat_url;
$link = $url . $title;
}
header(Location: http://www.prnewsnow.com/$link;);
exit;
/* Sudo code that I am trying to create to relieve server stress.
function randomRow(table, column) {
var maxRow = query(SELECT MAX($column) AS maxID FROM $table);
var randomID;
var randomRow;
do {
randomID = randRange(1, maxRow.maxID);
randomRow = query(SELECT * FROM $table WHERE $column = $randomID);
} while (randomRow.recordCount == 0); return randomRow;
}
*/
?
[/snip]
Try this ...
SELECT * FROM foo ORDER BY RAND(NOW()) LIMIT 1;
12000 rows is not huge at all, so this should be pretty quick
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.10.2 - Release Date: 4/21/2005
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: zip code search within x miles
Http://www.gunmuse.com Ok I use a storelocator. First if you have 8000 + records it becomes an issue. BUT Lat and long is in minutes and minutes can be used to estimate miles. By Breaking down the lat and long, Breaking down the Zip to a two digit prefix 88254 becomes 88 for indexing (Because the post offices goes in order folks with some exceptions) Then with a wide lasso you can rope your results to do your math check with. Break your lat and long fields up in hours minutes and seconds and filtering down becomes very easy to do. Learning to read a map before determining the key and distance calculation would help better understand this problem. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Richard Lynch [mailto:[EMAIL PROTECTED] Sent: Monday, April 25, 2005 12:05 AM To: Hank Cc: MySql Subject: Re: zip code search within x miles On Tue, April 19, 2005 8:55 am, Hank said: Talk about over complicating things... here's the above query simplifed. I can not figure out why they were self joining the table three times: SELECT b.zip_code, b.state, (3956 * (2 * ASIN(SQRT( POWER(SIN(((a.lat-b.lat)*0.017453293)/2),2) + COS(a.lat*0.017453293) * COS(b.lat*0.017453293) * POWER(SIN(((a.lng-b.lng)*0.017453293)/2),2) AS distance FROM zips a, zips b WHERE a.zip_code = '90210' GROUP BY distance having distance = 5; You'd have to time it, and *MAYBE* with enough indices this will all work out, but you'd probably be better off doing two queries. One to look up the long/lat for 90210, and another on just zips to calculate the distance. Benchmark on your own hardware and see for yourself. I could be 100% wrong. -- Like Music? http://l-i-e.com/artists.htm -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Client Side Query cache
Linux does some sort of page caching automatically and its the reason for there never being any Free memory in a Linux system. So if you read your db via XML and php and develop a page from that using a CSS style sheet Linux will cache it as a Page.(Took us forever to catch that one, something to do with CSS and no one really knows the answer that we have seen) As for a query cache edit your My.cnf and query_cache_type 1 2 = Cache only if script says to(my suggested setting, make your coders code for caching instead of catch all settings) This does cause some grief if the data changed you can see a cached result instead.(Even though your not supposed to). Also if your running in PPC feeds you can't cache those as you will hit their timeouts and get no redirect results. Our PPC XML Feed (http://www.firebasesofware.com) only allows the link to be valid for 3 minutes before your visitor will be redirected to our front door. This is not a traffic grab. Just 94% of cached results are typically fraud clicks so we don't allow the caching of results. We are more cautious than most (they use 5 minutes) only because we have a very high paying feed and it attracts the low life's of the internet world. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Mister Jack [mailto:[EMAIL PROTECTED] Sent: Thursday, April 14, 2005 4:49 AM To: mysql@lists.mysql.com Subject: Client Side Query cache Hi, I was wondering if there is any query cache code/lib somewhere to cache certains queries ? I'm always doing the same queries, (and the result never change, so I could spare the round-trip to the server), but caching each tine the data for it is a bit of work. Thanks, for your suggestions -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Relative Numeric Values
Wouldn't creating completely unique keys for every animal be a waste of good indexing? It doesn't matter if you 30 or 3000 animals start with the same 3 letters if your using a 3 letter key for speed. As long as you avoided searching through 30,000,000 records. The method you described is of no speed benefit. If you have 30,000,000 records and wind up with 30,000 keys as a result that is a speed improvement. Putting 30,000,000 keys in there all your doing is limiting the amount of characters searched and not the records searched. Indexing everything only slows the MySql down. Stick with your original plan, and reduce to 2 characters for your index if the speed isn't what your looking for still, or throw hardware at the problem at that point. That will reduce the number of records for the first glance at the index. My search engine on a small dual Xeon runs through 1.7 million records with a 2 letter index for keywords in about .2 seconds It only has 8142 keys in the 2 letter index. And I am crawling about 8000 pages a day adding content without seeing a speed drop at this point. When we get to a point of bottlenecking on searches I intend to make a index jumping call. Find * where 2 letter index equals 'ab' and 3 letter index equals 'abcd' I am sure there will be a better way to write that because at that time I am certain abcd may reside on different servers. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Kim Briggs [mailto:[EMAIL PROTECTED] Sent: Saturday, April 16, 2005 10:23 PM To: David Blomstrom Cc: mysql@lists.mysql.com Subject: Re: Relative Numeric Values David, In reading through miscellaneous database design text on the web, I read just the other day that you should not try to include meaningful data in your key values. I assume there will be some kind of lookup tables for species, phylum, whatever. Trying to make your key field smart seems like way too much overhead and complexity. I'm wondering why, if the database is enormous, are you being so short and cryptic with the user-friendly values? my $.02 KB On 4/16/05, David Blomstrom [EMAIL PROTECTED] wrote: I think my question is more oriented towards PHP, but I'd like to ask it on this list, as I suspect the solution may involve MySQL. I'm about to start developing an enormous database focusing on the animal kingdom and want to find a key system more user friendly than the traditional scientific name. So imagine instead a page with the following in the head section: $AnimalID = 'canlup'; This page displays information on the wolf, based on the first three letters of its genus and species name, Canis lupus. Now imagine a page with this value: $AnimalID = 'bal'; This page displays information on the whale family Balaenidae. But what about the whale family Balaenopteridae, which begins with the same three letters? I could solve this problem by adding a numerical key to my database and displaying the following: $AnimalID = 'bal23'; $AnimalID = 'bal24'; The problem with this is that it makes it much harder to work with my data. When tweaking a page or writing a script, I can easily remember that bal = Balaenidae, but I can't possibly remember which numeral is associated with each mammal family. Also, what happens if I add or subtract rows from my database table, and the above values suddenly change to bal27 and bal28? So here's what I think I'd like to do: $AnimalID = 'canlup1'; $AnimalID = 'bal1'; $AnimalID = 'bal2'; The page with canlup1 will display the FIRST (and only) instance of canlup in the database - the wolf. The page with bal1 will display the first instance of bal, which will always be Balaenidae, whether the absolute value is bal27 or bal2884. A page with bal2 will always display the next mammal family that begins with bal, Balaenopteridae. So I THINK all I need to do is create a variable that reflects a particular value's ordinal position in a database... abc1 abc2 abc3, etc. Plus, I'll have to join two or three fields together to form a key; e.g. animals.species + animals.numerals Does anyone know how I can do this? Thanks. __ Do you Yahoo!? Plan great trips with Yahoo! Travel: Now over 17,000 guides! http://travel.yahoo.com/p-travelguide -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: zip code search within x miles
We convert the zip code into a Lat and long. Run the math looking for all other zips in that area, then convert that back to lat long for a mileage calculation of each. I know there's a better way to do this we just haven't seen the benefit in rewriting it now. Watch PHP a lot of this in coming out in functions, and Perl already has some functions to do this I believe you just need the db. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Scott Haneda [mailto:[EMAIL PROTECTED] Sent: Friday, April 15, 2005 4:38 PM To: MySql Subject: zip code search within x miles How are sites doing the search by zip and coming up with results within x miles? Is there some OSS zip code download that has been created for this? -- - Scott HanedaTel: 415.898.2602 http://www.newgeo.com Novato, CA U.S.A. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: zip code search within x miles
I have a copy of the Zip Code db for MySql. Its a few years old but should be 99% accurate compared to new ones. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Scott Gifford [mailto:[EMAIL PROTECTED] Sent: Friday, April 15, 2005 6:01 PM To: Scott Haneda Cc: MySql Subject: Re: zip code search within x miles Scott Haneda [EMAIL PROTECTED] writes: How are sites doing the search by zip and coming up with results within x miles? Is there some OSS zip code download that has been created for this? Zipdy does most of what you want; it needs to be modified to support MySQL instead of PostgreSQL, but that shouldn't be too hard. It also has the great circle function you need to calculate the distances correctly. You can get it from: http://www.cryptnet.net/fsp/zipdy/ If you're using Perl, Geo::PostalCode works very well, though it doesn't use an SQL database at all: http://search.cpan.org/~tjmather/Geo-PostalCode-0.06/lib/Geo/PostalCode. pm ---ScottG. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Boolean searches
When using the Boolean search capabilities, I am finding it terribly slow. Since I am sure MySql uses this on their own site. What version of MySql is the fastest at this 5.0? I am using the PHPMySearch and I really think he did fair job on the crawler part. I built an XML converter for the results so I can now crawl websites and output XML its just the search on even 5000 rows (20mb of data) is very slow. Indexing seems to done properly to support this. I played with tweaking the character settings of the FULLTEXT search took it to 5 tried it at 3 no luck. I just think it should be faster than it is. The default indexes are. URL UNIQUE 5451 URL expiresFlag INDEX 1 expiresFlag title FULLTEXT 1 title keywords body_1 FULLTEXT 1 body_1 body_2 Maybe his PHP is querying the database wrongly or is out of date. I am running 4.1.8 Mysql What would be a good example on how to query the db for bring back ranked results based on what it found? ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
RE: WARNING!!!! abuser on this list?
URL UNIQUE 5451 URL expiresFlag INDEX 1 expiresFlag title FULLTEXT 1 title keywords body_1 FULLTEXT 1 body_1 body_2 Whoops on the html copy, calm down there guy. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: l'[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, April 08, 2005 11:52 AM To: mysql@lists.mysql.com Subject: WARNING abuser on this list? When I opened the message send by: [EMAIL PROTECTED] subject: Boolean searches a pop up box appeared stating Connect to 70.84.29.164 Web host manager Username mymachine/administrator password: Has anybody encountered this problem when you cliked on his email Should we FLAME that guy? In the mean time I am sending an email to [EMAIL PROTECTED] owning that IP address. Here is the headers of his email: Received: (qmail 7362 invoked by uid 109); 8 Apr 2005 17:13:25 - Mailing-List: contact [EMAIL PROTECTED]; run by ezmlm List-ID: mysql.mysql.com Precedence: bulk List-Help: mailto:[EMAIL PROTECTED] List-Unsubscribe: mailto:[EMAIL PROTECTED] List-Post: mailto:mysql@lists.mysql.com List-Archive: http://lists.mysql.com/mysql/182353 Delivered-To: mailing list mysql@lists.mysql.com Received: (qmail 7312 invoked from network); 8 Apr 2005 17:13:24 - Received-SPF: pass (lists.mysql.com: local policy) Reply-To: [EMAIL PROTECTED] From: [EMAIL PROTECTED] To: Mysql mysql@lists.mysql.com Subject: Boolean searches Date: Fri, 8 Apr 2005 11:13:13 -0600 Message-ID: [EMAIL PROTECTED] MIME-Version: 1.0 Content-Type: multipart/related; boundary==_NextPart_000_0098_01C53C2B.F4751020 X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.6604 (9.0.2911.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2527 X-PopBeforeSMTPSenders: [EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED],dealerfin [EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED] .com,gunmuse,[EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED] e.com,[EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED],paymen [EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED] ,redmoon,[EMAIL PROTECTED],[EMAIL PROTECTED],[EMAIL PROTECTED] gunsales.com X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - pistol.gunmuse.us X-AntiAbuse: Original Domain - lists.mysql.com X-AntiAbuse: Originator/Caller UID/GID - [0 0] / [47 12] X-AntiAbuse: Sender Address Domain - gunmuse.com X-Source: X-Source-Args: X-Source-Dir: X-Virus-Checked: Checked X-Spam-Checker-Version: SpamAssassin 3.0.2 (2004-11-16) on c.spam X-Spam-Status: No, score=1.1 required=5.0 tests=DNS_FROM_AHBL_RHSBL, HTML_MESSAGE,HTML_TAG_EXIST_TBODY,NORMAL_HTTP_TO_IP,NO_REAL_NAME, WEIRD_PORT autolearn=disabled version=3.0.2 X-Spam-Level: * x-html thanks Laurie -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Boolean searches
MY first post had html in it where I posted my indexin that I am using. URL UNIQUE 5451 URL expiresFlag INDEX 1 expiresFlag title FULLTEXT 1 title keywords body_1 FULLTEXT 1 body_1 body_2 Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, April 08, 2005 11:13 AM To: Mysql Subject: Boolean searches When using the Boolean search capabilities, I am finding it terribly slow. Since I am sure MySql uses this on their own site. What version of MySql is the fastest at this 5.0? I am using the PHPMySearch and I really think he did fair job on the crawler part. I built an XML converter for the results so I can now crawl websites and output XML its just the search on even 5000 rows (20mb of data) is very slow. Indexing seems to done properly to support this. I played with tweaking the character settings of the FULLTEXT search took it to 5 tried it at 3 no luck. I just think it should be faster than it is. The default indexes are. Maybe his PHP is querying the database wrongly or is out of date. I am running 4.1.8 Mysql What would be a good example on how to query the db for bring back ranked results based on what it found? Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183
RE: mysql_query Looping
We overcame that problem to infinite levels with our Blog Software. Instead of loops we use Anchor Points in the url to tell the navigation where it was at all times. This allows for a more dynamic navigation system as you can have Nav trees that not Expand with more subcatorgies but also collapse back into a single point. Example http://mydomain.com/CAT1/CAT2/CAT3/CAT4/ARTICLE1.html Now You have also a tree like http://mydomain.com/CAT1/CAT4/Article1.html http://mydomain.com/CAT1/CAT3/Article1.html There is no duplication of data but Say a privacy policy or Contact info page that needs to be everywhere doesn't need multiple links in the db to make it happen because we are looking at two anchor points. CAT1 and Article1.html What's in the middle doesn't matter because the its not on that page at that time. (YET IT IS IN THE URL) This means you can build freely pages without regard to where they need to go or how to move them later. As for your Expanding TREE (Like DHTML or JavaScript) for the next level down to give it a windows appearance. That would simply be a matter of calling the next anchor points on mouse over and showing the display Some javascript(DHTML) and CSS styling). Its a very complicated Nav system but it allows us to be so versatile when we decide to use third party programs we can create a Nav structure within our domain's website that will run it entirely all from a clean interface no coding required. Just input urls and hit enter. Here's some links to our software its free to use but not GPL as we have done some ground breaking stuff with it. You may not want to try and reinvent the wheel when all you need to do is create a CSS style sheet on our system to do a mouse over menu. http://www.firebasesoftware.com/firebase_downloads/firebase2.0_client_linux. zip http://www.firebasesoftware.com/firebase_downloads/firebase2.0_client_window s.zip Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Celebrity Institute [mailto:[EMAIL PROTECTED] Sent: Friday, April 08, 2005 1:32 PM To: mysql@lists.mysql.com Subject: mysql_query Looping Hi, just got intro to the list and i hope I can find some help here. I'm trying to figure out the true syntax of a project im working on, basicly theirs a menu on my site that has several catagoies and in those catagories are secondary, and some times 3rd level and 4th level sub lists, i want to do a on click open up and ive got a good script for on click fold out lists but its the SQL syntax loops im not sure how to configure so im trying to figure out at this point how to make a nested list of the items. Then I'll worry about colasping them. heres the psudo code im useing as my guide for my layout Quote ADMIN SIDE Category Tool We would need to set a order of display for the category, and where its bound, (main catagoies would be bound to Main Menu, while subs would be bound to their master). Code wise we need something like if Master = Main Menu display these entries order by Order of display for the main link and then to show the sub links we would need a nested loop to the effect recursive if master = Thisentry[x] display these entries order by Order of display 1. Beauty and Health 1.1. Cosmetics 1.2. Diet Nutrition 1.3. Fashion 1.4. Fitness 2. Computer/Electronics 2.1. Software 2.2. Hardware 2.3. Internet 2.4. Photography 2.5. Wireless 2.6. Audio 2.7. Video 3. Home and Child Care 3.1. Art and Decoration 3.2. Career/Work 3.3. Eating And Dinning 3.4. Education 3.5. Gifts 3.6. Household 3.6.1. Bedding 3.6.2. Flooring 3.6.3. Furniture 3.6.4. Houseware/Appliances 3.6.5. Gaardening 3.6.6. Tools 3.7. Pet 3.8. Staffing Services 4. On The Go 4.1. Autos 4.2. Planes 4.3. Travel Getaways 4.4. Yacts 5. Recreation 5.1. Movies 5.2. Music 5.3. TV 5.4. Radio 5.5. Reading 5.6. Games 6. Services 6.1. Business 6.2. Charities 6.3. Finance 6.4. Insurance 6.5. Legal 6.6. Medical 6.7. Real Estate 7. Shopping (Todays Special and Hot Stuff sub areas) 7.1. Home 7.2. Apparel and Accessories 7.3. Beauty and Health 7.4. Books, Movies Music 7.5. Computing and Office 7.6. Gifts, Flowers Gourmet 7.7. Jewelry Watches 7.8. Sports Outdoors 7.9. Toys, Kids Baby 7.10. A-Z Store Directory 8. Sports 8.1. Clothing and Gear 8.2. Equipment 9. On The Runway this is pre-coding im basicly working on my design documentation i know what i want in psudo-code and am trying to figure out what i need to do in real code to get there. Basicly i know how i need to set up the DB but am unsure on just how precisly to properly display that data in a format similar to the one above where's some titles are sub section of a link or sub sections of a sub section something to the effect of code
FW: GWAVA Sender Notification (Spam)
This is what I call WAY OVER REACTING. This member turned my email into the spam report immediately without thinking. Now I am recieving these. ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183 -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]Sent: Friday, April 08, 2005 12:51 PMSubject: GWAVA Sender Notification (Spam) A message sent by you was blocked by GWAVA - Content protection for Novell GroupWise. The message was blocked for the following reason(s): Spam The message contained the following information: Subject: RE: WARNING abuser on this list? From: [EMAIL PROTECTED] Recipient(s): [No To Addresses] [No Cc Addresses] [EMAIL PROTECTED], [EMAIL PROTECTED] The following information details the events that prevented delivery of this message: Event Details Spam The message was identified as potential spam
RE: Performance Tuning - Table Joins
Your not indexing properly this should be a blink of a search. Or your looping your loops when you search. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: mos [mailto:[EMAIL PROTECTED] Sent: Monday, April 04, 2005 1:30 PM To: MySQL list Subject: Re: Performance Tuning - Table Joins At 12:22 PM 4/4/2005, you wrote: I have been struggling to maintain decent performance on a web/database server for a good 6 months now due to MySQL performance issues. I have decided that my best option at this point is to take it to the list, so in advance, I thank you all for taking a look. There is no error messages that can be posted, so I will try and describe what's happening as best I can. I am joining 3 tables in one query. I have had numerous people examine the queries and all have given their stamp of approval. What happens when I run it is MySQL takes the processor for a ride, spiking it to 100% until I restart mysqld. The tables range from 50,000 to 85,000 records, and the join is only supposed to return 1 record. My question to you is this: are there changes I can make to the configuration to improve performance? --or-- is data de-normalization my best option? Is there any more information you need from me to answer this question? Current setup: 2.4ghz Pentium 4, 1gb ram, 360gb 4-disc raid 5 array w/ 3ware chassis and card, fedora core 3 w/ all patches and updates, selinux -disabled-, mysql 4.1.10a, MyISAM table format. Again, thank you all in advance, Jason Jason, Try running Analyze Table on each of the tables. This will rebalance the index and get rid of deleted space. Returning one row from a 3 table join should take only ms if you're using indexes properly. Mike -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: license question
MySql loses money from many vendors on this very point. Of which they do not budge. We have a Point of Sale software company who can distribute Oracle cheaper. They only require a percentage of the final product price that their product is packaged with. When the company explained they would rather use MySql an pay them the same rates MySql refused. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Daevid Vincent [mailto:[EMAIL PROTECTED] Sent: Wednesday, March 30, 2005 10:18 PM To: 'Pat Ballard'; mysql@lists.mysql.com Subject: RE: license question As my company and I understand it, if you intend on distributing mySQL on this appliance and the appliance is a sealed box with your own proprietary code (like PHP or C or Java or whatever) that interfaces to the STOCK/Untouched RDBMS, you NEED a mySQL Commercial License. This license is a ridiculous $600 per unit which makes it completely unrealistic for any large scale deployment!!! I mean, I don't mind paying someone for their work, but I was thinking more like $50 per unit, not 10 times that. If someone from mySQL can clarify that would be great, but this is how I read the license and that's why we've stuck to v4.0.18 which was GPL. http://www.mysql.com/company/legal/licensing/opensource-license.html Our software is 100% GPL (General Public License); if yours is 100% GPL compliant, then you have no obligation to pay us for the licenses. Free use for those who never copy, modify or distribute. As long as you never distribute the MySQL Software in any way, you are free to use it for powering your application, irrespective of whether your application is under GPL license or not. If you are a private individual you are free to use MySQL software for your personal applications as long as you do not distribute them. If you distribute them, you must make a decision between the Commercial License and the GPL. http://www.mysql.com/company/legal/licensing/commercial-license.html Building a hardware system that includes MySQL and selling that hardware system to customers for installation at their own locations. If you include the MySQL server with an application that is not licensed under the GPL or GPL-compatible license, you need a commercial license for the MySQL server. -Original Message- From: Pat Ballard [mailto:[EMAIL PROTECTED] Sent: Wednesday, March 30, 2005 4:03 PM To: mysql@lists.mysql.com Subject: license question Suppose i distribute MySQL-4.1 with an appliance, which is a sealed x86 machine running a Linux distribution made by another entity (ok, it's Red Hat). I don't write any code that's directly linked to MySQL, I'm only using the existing php-mysql, etc., packages already provided by the distribution, plus some third-party apps that are under GPL and link to MySQL (applications that access MySQL, not written by me, but are Open Source GPL projects off SourceForge and other places - i just bundle them with the appliance). Any code that I write personally is PHP and sits on top of the php-mysql module provided by Red Hat. The end-user has no direct visibility to the database, in fact, the end-user might never know it's MySQL - all that is visible is the PHP interface, via Apache. In this case, what's the license? Is MySQL still free (under GPL)? -- Pat Ballard -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Slow access Apache + PHP + MySQL
Your request is just to broad to answer. There are lots of places to look but I doubt php is causing your slowdowns. Poorly written php code will cause slowdowns not the server compiled php though. Lots of server monitoring script monitoring and logging you can start to trace down problems like this. You also missed an important part of any server with slowdowns and that is the network itself can cause load and grief. I can tell you that most default setups are very tight on resource use, and the helpful information others will point you to on the web just wasn't written for 4 gig ram and 64bit servers. More along the line of a nice high-performance PIII. So be careful what you read. Learn the math to do it right and not the exampled settings you see in forums. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Andre Matos [mailto:[EMAIL PROTECTED] Sent: Saturday, March 26, 2005 8:47 AM To: 'mysql@lists.mysql.com ' Subject: Slow access Apache + PHP + MySQL Hi List, I have 4 web based systems developed using PHP4 and MySQL accessed for 10 users. The Web Server and Database Server were running ok on a Mac OS X 10.3 G4 dual. However, since we move to a new server, the access becomes very slow. This was not expected since we move to a 64 bits high performance machine. Now, we are using MySQL version 4.1.9 with Apache 2.0.52 and PHP 4.3.10, all compiled and running on a Linux Fedora X86_64. My first thought was the systems, but since I have not changed 3 of the 4 systems, I start to look to the database. I monitored the MySQL using MySQL Administrator, but I couldn't identify any problem. It looks ok, but not completely sure if really is. The system administrator told me that could be the PHP session, but again, he also was not complete sure about this. It is a big problem since I need to check in 3 places: MySQL, Apache, or PHP. Does anyone had this kind of problem or has any suggestion or direction to help me to identify and solve this issue? Any help will be appreciated!!! Thanks. Andre -- Andre Matos [EMAIL PROTECTED] -- Andre Matos [EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Heap table says its Full?
I took a guess at that yesterday. I left the tmp_table_size 128M added the line max_heap_table_size 500M But to no avale. Still limited in the number to 12.7M I am using 4.1.8 as installed by Cpanel. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Harrison Fisk [mailto:[EMAIL PROTECTED] Sent: Friday, March 25, 2005 1:30 AM To: [EMAIL PROTECTED] Cc: Mysql Subject: Re: Heap table says its Fuul? Hi, On Mar 24, 2005, at 6:07 PM, [EMAIL PROTECTED] wrote: Mysql is telling me my Heap table is full. Now I set it to 128M. my.cnf line tmp_table_size = 128M Try changing the setting called max_heap_table_size. tmp_table_size only has to do with internal temporary tables that are used to resolve a query (ie. when you see a 'Using Temporary' in the EXPLAIN) The Table filled up at 12.7M This appears to be very close to 128M with a decimal out of place. Did I find a Bug? Am I doing something wrong? Is the tmp_table_size a PER TABLE or for all mysql heap tables? I can't seem to get past this 12.7M mark I need 128M of heap to run my looping searches with. CREATE TABLE `fsearch_searchheap` ( `searchAffid` int(11) NOT NULL default '0', `searchKeyword` varchar(100) NOT NULL default '', `searchReferrer` varchar(100) NOT NULL default '', `searchIp` varchar(15) NOT NULL default '', KEY `searchAffid` (`searchAffid`), KEY `searchKeyword` (`searchKeyword`) ) ENGINE=MEMORY DEFAULT Select * from fsearch_search; Regards, Harrison -- Harrison C. Fisk, Trainer and Consultant MySQL AB, www.mysql.com Get a jumpstart on MySQL Cluster -- http://www.mysql.com/consulting/packaged/cluster.html -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Heap table says its Full?
Ok, Never mind my last statement because I didn't change anything and it worked this morning. Next problem I copied a 21MB db to the heap and it reported 248M of data once there? Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, March 25, 2005 8:12 AM To: Harrison Fisk Cc: Mysql Subject: RE: Heap table says its Full? I took a guess at that yesterday. I left the tmp_table_size 128M added the line max_heap_table_size 500M But to no avale. Still limited in the number to 12.7M I am using 4.1.8 as installed by Cpanel. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Harrison Fisk [mailto:[EMAIL PROTECTED] Sent: Friday, March 25, 2005 1:30 AM To: [EMAIL PROTECTED] Cc: Mysql Subject: Re: Heap table says its Fuul? Hi, On Mar 24, 2005, at 6:07 PM, [EMAIL PROTECTED] wrote: Mysql is telling me my Heap table is full. Now I set it to 128M. my.cnf line tmp_table_size = 128M Try changing the setting called max_heap_table_size. tmp_table_size only has to do with internal temporary tables that are used to resolve a query (ie. when you see a 'Using Temporary' in the EXPLAIN) The Table filled up at 12.7M This appears to be very close to 128M with a decimal out of place. Did I find a Bug? Am I doing something wrong? Is the tmp_table_size a PER TABLE or for all mysql heap tables? I can't seem to get past this 12.7M mark I need 128M of heap to run my looping searches with. CREATE TABLE `fsearch_searchheap` ( `searchAffid` int(11) NOT NULL default '0', `searchKeyword` varchar(100) NOT NULL default '', `searchReferrer` varchar(100) NOT NULL default '', `searchIp` varchar(15) NOT NULL default '', KEY `searchAffid` (`searchAffid`), KEY `searchKeyword` (`searchKeyword`) ) ENGINE=MEMORY DEFAULT Select * from fsearch_search; Regards, Harrison -- Harrison C. Fisk, Trainer and Consultant MySQL AB, www.mysql.com Get a jumpstart on MySQL Cluster -- http://www.mysql.com/consulting/packaged/cluster.html -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Heap table says its Fuul?
Mysql is telling me my Heap table is full. Now I set it to 128M. my.cnf line tmp_table_size = 128M The Table filled up at 12.7M This appears to be very close to 128M with a decimal out of place. Did I find a Bug? Am I doing something wrong? Is the tmp_table_size a PER TABLE or for all mysql heap tables? I can't seem to get past this 12.7M mark I need 128M of heap to run my looping searches with. CREATE TABLE `fsearch_searchheap` ( `searchAffid` int(11) NOT NULL default '0', `searchKeyword` varchar(100) NOT NULL default '', `searchReferrer` varchar(100) NOT NULL default '', `searchIp` varchar(15) NOT NULL default '', KEY `searchAffid` (`searchAffid`), KEY `searchKeyword` (`searchKeyword`)) ENGINE=MEMORY DEFAULT Select * from fsearch_search; ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
RE: Auto loading a table
Thanks that gives me options, yes table was already created what I wanted was for the table itself to know that when MySql reloads to go an get all from another table. I was understanding this was just something I did when I created the table the first time as a Character of the table to know on load select * from Test2 This allows me to maintain stability and speed at the same time. While I write to 2 tables I always read from 1 and reading is done 95 times more often at least. I have set the My.cnf to 128M for memory tables as default but it appears I still stop at the 10M limit on memory tables anyway. Should I add something into the creation of the table to override the defaults locally with that table? Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Michael Stassen [mailto:[EMAIL PROTECTED] Sent: Monday, March 14, 2005 8:28 AM To: [EMAIL PROTECTED] Cc: Gleb Paharenko; mysql@lists.mysql.com Subject: Re: Auto loading a table [EMAIL PROTECTED] wrote: [Donny Lairson] Quick bump I never got an answer I have a table fsearch_temp I use it as a memory table to keep things light and fast but after a restart I want to repopulate some data automatically. So I thought I just said load from TEST2 which would by a myisam table containing the hardbackup I need. But obviously not the right way of saying this. I must be reading the instructions wrong can someone clarify this for me? snip Which instructions are you reading? I expect you get a syntax error, right? From the manual http://dev.mysql.com/doc/mysql/en/create-table.html, the correct syntax is CREATE [TEMPORARY] TABLE [IF NOT EXISTS] tbl_name [(create_definition,...)] [table_options] [select_statement] Gleb Paharenko wrote: Hello. ) ENGINE = MEMORY LOAD FROM TEST2 DEFAULT CHARSET = utf8 AUTO_INCREMENT =0 You should use select statement, not LOAD. For example: CREATE TABLE . SELECT * FROM TEST2; And table options like DEFAULT CHARSET you should put before select statement. See: http://dev.mysql.com/doc/mysql/en/create-table.html I think this is accurate but misleading. CREATE ... SELECT adds columns from the SELECT to the columns defined in the CREATE, so you cannot fix this simply by getting the last line right. You have to leave out the column definitions. On the other hand (from the manual page you cite), CREATE TABLE ... SELECT does not automatically create any indexes for you. This is done intentionally to make the statement as flexible as possible. If you want to have indexes in the created table, you should specify these before the SELECT statement... so you do need to keep the index definitions. Thus, assuming fsearch_temp's create_definition matches that of table TEST2, to create fsearch_temp as a copy of TEST2, you would CREATE TABLE fsearch_temp ( PRIMARY KEY (fsearchId), KEY fsearchIp (fsearchIp) ) ENGINE = MEMORY DEFAULT CHARACTER SET utf8 SELECT * FROM TEST2; but I don't think this is what you want, either. First, there is this caveat (from the manual): Some conversion of column types might occur. For example, the AUTO_INCREMENT attribute is not preserved, and VARCHAR columns can become CHAR columns. To avoid that, you need to first CREATE the table, then populate it with a copy of TEST2 in a separate INSERT ... SELECT statement. See the manual for details http://dev.mysql.com/doc/mysql/en/insert-select.html. In any case, MEMORY tables don't go away unless they are dropped. Only the rows disappear when mysql stops. If you've previously created this table and haven't dropped it, it should still exist as an empty table on startup. In that case, you only need to reload the rows. INSERT INTO fsearch_temp SELECT * FROM TEST2; Michael -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Auto loading a table
Your rudeness is not warranted in any manner. If the lists answer was to Hire a consultant with every question what would be the point of peer to peer help. The list gets busy and at times and I got excellent responses from qualified and courteous people, from a simple bump after waiting a respectful 4 days. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: sol beach [mailto:[EMAIL PROTECTED] Sent: Sunday, March 13, 2005 3:11 PM To: [EMAIL PROTECTED] Subject: Re: Auto loading a table With free advice, you get what you paid for it. Nobody here owes you or anyone a response. If you expect answers, pay a consultant for them. On Sun, 13 Mar 2005 14:23:40 -0700, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: [Donny Lairson] Quick bump I never got an answer -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Auto loading a table
[Donny Lairson] Quick bump I never got an answer I have a table fsearch_temp I use it as a memory table to keep things light and fast but after a restart I want to repopulate some data automatically. So I thought I just said load from TEST2 which would by a myisam table containing the hardbackup I need. But obviously not the right way of saying this. I must be reading the instructions wrong can someone clarify this for me? CREATE TABLE `fsearch_temp` ( `fsearchId` bigint( 19 ) unsigned NOT NULL AUTO_INCREMENT , `fsearchHost` varchar( 100 ) NOT NULL default '', `fsearchSite` varchar( 255 ) NOT NULL default '', `fsearchDescription` varchar( 255 ) NOT NULL default '', `fsearchUrl1` varchar( 255 ) NOT NULL default '', `fsearchUrl2` varchar( 255 ) NOT NULL default '', `fsearchUrl3` varchar( 255 ) NOT NULL default '', `fsearchUrl4` varchar( 255 ) NOT NULL default '', `fsearchTime` int( 11 ) NOT NULL default '0', `fsearchIp` varchar( 22 ) NOT NULL default '', `fsearchKeyword` varchar( 100 ) NOT NULL default '', `fsearchBid` varchar( 11 ) NOT NULL default '0', `fsearchClicked` varchar( 6 ) NOT NULL default 'no', PRIMARY KEY ( `fsearchId` ) , KEY `fsearchIp` ( `fsearchIp` ) ) ENGINE = MEMORY LOAD FROM TEST2 DEFAULT CHARSET = utf8 AUTO_INCREMENT =0 Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183
Auto loading a table
I have a table fsearch_temp I use it as a memory table to keep things light and fast but after a restart I want to repopulate some data automatically. So I thought I just said load from TEST2 which would by a myisam table containing the hardbackup I need. But obviously not the right way of saying this. I must be reading the instructions wrong can someone clarify this for me? CREATE TABLE `fsearch_temp` ( `fsearchId` bigint( 19 ) unsigned NOT NULL AUTO_INCREMENT ,`fsearchHost` varchar( 100 ) NOT NULL default '',`fsearchSite` varchar( 255 ) NOT NULL default '',`fsearchDescription` varchar( 255 ) NOT NULL default '',`fsearchUrl1` varchar( 255 ) NOT NULL default '',`fsearchUrl2` varchar( 255 ) NOT NULL default '',`fsearchUrl3` varchar( 255 ) NOT NULL default '',`fsearchUrl4` varchar( 255 ) NOT NULL default '',`fsearchTime` int( 11 ) NOT NULL default '0',`fsearchIp` varchar( 22 ) NOT NULL default '',`fsearchKeyword` varchar( 100 ) NOT NULL default '',`fsearchBid` varchar( 11 ) NOT NULL default '0',`fsearchClicked` varchar( 6 ) NOT NULL default 'no',PRIMARY KEY ( `fsearchId` ) ,KEY `fsearchIp` ( `fsearchIp` ) ) ENGINE = MEMORY LOAD FROM TEST2 DEFAULT CHARSET = utf8 AUTO_INCREMENT =0 ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
Loading data on startup
I need to copy data from TABLE A to TABLE B(Memory Table)on MySql startup or restart. MySql --init-file on startup is obviously something I need to use could I get an example of what a sql would look like to start the memory table and completely copy data from table A. ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
Is there a limit on Auto-increment
I am using the memory table in 4.1 to auto increment is there a limit to how big that number can get? ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
Memory Tables and My.cnf config may not be right
My Memory table hit 16Mb and locked up. Is there something in my.cnf that I don't have correct. I thought I set it to 128MB memory tables. max_connections = 3500max_user_connections = 1500key_buffer = 750Mmyisam_sort_buffer_size = 130Mjoin_buffer_size = 128Mread_buffer_size = 1Msort_buffer_size = 128Mread_rnd_buffer_size = 16Mtable_cache = 28192thread_cache_size = 1512wait_timeout = 7connect_timeout = 10max_allowed_packet = 16Mmax_connect_errors = 20query_cache_limit = 8Mquery_cache_size = 32Mquery_cache_type = 0skip-innodbthread_concurrency = 8safe-show-databaseinteractive_timeout= 15tmp_table_size = 12800 ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
Delete without Overhead on a MEMORY.
We are getting lots of Overhead in our MEMORY table when we delete rows that are to old. So How do we delete from the table an not consumer MEMORY that we want later? ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
RE: Memory Tables and My.cnf config may not be right
Don't want this to roll to far down the list. My Memory table hit 16Mb and locked up. Is there something in my.cnf that I don't have correct. I thought I set it to 128MB memory tables. max_connections = 3500 max_user_connections = 1500 key_buffer = 750M myisam_sort_buffer_size = 130M join_buffer_size = 128M read_buffer_size = 1M sort_buffer_size = 128M read_rnd_buffer_size = 16M table_cache = 28192 thread_cache_size = 1512 wait_timeout = 7 connect_timeout = 10 max_allowed_packet = 16M max_connect_errors = 20 query_cache_limit = 8M query_cache_size = 32M query_cache_type = 0 skip-innodb thread_concurrency = 8 safe-show-database interactive_timeout= 15 tmp_table_size = 12800 Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183
RE: Yeah worked liked a dream
Ok folks your little words of wisdom crunched our problem out and and we built a metacrawler that is faster than momma and dogpile because we now NEVER touch a harddrive while searching. We will be applying this search to our Blogging software at firebasesoftware.com by Monday. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Tuesday, February 22, 2005 3:17 PM To: Mysql Subject: Memory Tables and My.cnf config may not be right My Memory table hit 16Mb and locked up. Is there something in my.cnf that I don't have correct. I thought I set it to 128MB memory tables. max_connections = 3500 max_user_connections = 1500 key_buffer = 750M myisam_sort_buffer_size = 130M join_buffer_size = 128M read_buffer_size = 1M sort_buffer_size = 128M read_rnd_buffer_size = 16M table_cache = 28192 thread_cache_size = 1512 wait_timeout = 7 connect_timeout = 10 max_allowed_packet = 16M max_connect_errors = 20 query_cache_limit = 8M query_cache_size = 32M query_cache_type = 0 skip-innodb thread_concurrency = 8 safe-show-database interactive_timeout= 15 tmp_table_size = 12800 Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183
Timed truncate of a table.
I am building a temporary memory table to store links in. Instead of having php truncating all links older than 10 minutes is there just a way to have the table do itself automatically? ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
Memory table questions
We are building two copies of our commonly used tables. When we have to write something we write it Both tables: TABLE A TABLE B MEMORY table When we read to run our script we only read from he MEMORY TABLE. My question is how do I tell TABLE B toclone TABLE Aafter a reboot automatically. That is actually part of the table build itself right? what does that SQL look like? ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
Lots of searches
We want to be able to search through 1 million words looking for matches and a count of how many times they have been searched. Example. Words like: guns shotguns longguns long guns shot guns So I want to search all million words for the use of the word 'gun' Then it would return me those 5 phrases since the pattern gun is in the phrase. My question is what is the best way to setup the db to make this search as fast as possible. Right now its taking 45-50 seconds and I would like to get it down to 4-5 seconds. ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
RE: Lots of searches
The memory deal gave me an idea. A little hotcopy to a Memory table once a day and perform the search from there. Since the search is a nicity for the users and not critical to be 100% accurate or up2date. How do you do a hotcopy with php then? Never even seen it done. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Brent Baisley [mailto:[EMAIL PROTECTED] Sent: Friday, February 18, 2005 1:41 PM To: [EMAIL PROTECTED] Cc: Mysql Subject: Re: Lots of searches There's nothing you can do except get really fast hard drives. Since you are searching for random parts of words you can't use an index. So you're stuck with doing a full table scan. Your entire table probably won't fit in memory, so you will be reading from disk. Thus, fast disk drives in a RAID setup. On Feb 18, 2005, at 1:56 PM, [EMAIL PROTECTED] wrote: We want to be able to search through 1 million words looking for matches and a count of how many times they have been searched. Example. Words like: guns shotguns longguns long guns shot guns So I want to search all million words for the use of the word 'gun' Then it would return me those 5 phrases since the pattern gun is in the phrase. My question is what is the best way to setup the db to make this search as fast as possible. Right now its taking 45-50 seconds and I would like to get it down to 4-5 seconds. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -- Brent Baisley Systems Architect Landover Associates, Inc. Search Advisory Services for Advanced Technology Environments p: 212.759.6400/800.759.0577 -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
A Heartfelt thank you.
I would just like to say thank you to everyone on the list who helped hunt down and find quirks while we coded up our latest software. We are absolutely thrilled with the resulting speed and power of our CMS software now. The ability to Optimize peoples websites while still keeping them dynamic is a huge leap in Search engine optimization. We are giving away the software for free if you would like to see the final result download a copy and please give us some comments from your perspective. http://www.firebasesoftware.com/firebase_downloads/FB-ReleaseVersion-1-1.zip Again you have been very helpful and I am sure we finally update our server to 4.1.8 we will be back with a boat load of questions. I know we are interested the sub search functionality of the new MySql. Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: David Blomstrom [mailto:[EMAIL PROTECTED] Sent: Sunday, February 13, 2005 3:18 PM To: mysql@lists.mysql.com Subject: Re: Where's my ODBC icon? --- Andrew Pattison [EMAIL PROTECTED] wrote: However, when I double-clicked odbccp32.cpl, I was rewarded with something similar to what I got before. Not sure what you are looking for then. The myODBC driver should not need configuring, beyond setting up data sources, which is exactly what the control panel applet does for you. There is no program to launch - you configure a data source to allow you to access data, then use your ODBC-capable program to connect to that data source. Start up the ODBC applet and change to the System DSN tab. Next, add a new data source which uses the myODBC driver to connect to the database you want to access via ODBC. Once you have done this, you can then connect to the data source from your ODBC-capable program using the name of the data source. OK, now I understand it a little better. My ultimate goal is to extract some data from some GIS files and import them into MySQL. So it looks like I'm going to be using ODBC to connect a software program called GeoClient (which I haven't begun to figure out yet) to another program called ArcExplorer (which isn't working for me). This should be interesting. :) Thanks. __ Do you Yahoo!? Meet the all-new My Yahoo! - Try it today! http://my.yahoo.com -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Import Excel data into table
NaviCat Thanks Donny Lairson President 29 GunMuse Lane P.O. box 166 Lakewood NM 88254 http://www.gunmuse.com 469 228 2183 -Original Message- From: Steve Grosz [mailto:[EMAIL PROTECTED] Sent: Thursday, January 13, 2005 2:56 PM To: mysql@lists.mysql.com Subject: Import Excel data into table Can anyone tell me a good way to import individual column data into a table? Is there a tool to assist with this? Thanks, Steve -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Pear abstraction class
We are using a pear abstraction class written by Mohammed J. Kabir Problem is its out of date for error reporting. Does anyone know where a current version of this is. Hopefully one that would be a plug in exchange for the one we have so that we don't have to rewrite the software. ThanksDonny LairsonPresident29 GunMuse LaneP.O. box 166Lakewood NM 88254http://www.gunmuse.com469 228 2183
Heap Help
I want to put a table in Ram (HEAP) with a field of at least 500 characters. I do I do this if Blob and text are not allowed? ThanksDonny LairsonPresidenthttp://www.gunmuse.com469 228 2183
Take a normal Table and Make it a Heap table
I have to pull in 200 search rows and store them temporarily in a Table called xmllinks This is so I can track the click on the one link of the 200 I bring down. Nothing is permanently stored in this table This is just a normal table in a db right now but during peak traffic times it bogs down the MySql. What do I have to do to move this one table into a heap of ram. I didn't code this software so please give me a little detail as to whether this is done at the software side or the MySql side. Of what I need to edit to stick this up into ram and verify that its there. I do want to verify its actually there somehow. ThanksDonny LairsonPresidenthttp://www.gunmuse.com469 228 2183
RE: Take a normal Table and Make it a Heap table
I agree and we are rewriting this application ourselves to accomidate these types of issues of making it faster faster faster. But I would like to patch what I have at the same time. Call me greedy Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: sol beach [mailto:[EMAIL PROTECTED] Sent: Friday, November 26, 2004 11:25 AM To: [EMAIL PROTECTED] Subject: Re: Take a normal Table and Make it a Heap table When your only tool is a hammer, then all problems are views as nails. A shovel is a great tool for creating a hole in the ground, but only when the right end is contacts the ground. You seems to be using the wrong end of the computer. With computers, you can have it good, fast, or cheap. Pick any two ( pay the price in the third). I wish you luck in re-inventing the wheel rolling your own custom SCABALBLE application. P.S. Scalability needs to be designed into the archituecture from the start. It rarely can be bolted together after the bottlenecks are encountered, because bottlenecks result from inappropriate original design decisions. On Fri, 26 Nov 2004 11:12:38 -0700, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I have to pull in 200 search rows and store them temporarily in a Table called xmllinks This is so I can track the click on the one link of the 200 I bring down. Nothing is permanently stored in this table This is just a normal table in a db right now but during peak traffic times it bogs down the MySql. What do I have to do to move this one table into a heap of ram. I didn't code this software so please give me a little detail as to whether this is done at the software side or the MySql side. Of what I need to edit to stick this up into ram and verify that its there. I do want to verify its actually there somehow. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Take a normal Table and Make it a Heap table
I ran this and got the following error. Why doesn't heap support autoincrement? Or does it and I need to do something different. CREATE TABLE `xmllinks2` ( `rowID` int(11) NOT NULL auto_increment, `affiliateID` int(11) NOT NULL default '0', `pluginName` varchar(255) NOT NULL default '', `linkID` int(11) NOT NULL default '0', `linkURL` varchar(255) NOT NULL, `bid` decimal(10,4) NOT NULL default '0.', `uniqueData` varchar(255) NOT NULL, `searchDate` datetime NOT NULL default '-00-00 00:00:00', PRIMARY KEY (`rowID`) ) TYPE=HEAP AUTO_INCREMENT=1425725 ; #1164 - The used table type doesn't support AUTO_INCREMENT columns Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, November 26, 2004 11:59 AM To: sol beach Cc: Mysql Subject: RE: Take a normal Table and Make it a Heap table I agree and we are rewriting this application ourselves to accomidate these types of issues of making it faster faster faster. But I would like to patch what I have at the same time. Call me greedy Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: sol beach [mailto:[EMAIL PROTECTED] Sent: Friday, November 26, 2004 11:25 AM To: [EMAIL PROTECTED] Subject: Re: Take a normal Table and Make it a Heap table When your only tool is a hammer, then all problems are views as nails. A shovel is a great tool for creating a hole in the ground, but only when the right end is contacts the ground. You seems to be using the wrong end of the computer. With computers, you can have it good, fast, or cheap. Pick any two ( pay the price in the third). I wish you luck in re-inventing the wheel rolling your own custom SCABALBLE application. P.S. Scalability needs to be designed into the archituecture from the start. It rarely can be bolted together after the bottlenecks are encountered, because bottlenecks result from inappropriate original design decisions. On Fri, 26 Nov 2004 11:12:38 -0700, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I have to pull in 200 search rows and store them temporarily in a Table called xmllinks This is so I can track the click on the one link of the 200 I bring down. Nothing is permanently stored in this table This is just a normal table in a db right now but during peak traffic times it bogs down the MySql. What do I have to do to move this one table into a heap of ram. I didn't code this software so please give me a little detail as to whether this is done at the software side or the MySql side. Of what I need to edit to stick this up into ram and verify that its there. I do want to verify its actually there somehow. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: Take a normal Table and Make it a Heap table
Can I put a MyISAM table into Ram permanently? Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Friday, November 26, 2004 11:59 AM To: sol beach Cc: Mysql Subject: RE: Take a normal Table and Make it a Heap table I agree and we are rewriting this application ourselves to accomidate these types of issues of making it faster faster faster. But I would like to patch what I have at the same time. Call me greedy Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: sol beach [mailto:[EMAIL PROTECTED] Sent: Friday, November 26, 2004 11:25 AM To: [EMAIL PROTECTED] Subject: Re: Take a normal Table and Make it a Heap table When your only tool is a hammer, then all problems are views as nails. A shovel is a great tool for creating a hole in the ground, but only when the right end is contacts the ground. You seems to be using the wrong end of the computer. With computers, you can have it good, fast, or cheap. Pick any two ( pay the price in the third). I wish you luck in re-inventing the wheel rolling your own custom SCABALBLE application. P.S. Scalability needs to be designed into the archituecture from the start. It rarely can be bolted together after the bottlenecks are encountered, because bottlenecks result from inappropriate original design decisions. On Fri, 26 Nov 2004 11:12:38 -0700, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: I have to pull in 200 search rows and store them temporarily in a Table called xmllinks This is so I can track the click on the one link of the 200 I bring down. Nothing is permanently stored in this table This is just a normal table in a db right now but during peak traffic times it bogs down the MySql. What do I have to do to move this one table into a heap of ram. I didn't code this software so please give me a little detail as to whether this is done at the software side or the MySql side. Of what I need to edit to stick this up into ram and verify that its there. I do want to verify its actually there somehow. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Complete php,mysql packages
Wamp servers are an excellent windows install for It elimininates those little hiccups http://www.wampserver.com/en/index.php Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: Danesh Daroui [mailto:[EMAIL PROTECTED] Sent: Wednesday, November 24, 2004 8:59 AM To: [EMAIL PROTECTED] Subject: MySQL and PHP Hi all, I have problem by using PHP and MySQL. I have installed MySQL Server 4.1.7 on a Linux machine with Apache. PHP interpreter has been installed on it by default but I am not sure if PHP modules for MySQL has been installed too or not. On MySQL download section there is only some extensions which has links to PHP site. I couldn't use them really. Can anybody help ? How can I install PHP modules for MySQL so I would work with mysql database through PHP ? Regards, Danesh -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: On but off topic Putting a file in Ram
Actually no. I have a file that is determined to be requested by mysql (Top 100 site) What I am wanting to do is put the images and or files into Ram to serve them from there instead of the harddrive and conserve hd resources for not known tasks. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: Eamon Daly [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 23, 2004 9:17 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: Re: On but off topic Putting a file in Ram The reason I ask this here. Is I have graphics that are loaded by Mysql and was wondering if I can do the same for them since some of these sites can call my server 10-20,000 times a day for that same graphic. I assume you mean that you have image data stored in a MySQL table somewhere and are using a SELECT to fetch and serve it. I think the general consensus would be something along the lines of Don't do that. Apache was /designed/ to serve files quickly, so let it do what it does best. Store just the filenames in MySQL and let Apache handle the rest. Once you've done that, you can do plenty of things to speed up or scale your system, such as mapping the files to memory with mod_file_cache, judicious use of a caching proxy, or the creation of a ramdisk. Eamon Daly - Original Message - From: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Monday, November 22, 2004 8:20 PM Subject: On but off topic Putting a file in Ram I have a small file that calls a search function at Findwhat in case Mysql locally overloads. I just put on a new partner who looks like they may call my server 40 million times a month. I know there is some way to put a file into Ram for super fast response. Question is how do I do this? Will it still write to Mysql from the Ram Drive? What is the downside of doing this? The reason I ask this here. Is I have graphics that are loaded by Mysql and was wondering if I can do the same for them since some of these sites can call my server 10-20,000 times a day for that same graphic. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: On but off topic Putting a file in Ram
Heap/Memory tables that is the phrase I couldn't remember. The data is stored in the file system. I have one file that that is linked to via JavaScript to run a php file and send an output. That file accesses MySql OR if I am overloaded it bypasses my local system and goes directly to Findwhat.com to produce the search. By putting that file into memory I should be able to handle any load fairly easily. Any suggestions on where I should read to learn how to use heap/memory on Linux/enterprise? Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: Victor Pendleton [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 23, 2004 11:10 AM To: [EMAIL PROTECTED] Cc: Eamon Daly; [EMAIL PROTECTED] Subject: Re: On but off topic Putting a file in Ram Is the actual data stored in the database or somewhere in the file system? If you do not have text or blob columns you may be able to use heap/memory tables. [EMAIL PROTECTED] wrote: Actually no. I have a file that is determined to be requested by mysql (Top 100 site) What I am wanting to do is put the images and or files into Ram to serve them from there instead of the harddrive and conserve hd resources for not known tasks. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: Eamon Daly [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 23, 2004 9:17 AM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: Re: On but off topic Putting a file in Ram The reason I ask this here. Is I have graphics that are loaded by Mysql and was wondering if I can do the same for them since some of these sites can call my server 10-20,000 times a day for that same graphic. I assume you mean that you have image data stored in a MySQL table somewhere and are using a SELECT to fetch and serve it. I think the general consensus would be something along the lines of Don't do that. Apache was /designed/ to serve files quickly, so let it do what it does best. Store just the filenames in MySQL and let Apache handle the rest. Once you've done that, you can do plenty of things to speed up or scale your system, such as mapping the files to memory with mod_file_cache, judicious use of a caching proxy, or the creation of a ramdisk. Eamon Daly - Original Message - From: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Monday, November 22, 2004 8:20 PM Subject: On but off topic Putting a file in Ram I have a small file that calls a search function at Findwhat in case Mysql locally overloads. I just put on a new partner who looks like they may call my server 40 million times a month. I know there is some way to put a file into Ram for super fast response. Question is how do I do this? Will it still write to Mysql from the Ram Drive? What is the downside of doing this? The reason I ask this here. Is I have graphics that are loaded by Mysql and was wondering if I can do the same for them since some of these sites can call my server 10-20,000 times a day for that same graphic. -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
How do you do the math
I see guys talk about doing the math to determine how many apache servers you need. How much ram you have or have left. My question is how do you determine how much your MySql can handle? I am running a dell dual 2.8 with 2 gig of ram Raid 5 Whats the max connections MySql can handle? I have mine set in the 1000's now because any lower my server seems to choke. ThanksDonny LairsonPresidenthttp://www.gunmuse.com469 228 2183
On but off topic Putting a file in Ram
I have a small file that calls a search function at Findwhat in case Mysql locally overloads. I just put on a new partner who looks like they may call my server 40 million times a month. I know there is some way to put a file into Ram for super fast response. Question is how do I do this? Will it still write to Mysql from the Ram Drive? What is the downside of doing this? The reason I ask this here. Is I have graphics that are loaded by Mysql and was wondering if I can do the same for them since some of these sites can call my server 10-20,000 times a day for that same graphic. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: Jonathan Duncan [mailto:[EMAIL PROTECTED] Sent: Monday, November 22, 2004 5:34 PM To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Subject: Re: MySQL Books Sasha, Plugs from authors are interesting, but plugs from readers are what really sell a book. I will check it out though. Thank you for the response. Jonathan Sasha Pachev [EMAIL PROTECTED] 11/19/04 5:36 pm Jonathan Duncan wrote: I have the MySQL first edition book by Paul. Still a great reference. However, it being a bit outdated I was hoping to get a more current book and one with more examples, since I learn best by example. The first book has good examples, but more would still help. Therefore, I was comparing reviews online for the following two books: -MySQL, Second Edition by Paul DuBois -Mastering MySQL 4 by Ian Gilfillan Any preferences between these two? Any better suggestions for learning MySQL front and back from a DBA perspective to an end user perspective? Jonathan: May I offer a shameless plug? MySQL Enterprise Solutions. Being the first book I've ever written, it does have its weaknesses, but also has its strengths. For every configuration variable in Chapter 14, and for every status variable in Chapter 15 I went to the source to make sure I understood what was going on behind the scenes before I wrote the description. It is also the only book that I know of so far that discusses MySQL internals (I am working on another one dedicated solely to MySQL Internals). It was written in 2002, so it does focus on 3.23-4.0. However, this is not that big of a minus. Due to the strong commitment of the MySQL team to backwards compatibility, most if not almost everything the book says applies to 4.1 and 5.0. It is just that the newer versions have some new features and options that the book does not cover. -- Sasha Pachev Create online surveys at http://www.surveyz.com/ -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
Thread concurrency
Just out of curiosity I recently turned this up from a 4 to a 16 to see what would choke. Nothing did. Matter of fact it seems like my load went down quite a bit. Is there something wrong with leaving it this way? Since it should be really simple tasks on Xeon processors shouldn't it be able to spawn even more than 16? What was the thinking behind the 2 and 4 numbers in this arena? ThanksDonny LairsonPresidenthttp://www.gunmuse.com469 228 2183
RE: DBManager 3.0
Nice interface, but its crippleware. Anything that may be of use is locked out from testing exactly what they would tell me. I tried to run reports. Need to purchase. Tried to diagram but not sure what that did for me other than tell me what tables I already had. could see some benefit for building a complex db but not needed to test this software. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: COS [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 16, 2004 6:25 AM To: PostgreSQL List; MySQL List Subject: ANN: DBManager 3.0 Hi, We are pleased to announce the awaited DBManager Professional 3.0. This version is a new GUI with lots of new features. DBManager is a client application to manage your databases for Windows environments. Some of the features in this version are - Manage all objects in the database server: MySQL 5, PostgreSQL 8, SQLite 3, Xbase, Interbase, Firebird, MSAccess(*), MSSQL Server(*), Sybase(*), Oracle(*) - New and improved design, Multi Document TAB Interface compatible with Visual Studio .NET - Create, build, import and export queries - Create PHP and ASP(*) scripts for the web to any database flavour(*) (PHP/native and ODBC/ASP) - Import structure and data from a variety of source including: Text Files, XML, MSAccess, MSExcel, Dbase III/Clipper/FoxPro, Paradox, ODBC, etc - Export structure and data to MSAccess, MSExcel, Text Files (CSV/Formatted), HTML, XML, SQL Dumps - User and Privilege (*) Manager - Task Builder (*) to automate processes - Diagram Designer (*) to create and mantain your database structure - New Table Designer - New Query Editor - Report and Form Builders (*) to create reports and Forms in HTML format - Monitor (*) the Activities of the Server, Database or Table with a simple click - DBManager Console to type your commands - Database Comparer (*) to keep your databases updated - and much, much more. (*) Available only in the Enterprise Edition If you want to know more about DBManager or download the new version please go to http://www.dbtools.com.br. Best Regards, Crercio O. Silva / DBTools Software -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
RE: newbie: relationships between fields
NaviCat search for it on google it will make your life much easier. Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -Original Message- From: Amer Neely [mailto:[EMAIL PROTECTED] Sent: Tuesday, November 16, 2004 1:43 PM To: MySQL Cc: Christian Kavanagh Subject: Re: newbie: relationships between fields -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 | Dear list, | | I'm hoping to move to MySQL from Access, mostly so I can use my Mac to build databases. So far I've been able to set up MySQL, connect to it, and create databases and tables. Getting to this point has required a paradigm shift or two on my part (my first question after it was installed was, okay, how the hell do I open up the app and start work?). | | Now I'd like to create some relationships between the tables in my database. But I'm having some trouble getting my head around how to do this - probably because I'm working with an Access paradigm. Imagine I had two tables: | [snip] Perhaps a good session with a relational database book would help you break the 'Access' paradigm :) I have had good success with 'Database Design For Mere Mortals', Michael J. Hernandez, 0-201-69471-9. 'MySQL' by Paul DuBois will also give you a good start to RDBMS concepts and design. - -- /* All outgoing email scanned by AVG Antivirus */ Amer Neely, Softouch Information Services Home of Spam Catcher North Bay Information Technology Networking Group W: www.softouch.on.ca E: [EMAIL PROTECTED] Perl | PHP | MySQL | CGI programming for all data entry forms. We make web sites work! -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (MingW32) Comment: For info see http://www.gnupg.org iEYEARECAAYFAkGaZlAACgkQ3RxspxLYVsWIlACgnWa+wSt1xO8QTws3cldjsI+3 suQAn0i5mmNVOMCBvY2bB4arjZQNKYVs =IyI1 -END PGP SIGNATURE- -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED] -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
4.1.8
Does 4.1.8 address any of the issues I am reading about in 4.1.7? Are all of these issues valid or are you finding that its Lazy coding? Thanks Donny Lairson President http://www.gunmuse.com 469 228 2183 -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe:http://lists.mysql.com/[EMAIL PROTECTED]
UTF-8 compliance
What does this all entail? Can we use Under_Scores in table names. ThanksDonny LairsonPresidenthttp://www.gunmuse.com469 228 2183
XML to Mysql
This may just sound stupid but I am going to ask anyway. We run a search engine and we bring in 3+ XML feeds from other search engines Via perl and PHP. So we can end up with 300 results listed for EACH SEARCH. They are only valid for that ONE SEARCH but we need to track every click for proper payment. As you may imagine this has a huge overhead to our mysql. We are growing and about to rewrite our search engine and would like ideas on how to make the best use of these temporary storage tables. Some of the tracking urls can be 440+ characters long so we can't use varchar. Also If you have 20 searches per minute coming in and 30 seconds between the clicks the table can get quick big quickly. So jumping directly to right stored link immediately if not quicker is a must. Just spitballing for the plan ahead. Wanting to make sure we use all the available speed assests at our disposal. ThanksDonny LairsonPresidenthttp://www.gunmuse.com469 228 2183
Temporary Upgrade for Cpanel?
I read on the Cpanel.net forum that you can start the 4.0.22 mysql with a --new switch to use it as 4.1.7 so you can see if its compatible with my software. Is this true and exactly how? ThanksDonny LairsonPresidenthttp://www.gunmuse.com469 228 2183
