I've an app that uses net/http. I register some handlers with http that need to fetch some stuff from a database before we can proceed to writing the response and be done with the request.
My question is in about which the best pratice is to connect to this database. I want this to work at one request per minute or 10 request per second.
I could connect to database within each handler every time a request comes in. (This would spawn a connection to mysql for each request?)
package main
import (
"database/sql"
_ "github.com/go-sql-driver/mysql"
"net/http"
"fmt"
)
func main() {
http.HandleFunc("/",func(w http.ResponseWriter, r *http.Request) {
db, err := sql.Open("mysql","dsn....")
if err != nil {
panic(err)
}
defer db.Close()
row := db.QueryRow("select...")
// scan row
fmt.Fprintf(w,"text from database")
})
http.ListenAndServe(":8080",nil)
}
I could connect to database at app start. Whenever I need to use the database I Ping it and if it's closed I reconnect to it. If it's not closed I continue and use it.
package main
import (
"database/sql"
_ "github.com/go-sql-driver/mysql"
"net/http"
"fmt"
"sync"
)
var db *sql.DB
var mutex sync.RWMutex
func GetDb() *sql.DB {
mutex.Lock()
defer mutex.Unlock()
err := db.Ping()
if err != nil {
db, err = sql.Open("mysql","dsn...")
if err != nil {
panic(err)
}
}
return db
}
func main() {
var err error
db, err = sql.Open("mysql","dsn....")
if err != nil {
panic(err)
}
http.HandleFunc("/",func(w http.ResponseWriter, r *http.Request) {
row := GetDb().QueryRow("select...")
// scan row
fmt.Fprintf(w,"text from database")
})
http.ListenAndServe(":8080",nil)
}
Which of these ways are the best or is there another way which is better. Is it a bad idea to have multiple request use the same database connection?
It's unlikly I will create an app that runs into mysql connection limit, but I don't want to ignore the fact that there's a limit.