getting the current date/time

>
> $ perl -MTime::HiRes=gettimeofday -e '$time=gettimeofday();print "$time\n"'
> 1171908671.04656
>

That's all very well if I was writing a perl program and the machine
is local, but I'm writing a Java program and the machines are remote.
As I already have a JDBC connection to each machine I'm interested in,
SQL is the easiest way for me to get the time on those machines.

donalmurtagh@yahoo.co.uk wrote:
>> $ perl -MTime::HiRes=gettimeofday -e '$time=gettimeofday();print "$time\n"'
>> 1171908671.04656
>>
>
> That's all very well if I was writing a perl program and the machine
> is local, but I'm writing a Java program and the machines are remote.
> As I already have a JDBC connection to each machine I'm interested in,
> SQL is the easiest way for me to get the time on those machines.
>

And definitely the wrong way. Since you're using Java, set up a jsp to
return it, or have a java object to run on the machine - or any of
several other ways of doing it.

But MySQL won't give you the detail you need, and it's like using a
hammer to drive in a screw.

--
==================
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
jstucklex@attglobal.net
==================

Why use sarcasm to answer a question -- why not just say the result
would always be 0? And why use profanity? I observed this newsgroup
was dead -- then realized it was a weekend. So I (being a novice)
threw out an answer I thought might help.

An observation is a complaint only to the tender skinned or guilty.



On Feb 18, 4:14 pm, Rik <luiheidsgoe...@hotmail.com> wrote:
> On Sun, 18 Feb 2007 19:59:16 +0100, <christop...@dailycrossword.com> wrote:
> > On Feb 18, 10:17 am, Rik <luiheidsgoe...@hotmail.com> wrote:
> >> On Sun, 18 Feb 2007 18:21:32 +0100, <christop...@dailycrossword.com>
> >> wrote:
>
> >> > in MySQL there is a DATE_FORMAT function which allows you to specify
> >> > the output of a date.
>
> >> But that still won't give you the _current_ microseconds...
>
> > DATE_FORMAT(date,format)
> > ...
> > %f Microseconds (000000..999999)
>
> > what do you mean by current? Oh, you are making a joke. Confusing,
> > since I just spent 20 minutes trying to figure out what you said I was
> > doing wrong.
>
> I'm not joking, the question was not to get the microsconds form _a_ date
> (which is hidiously easy with date_format), but a timestamp _with_ the
> microseconds of the exact time it's called:
>
> <donalmurt...@yahoo.co.uk> wrote:
> > Is there any way that I can (with SQL) retrieve the current date/time
> > that shows fractions of a second (milliseconds, microseconds, etc.).
> > If I execute:
>
> > select now()
>
> > It returns the current date/time, but only to the nearest second. I
> > would like to get a more precise measure of the current date/time and
> > if I could get it in the following format, that would be even better:
>
> > yyyy-mm-dd hh:mm:ss.ffff
>
> DATE_FORMAT(NOW(),'%f') will always result in 000000
>
> So, in short, before you complain about the quality of a newsgroup don't
> talk out of your ass.
> --
> Rik Wasmus

<christopher@dailycrossword.com> wrote:
> Rik <luiheidsgoe...@hotmail.com> wrote:
>> >> > in MySQL there is a DATE_FORMAT function which allows you to
>> specify
>> >> > the output of a date.
>>
>> >> But that still won't give you the _current_ microseconds...
>>
>> > what do you mean by current? Oh, you are making a joke. Confusing,
>> > since I just spent 20 minutes trying to figure out what you said I was
>> > doing wrong.
>>
>> I'm not joking, the question was not to get the microsconds form _a_
>> date
>> (which is hidiously easy with date_format), but a timestamp _with_ the
>> microseconds of the exact time it's called:
>> DATE_FORMAT(NOW(),'%f') will always result in 000000
>> So, in short, before you complain about the quality of a newsgroup don't
>> talk out of your ass.
>Why use sarcasm to answer a question -- why not just say the result
> would always be 0? And why use profanity? I observed this newsgroup
> was dead -- then realized it was a weekend. So I (being a novice)
> threw out an answer I thought might help.
>
> An observation is a complaint only to the tender skinned or guilty.

I'm sorry, a was a bit peeved by:
a. A wrong answer.
b. The fact that I told you it was a wrong answer was considered a joke.
c. (mostly) about being declared dead.

Observations are not always true, but forgive my rudeness.
--
Rik Wasmus

>
> And definitely the wrong way. Since you're using Java, set up a jsp to
> return it, or have a java object to run on the machine - or any of
> several other ways of doing it.
>
> But MySQL won't give you the detail you need, and it's like using a
> hammer to drive in a screw.
>

I agree that using SQL isn't necessarily the best way, but a JSP?
You want me to run a webserver just so I can get the current time on a
box?
Maybe you assumed this is a web app, but it's not. I'm running a J2SE
process on a box that needs to detect the time on other boxes.
Ideally, I would use RMI, but that's *a lot* of overhead just to get
the system time.

> Wouldn't it be FAR easier just to have the java apps open an independant
> connection between themselves and use the java timestamp, which DOES go
> to milliseconds, to detect if they're out of sync?
>

If there was already a Java process running on each box I'm interested
in then I could use sockets or RMI to pass the timestamps around.
However, the actual scenario is one java (J2SE) process running on a
box which needs to get the time on other remote boxes.

donalmurtagh@yahoo.co.uk wrote:
>> And definitely the wrong way. Since you're using Java, set up a jsp to
>> return it, or have a java object to run on the machine - or any of
>> several other ways of doing it.
>>
>> But MySQL won't give you the detail you need, and it's like using a
>> hammer to drive in a screw.
>>
>
> I agree that using SQL isn't necessarily the best way, but a JSP?
> You want me to run a webserver just so I can get the current time on a
> box?
> Maybe you assumed this is a web app, but it's not. I'm running a J2SE
> process on a box that needs to detect the time on other boxes.
> Ideally, I would use RMI, but that's *a lot* of overhead just to get
> the system time.
>

Ok, so you're not using a webserver there. Fine. But you think RMI is
more overhead than invoking a MySQL query? I doubt RMI for something
this basic would be that much overhead.

--
==================
Remove the "x" from my email address
Jerry Stuckle
JDS Computer Training Corp.
jstucklex@attglobal.net
==================

>
> Ok, so you're not using a webserver there. Fine. But you think RMI is
> more overhead than invoking a MySQL query? I doubt RMI for something
> this basic would be that much overhead.
>

In terms of runtime performance, probably not, but in terms of effort
and architecture, introducing RMI into a system that doesn't already
use it, just for the sake of getting the time seems like overkill. The
reason I want to get the time via SQL is because I know there will
already be a MySql server running on the machine(s) I'm interested in.