With the caveat that it's not *actually* waking up right at the specified 
microsecond.
It's *at least* that number of microseconds, and then you have to add the 
effect of the round-robin timeslice scheme.

"Your wait time will be at least 84213 microseconds...."
"Your wait time will be at least 65432 microseconds...."

W. S. J.



-----Original Message-----
From: John Thompson <[email protected]>
To: U2 Users List <[email protected]>
Sent: Tue, Aug 28, 2012 9:51 am
Subject: Re: [U2] [ud] Sub-second delay?


Oh and I should have said usleep handles time in microseconds.
There is a million microseconds in a second.  I actually had to google that
:(

On Tue, Aug 28, 2012 at 12:49 PM, John Thompson <[email protected]>wrote:

> Coming late to the party because I've been writing documentation for the
> past few days...  which may cause me to go to sleep...
>
> Anyway, AIX and Linux should have a command called usleep I believe.
>
> Could you shell out and execute usleep?  Just a thought...
> (This shell out code is UV specific, not sure what it might be in Unidata)
>
> EXECUTE 'SH -c "usleep 200000"' ; *Sleep for 0.2 seconds"
>
> Of course, the execute command probably takes longer than 0.2 seconds...
>
> On the AIX Print spooler question.  Is it System V printing or AIX native
> printing?
>
> If its System V printing:
> I found this in a redbook (Ugh, redbooks...)
> http://www.redbooks.ibm.com/redbooks/pdfs/sg246018.pdf
>
> Notice they put FUN, in the title, sheesh...
>
> 8.3.8  Cleaning up and starting over
> There should not be a necessity to clear out all the queues and start
> over, but
> you might have a need for this if you clone a system.
> 8.3.8.1  Resetting the print subsystem job number
> The job number for the local jobs is stored in /spool/tmp/host-name/.SEQF
> and contains four fields separated by colons, as shown here:
> 1:999:1:598
> The first two numbers are the limits of the job numbers. The next number
> shows where this job sequence started, and the last number (598) is the
> number of the next print request. To change the sequence number, just
> change this last value, and recycle the print subsystem with lpshut and
> lpsched.
> 8.3.8.2  Cleaning out old print files
> If you are just having problems, you may want to start out by cleaning out
> the
> job request files. First, check to see the job names and cancel them, as
> shown here:
> # lpstat -o
> # cancel fileps-124
> If you are not able to cancel the files, then remove the individual files
> from the
> /spool/tmp/host-name directory and the /var/spool/lp/requests/host-name
> directory.
>
> Thats probably not what you want though.
>
> If its the AIX Print system, they probably do it differently.
>
> I didn't find much in here after poking around for a few minutes:
>
> http://publib.boulder.ibm.com/infocenter/pseries/v5r3/topic/com.ibm.aix.printergd/doc/printrgd/printrgd_pdf.pdf
>
> I did find that the maximum copies per page in the print job is 999, so it
> might make sense that the maximum number of total jobs is 999.
>
> Also keep in mind, that if you did find a way to increase that job number
> greater than 999, you would have to be wary of exceeding maximum number of
> inodes of the filesystem the jobs would be queued up on.
>
> You would be able to get that with a df command.  Just use the switch to
> show inodes I believe.
>
> Other than that, I'm afraid thats all I've got- if that helps.
>
>
>
> On Tue, Aug 28, 2012 at 11:51 AM, Wjhonson <[email protected]> wrote:
>
>>
>> Example code ?
>>
>>
>>
>> -----Original Message-----
>> From: Daniel McGrath <[email protected]>
>> To: U2 Users List <[email protected]>
>> Sent: Tue, Aug 28, 2012 1:57 am
>> Subject: Re: [U2] [ud] Sub-second delay?
>>
>>
>> I needed this once for a test program I was writing (not production code).
>>
>> To do it, I cheated a bit and used the socket API and used a blocking
>> command,
>> which did support timeouts with millisecond granularity. Works like a
>> charm; at
>> least in the cases when you don't have to worry about people connecting
>> to that
>> socket and affecting your programs timing :)
>>
>> Regards,
>> Dan
>>
>> -----Original Message-----
>> From: [email protected] [mailto:
>> [email protected]]
>> On Behalf Of Kevin King
>> Sent: Monday, August 27, 2012 9:46 AM
>> To: U2 Users List
>> Subject: [U2] [ud] Sub-second delay?
>>
>> Is there anything in Unidata (7.1, in particular) that can do a sub-second
>> delay, like maybe a half second?
>> _______________________________________________
>> U2-Users mailing list
>> [email protected]
>> http://listserver.u2ug.org/mailman/listinfo/u2-users
>> _______________________________________________
>> U2-Users mailing list
>> [email protected]
>> http://listserver.u2ug.org/mailman/listinfo/u2-users
>>
>>
>> _______________________________________________
>> U2-Users mailing list
>> [email protected]
>> http://listserver.u2ug.org/mailman/listinfo/u2-users
>>
>
>
>
> --
> John Thompson
>



-- 
John Thompson
_______________________________________________
U2-Users mailing list
[email protected]
http://listserver.u2ug.org/mailman/listinfo/u2-users

 
_______________________________________________
U2-Users mailing list
[email protected]
http://listserver.u2ug.org/mailman/listinfo/u2-users

Reply via email to