El 24/may/2006 a las 13:04 -0300, mikshaw me decía:
> According to the wmii-3 guide, section 6.5, removing the status label
> before running a new instance of status "ensures that the status label
> will not exist, hence all the writes made from any previously running
> status script will fail, thus they will finish." This works fine if
> you leave the sleep time at the default 1 second. However, as was
> suggested elsewhere I extended the sleep time to 20sec to reduce the
> system load a little. This results in the previous status scripts
> not finishing, because they are usually sleeping during the 2-second
> period that the status label is non-existent. By the time they wake
> up, the status label has been recreated and so they continue their
> loop. This results in multiple instances of status and multiple
> instances of sleep. After several runs of wmiirc the status bar is
> constantly being written, which negates the original purpose of
> extending the status sleep time.
This is not tested sop maybe i'm wrong, but you could divide your
status script in two (or more) scripts: one (or more) that provides
content, and the proper status script that just read the generated
content (maybe from files written by those content-generators), so
your status script could have a sleep of 1 second (or whatever
appropriate to avoid the problem you mention) and the content-provider
scripts could have whatever fits better.
For example, for the hour i don't care about the seconds, so i have my
clock-generator script with a sleep 60.
For some other informational process i even got more sleepiness.
-- Fernando Canizo - http://www.lugmen.org.ar/~conan/ TEACHER IS NOT A LEPER - BartntReceived on Thu May 25 2006 - 01:25:15 UTC
This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 16:06:47 UTC