Paul,

After thinking about this more....I believe it is more likely to screw up if 
(at the same time) my script is running and another script / command is trying 
to access 1wire data.
 
Observations
a/  I have noticed missing data in my trend files if I have logged into the 
slug and have been reading data via the command prompt.  ie When I check 
calibrations of my devices I generally do a few manual "cat" commands.  The 
next day or so I see my trend files are missing some data around the same time 
as I did my calibrations.  The missing data is NOT the same device as I am 
calibrating.  My cat command could of been requested at the same time my script 
is running.

b/ The last time it stopped, i was away on business and I think it could of 
been the time I logged in via my web page and executed a cgi script.  My web 
front end shows trends however I also have a CGI file that just does a read of 
the 1wire uncached data and reports "instantaneous data".  (I ahve included cgi 
file below)

So I am thinking that when the 1wire file systems get hit with two read 
requests it is more likely to stop responding.

See the following CGI script output
http://img.photobucket.com/albums/v319/rjconway/web1.jpg


CGI FILE #####################################################################
#!/bin/bash
etime=$(/bin/date)

### Read 1wire data and convert into engineering units #ph was 2.6667##
PH=$(/opt/bin/awk '{printf "%4.2f\n",((($0-1)/1.6)+5.5)}' 
/tmp/1wire/20.B71F05000000/volt.A)
US=$(/opt/bin/awk '{printf "%5.1f\n",($0*100)}' 
/tmp/1wire/20.B71F05000000/volt.B)
TempAqua=$(/opt/bin/awk '{printf "%3.1f\n",((($0-1)*5)+20)}' 
/tmp/1wire/20.B71F05000000/volt.C)
LITRES=$(/opt/bin/awk '{printf "%3.1f\n",(($0-2.900)*60)}' 
/tmp/1wire/20.952A05000000/volt.D)
TempOut=$(/opt/bin/awk '{printf "%3.1f\n",($0)}' 
/tmp/1wire/28.11BF97000000/temperature)
TempAmb=$(/opt/bin/awk '{printf "%3.1f\n",($0)}' 
/tmp/1wire/28.D70598000000/temperature)
TempRes=$(/opt/bin/awk '{printf "%3.1f\n",($0)}' 
/tmp/1wire/28.A6EE97000000/temperature)

set -f
echo Content-type: text/plain
echo
echo "             DEMAND REPORT"
echo -------------------------------------------
echo Report Date  $etime
echo
echo
echo Aquaium Data
echo " pH =" $PH
echo " Conductivity =" $US uS
echo " Aquarium     = "$TempAqua " DegC"
echo " Reservoir Temperature = "$TempRes DegC
echo " Litres Available      = "$LITRES L
echo
echo Weather Data
echo " Outside Temperature =" $TempOut DegC
echo " House   Temperature =" $TempAmb DegC
echo
echo --------------------------------------------
echo "                 Page 1/1"
###############################################################



----- Original Message ----- 
  From: Paul Alfille 
  To: [email protected] 
  Sent: Monday, March 12, 2007 10:06 PM
  Subject: Re: [Owfs-developers] Reading 1 wire data every minute with anNSLU2 
(SLUG)


  I use an "inittab" script -- it gets automatically restarted. (And a delay 
loop).

  Are you running your script against owserver using owshell? Can you tell what 
the point of failure is? Is owserver still working after the "crash" -- can you 
access from another machine. It would be nice to find the underlying problem.

  Paul Alfille


  On 3/12/07, Rob Conway <[EMAIL PROTECTED]> wrote:
    I am using an NSLU2 (Unslung 6.8), I currently use a simple bash script to 
gather 8 off 1wire variables and do some calc's before storing the data in 
trend files.  I have it scheduled in crontab to run every 1 minute.  What 
happens is that about ~3 times a month the NSLU2 grinds to a halt.  It seems 
that my script somehow screws up and I just fill up the slug with scripts never 
finishing.  I think its because it cannot access the 1wire file system to read 
the data...

    Is this the best way to handle scheduling of scripts, using a 1 minute 
crontab entry ?
    I was thinking to just use a loop in a single script and just sleep for 60 
seconds...that way if it does screw up at least the NSLU2 will continue to run. 
 I could have another script testing a "PID entry" to make sure its still 
running and if not just restart it automatically.

    I know its slightly off topic however was keen to see if other NSLU2 -1wire 
users have implemented the same type of continuous monitoring and what their 
maximum uptime was.  The SLUG is lightly loaded and dedicated for data 
monitoring with owfs, NTP, thhtpd, and rrdtool....I want to move onto relay 
output control  thus trying to sort this problem out for reliability before I 
move into closed loop control.

    owfs version as seen from #ipkg list_installed
    owfs - 2.6p1-3 - 1-Wire filesystem (http://owfs.sf.net)
    owlib - 2.6p1-3 - 1-wire library
    fuse-utils - 2.6.0-pre2-3 - Fuse


    ################################################################
    #!/bin/bash
    ###  Force Simultaneous 1wire conversion ###
    echo 1 >/tmp/1wire/simultaneous/voltage
    echo 1 >/tmp/1wire/simultaneous/temperature

    ### Read 1wire data and convert into engineering units ##
    PH=$(/opt/bin/awk '{printf "%4.2f\n",((($0-1)/1.6)+5.5)}' 
/tmp/1wire/20.B71F05000000/volt.A)
    US=$(/opt/bin/awk '{printf "%5.1f\n",($0*100)}' 
/tmp/1wire/20.B71F05000000/volt.B)
    TempAqua=$(/opt/bin/awk '{printf "%3.1f\n",((($0-1)*5)+20)}' 
/tmp/1wire/20.B71F05000000/volt.C)
    LITRES=$(/opt/bin/awk '{printf "%3.1f\n",(($0-2.900)*60)}' 
/tmp/1wire/20.952A05000000/volt.D)
    TempOut=$(/opt/bin/awk '{printf "%3.1f\n",($0)}' 
/tmp/1wire/28.11BF97000000/temperature)
    TempAmb=$(/opt/bin/awk '{printf "%3.1f\n",($0)}' 
/tmp/1wire/28.D70598000000/temperature)
    TempRes=$(/opt/bin/awk '{printf "%3.1f\n",($0)}' 
/tmp/1wire/28.A6EE97000000/temperature)
    Accum_flow=$(/opt/bin/awk '{printf "%d\n",($0/4100)}' 
/tmp/1wire/uncached/1D.EBEF0900000034/counters.A)

    ### Calculations based upon Data ###
    C12=$(/opt/bin/awk -v PH="$PH" '{printf "%3.1f\n",((3*$0)*(10^(7-PH)))}' 
/public/kh_value)
    Todayflow=$(/opt/bin/awk -v AF="$Accum_flow" '{printf "%d\n",(AF-$0)}' 
/public/lastflow.txt)
    Weekflow=$(/opt/bin/awk -v AF="$Accum_flow" '{printf "%d\n",(AF-$0)}' 
/public/weekflow.txt)

    #################################################################

    cheers rob -Australia

    -------------------------------------------------------------------------
    Take Surveys. Earn Cash. Influence the Future of IT
    Join SourceForge.net's Techsay panel and you'll get the chance to share 
your 
    opinions on IT & business topics through brief surveys-and earn cash
    http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
    _______________________________________________
    Owfs-developers mailing list
    [email protected]
    https://lists.sourceforge.net/lists/listinfo/owfs-developers 






------------------------------------------------------------------------------


  -------------------------------------------------------------------------
  Take Surveys. Earn Cash. Influence the Future of IT
  Join SourceForge.net's Techsay panel and you'll get the chance to share your
  opinions on IT & business topics through brief surveys-and earn cash
  http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV


------------------------------------------------------------------------------


  _______________________________________________
  Owfs-developers mailing list
  [email protected]
  https://lists.sourceforge.net/lists/listinfo/owfs-developers
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Owfs-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/owfs-developers

Reply via email to