setroot and setsite in a tcl proc for windows

[Jason Garrant]

Hi,

I am looking for some information on how to change the site in a tcl proc.  basically I have tried multiple incarnations of the the code below:

catch {exec cmd /c call setroot $root $site} execResponse

echo “Response: $execResponse”

I have tried using -clear, before this, also tried setsite as well, and setsite with -clear.  All come back with no response.  When I do showroot it always comes back with the site I am running the proc from.  

I am trying to do this, because I am setting up a uPoC protocol thread, that calls this proc, to cycle and delete SMAT and Log files.  I need to change the site because the following lines are only working for the site the thread is currently running in:

catch { exec — hcicmdnt -p $process -c “. output_cycle” } execResponse

catch { exec — hcicmdnt -p $processName -c “$threadName save_cycle in” } execResponse

catch { exec — hcicmdnt -p $processName -c “$threadName save_cycle out” } execResponse

Thanks in advance,

Jay

Topic

[Max Drown (Infor)]

You need to use eval. Here’s an example.

Code:

global HciRoot
set debug 0
set hciSite [lindex $argv 0]
set date [clock format [clock seconds] -format %Y%m%d%H%M%S]
set archDir /data/dbdump/error

# Set up the environment for Cloverleaf commands
eval [exec $HciRoot/sbin/hcisetenv -root tcl $HciRoot $hciSite]

# Get the number of lines in the error database. Ignore the header and footer (8 lines).
catch { exec hcidbdump -e -U [random 999999] | wc -l } numLines
set numLines [expr $numLines – 8]

-- Max Drown (Infor)

[Jason Garrant]

Hi Max,

It’s saying there is no such file or directory.  I followed the path, and it is there, but it is still coming back with the error.

[Max Drown (Infor)]

Do you have the script header in place? Something like “#!/hci/quovadx/qdx5.6/integrator/bin/tcl”?

-- Max Drown (Infor)

[Max Drown (Infor)]

This is how we call that particular script from cron:

Code:

03 01 * * * . ~/.profile.cron; /hci/quovadx/qdx5.6/integrator/sitebni/iig/bin/phs2idx_orderNum_purge.tcl

Here’s ~/.profile.cron

Code:

export HOST=bhciigp
export CL_INSTALL_DIR=/hci/quovadx/qdx5.6
export HOSTNAME=bhciigp.boone.org

eval `${CL_INSTALL_DIR}/integrator/sbin/hcisetenv -root ksh $*`

-- Max Drown (Infor)

[Mark Thompson]

Hi Max,

Running 6.1.2 on AIX.  Attempting to change site from met_wrk to s1_wrk:

Code:

$ hcitcl
hcitcl>echo $HciSite
met_wrk
hcitcl>eval [exec $HciRoot/sbin/hcisetenv -root tcl $HciRoot s1_wrk]
/hci/wrk/cis6.1/integrator/s1_wrk/exec/databases/rdm.taf
hcitcl>echo $HciSite
met_wrk

Doesn’t appear to change the site.

- Mark Thompson
HealthPartners

[Keith McLeod]

You may want to try some of this code to help change sites within tcl .

if  ![catch {netcfgLoad} err] {

      NCFG:ChangeSite $sitedir

      netcfgLoad

}

[Jim Kosloskey]

Here is what I have used:

set ret_cd [catch {eval [exec ${::env(HCIROOT)}/sbin/hcisetenv -site tcl $sitename]} err]

   if {$ret_cd} {

       echo “$module Switch to site >$sitename< failed…"

       echo “$module with error >$err<. Aborting!"

       return 0

   }

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Charlie Bursell]

Jim, Max

I used to do it this way but it does not work anymore.  I complained about it before I left.

I think the best method now is to use the Cloverleaf library call like stated by Keith.

[Donna Bailey]

Just getting a new script running on windows for cycle save(from fellow clovertecher)….I have a section with:

global env

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

#  Pull site & root from environment  #

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

   set WrkDir [setHciDirs]

   set site $env(HCISITE)

   set rootdir $env(HCIROOT)

echo “HCISITE=$site  HCIROOT=$rootdir”

Not sure if that helps..this tcl is being called from a bat script…

Donna

Donna Bailey
Tele: 315-729-3805
dbailey@microstar.health
Micro Star Inc.

[Jim Kosloskey]

Thanks Charlie.

I checked the release notes for 6.1 through 6.2 and did not see any mention regarding this being changed.

So in my opinion this is a bug.  Especially since it affects code already in place at some clients.

I would report it to Infor but I do not have access to Xtreme.

Anyone else who agrees this is a bug who does have access to Xtreme might report it as such.

At least to see what Infor’s explanation as to why this no longer functions.

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Robert Kersemakers]

A colleague of mine dug into this when we first migrated from CL5.5 to CL6.0. We now have these lines at the beginning of scripts called by cron:

Code:

# load function for execution in cron
FPATH=${CL_INSTALL_DIR}/integrator/kshlib; export FPATH
       . $FPATH/setroot
       . $FPATH/setsite

On further inspection: we execute a script first when calling from cron:

Code:

55 23 * * * /home/hci/sethcienv daily_housekeeping_cloverleaf > /dev/null 2>&1

The ‘sethcienv’ checks whether the cluster is running on that server; if so then the linux-profile and CL-profile are executed:

Code:

# Controleren of /prod/cloverleaf/sites60 bestaat
if [[ -d /prod/cloverleaf/sites60 ]]
then
  . /etc/profile
  . /home/hci/.profile
 # script uitvoeren
 echo “+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++”
 $*
fi
exit 0

This may help some people here.

Zuyderland Medisch Centrum; Heerlen/Sittard; The Netherlands

[Steve Williams]

I’m converting an old AIX script to TCL (portable to the Windows Server platform) on Clvf v6.1. I tested both Jim’s posted code and Keith’s posted code and here’s what I found on my WinServer 2012 environment.

The older code now generates a file not found error. I played around with it but could not get it to work. I can run it directly from the cmdline using the full path name, but from within the hcitcl environemnt it’s lost somehow.

The newer code posted by Keith, using NCFG:ChangeSite, worked right out of the box.

Disclaimer: Your mileage may vary…

[Author]

Replies