Good idea!
I had a similar situation, I was wanting to check the reason of a proto error, and the time that it occured, and I got around it by setting the alert to exec a batch file. The batch file called an .htc file. That way I could check anything that I wanted about the status of a thread.
( I am on windows 2000)
my .bat has:
DownProtError to_Output “Open failed” 30
(.htc file name) (thread) (error string) (time)
and my .htc has:
proc main {argv argc} {
set threadName [lindex $argv 0]
set errorString [lindex $argv 1]
set recently [lindex $argv 2]
catch {msiAttach} emsg
set nowtime [clock seconds]
msiGetStatSample $threadName var
keylget var PLASTERROR errortime
keylget var LASTUPDATE updatetime
keylget var PLASTERRTEXT ErrorText
if {[regexp $errorString $ErrorText]} {
if {[expr $nowtime – $errortime] < $recently} {
catch {exec cmd.exe /c hcicmd -p testP -c “to_Output pstop”} result
}}
}
main $argv $argc