Hello
I use the search below in order to calculate a volume percentage
| inputlookup host.csv
| lookup lookup_patch "Computer" as host output FileName, StateName
| search StateName="Non-Compl"
| stats dc(host) as host by StateName FileName
| stats sum(host) as NbNonCompliantPatchesIndHost
| appendcols
[| inputlookup host.csv
| stats dc(host) as NbIndHost]
| eval Perc=round((NbNonCompliantPatchesIndHost/NbIndHost)*100,2)
| table Perc, NbIndHost, NbNonCompliantPatchesIndHost
No I need to calculate the volume percentage by SITE
So I add a clause "by SITE" in my stats command but it's wrong because sometimes the percentage is over 100% because NbNonCompliantPatchesIndHost is > NbIndHost
What is the solution to do this please?
| inputlookup host.csv
| lookup lookup_patch "Computer" as host output FileName, StateName
| lookup fo_all HOSTNAME as host output SITE
| search StateName="Non-Compl"
| stats dc(host) as host by StateName FileName SITE
| stats sum(host) as NbNonCompliantPatchesIndHost by SITE
| appendcols
[| inputlookup host.csv
| lookup fo_all HOSTNAME as host output SITE
| stats dc(host) as NbIndHost by SITE]
| eval Perc=round((NbNonCompliantPatchesIndHost/NbIndHost)*100,2)
| table SITE Perc, NbIndHost, NbNonCompliantPatchesIndHost
One factor to consider is how appendcols works.
In the first query, each subsearch returned a single result so appending one to the other worked well.
In the second query, each subsearch returns multiple results which may not be the same. If each list of sites is not identical, then appending one to the other will produce nonsense.
To prevent this, use append instead of appendcols then use stats to combine to two results.
| inputlookup host.csv
| lookup lookup_patch "Computer" as host output FileName, StateName
| lookup fo_all HOSTNAME as host output SITE
| search StateName="Non-Compl"
| stats dc(host) as host by StateName FileName SITE
| stats sum(host) as NbNonCompliantPatchesIndHost by SITE
| append
[| inputlookup host.csv
| lookup fo_all HOSTNAME as host output SITE
| stats dc(host) as NbIndHost by SITE]
| stats values(*) as * by SITE
| eval Perc=round((NbNonCompliantPatchesIndHost/NbIndHost)*100,2)
| table SITE Perc, NbIndHost, NbNonCompliantPatchesIndHost
Note that this may not solve the problem, but will eliminate one possible source of error.
One factor to consider is how appendcols works.
In the first query, each subsearch returned a single result so appending one to the other worked well.
In the second query, each subsearch returns multiple results which may not be the same. If each list of sites is not identical, then appending one to the other will produce nonsense.
To prevent this, use append instead of appendcols then use stats to combine to two results.
| inputlookup host.csv
| lookup lookup_patch "Computer" as host output FileName, StateName
| lookup fo_all HOSTNAME as host output SITE
| search StateName="Non-Compl"
| stats dc(host) as host by StateName FileName SITE
| stats sum(host) as NbNonCompliantPatchesIndHost by SITE
| append
[| inputlookup host.csv
| lookup fo_all HOSTNAME as host output SITE
| stats dc(host) as NbIndHost by SITE]
| stats values(*) as * by SITE
| eval Perc=round((NbNonCompliantPatchesIndHost/NbIndHost)*100,2)
| table SITE Perc, NbIndHost, NbNonCompliantPatchesIndHost
Note that this may not solve the problem, but will eliminate one possible source of error.
Thanks it works but I have another question
I filter the results of my search from a SITE dropdown list in order to display the volume percentage for a SITE
I added the token filters in the example below
But with your "stats values" command all the SITE are displayed. How to display only the choice done from the dropdown list please?
| inputlookup host.csv
| lookup lookup_patch "Computer" as host output FileName, StateName
| lookup fo_all HOSTNAME as host output SITE
| search StateName="Non-Compl" | search SITE=$tok_filtersite$
| stats dc(host) as host by StateName FileName SITE
| stats sum(host) as NbNonCompliantPatchesIndHost by SITE
| append
[| inputlookup host.csv
| lookup fo_all HOSTNAME as host output SITE
| search SITE=$tok_filtersite$
| stats dc(host) as NbIndHost by SITE]
| stats values(*) as * by SITE
| eval Perc=round((NbNonCompliantPatchesIndHost/NbIndHost)*100,2)
| table SITE Perc, NbIndHost, NbNonCompliantPatchesIndHost
perfect rich!
many thanks to you