index=source earliest=-2h sourcetype=e | bucket _time span=1h |stats count by code _time| delta count as difference | eval percdif=round(abs(difference/count)*100,0)|table code, count, difference, percdif|sort -percdif -count|where count>1100|fields code|head 10
I am using this query. output basically will return error codes sorted with high percentage difference (errors are increasing). I am using fields command to only output top 10 errors in table
since this query will show as stats table, how can I use this output to further pipe or do a timechart?
If your timechart query's timerange is different from last 2 hr, then try like this
index=source [search index=source earliest=-2h sourcetype=e | bucket _time span=1h |stats count by code _time| delta count as difference | eval percdif=round(abs(difference/count)*100,0)|table code, count, difference, percdif|sort -percdif -count|where count>1100|fields code|head 10]
| timechart count by code
It did not work, returns 0 results
How about this
index=source sourcetype=e [search index=source earliest=-2h sourcetype=e | bucket _time span=1h |stats count by code _time| delta count as difference | eval percdif=round(abs(difference/count)*100,0)|table code, count, difference, percdif|sort -percdif -count|where count>1100|fields code|head 10]
| timechart count by code
does this work if I pass more than one field (code and percdif) to timechart as well?
It'll. As long as that field is present in main search.
Thanks working now
What's your final expected output (that you want to generate for these top 10 error codes)?
Final output should be a timechart of the top 10 error codes (of query I posted above)