get the wrong mistake number when query time is short
35 views
Skip to first unread message
何吉林
unread,
Jul 31, 2020, 12:22:06 AM7/31/20
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to Prometheus Users
suppose I monitor a job with 100 instance.when I query with 1 minute :
count(sum(metrics_name{xxx}[1m]) by (instance)),
what i get is less than 100.
when I use 10 minutes,
count(sum(metrics_name{xxx}[10m]) by (instance)),
what i get is 100.
Brian Candler
unread,
Jul 31, 2020, 3:10:47 AM7/31/20
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to Prometheus Users
That query cannot even be run - metrics_name{xxx}[1m] is a range vector, but sum() only works over an instant vector.
What's the real query you're doing?
何吉林
unread,
Jul 31, 2020, 3:19:36 AM7/31/20
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to Prometheus Users
sorry, the increase function was lost, the whole query should be
count(sum(increase(metric_name{xxx}[1m])) by (instance) )
在 2020年7月31日星期五 UTC+8下午3:10:47,Brian Candler写道:
Brian Candler
unread,
Jul 31, 2020, 6:34:43 AM7/31/20
Reply to author
Sign in to reply to author
Forward
Sign in to forward
Delete
You do not have permission to delete messages in this group
Copy link
Report message
Show original message
Either email addresses are anonymous for this group or you need the view member email addresses permission to view the original message
to Prometheus Users
What's your scrape interval?
increase(metric_name{xxx}[1m])
requires at least two data points in the 1m window. So if you are scraping at 1m intervals, you need at least [2m] to guarantee two data points in the window at any time.