You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
1 - I have in fact, at the execution time of the query 5 idle in transaction queries since the warning and critical are not passed on the query for filtering.
2 - nb_idle variable is incremated in the for loop for EACH (5 in total) idle in transaction queries before checking if c_limit or w_limit has been reached. But only one has reach the treshold.
Robin,
The text was updated successfully, but these errors were encountered:
In fact, the service seems to react as documented - but the documentation is not very clear in this aspect. The service reports the total number of idle transactions and the oldest one that reached one of the limit.
Maybe we could improve the reported message. I'm not kind to change the current behavior, because an upgrade to a latter release of check_pgactivity would have an impact for users using graphing tools.
Thanks for the report, I leave the issue open for now, I'm not against a change of the reported message.
Hello,
Today I experienced something strange, my monitoring told me I had 5 queries stuck in idle in transaction state but only had 1 on my production.
What happens here is :
1 - I have in fact, at the execution time of the query 5 idle in transaction queries since the warning and critical are not passed on the query for filtering.
2 -
nb_idle
variable is incremated in the for loop for EACH (5 in total) idle in transaction queries before checking ifc_limit
orw_limit
has been reached. But only one has reach the treshold.Robin,
The text was updated successfully, but these errors were encountered: