Long-term queries over time: Inconsistency

Dears,
after updating to the latest version (v 3.2.1) I found out that the reported ratio of blocked DNS requests is invalid as it's surpassing the 100% mark.

Can you share the relevant section of your log (/var/log/pihole.log) privately? How do I debug my Pi-hole installation?

Also, does the issue stay after a restart of FTL? sudo service pihole-FTL restart

@Mcat12: Thank you for your responses. I've noticed that the problem appears in the graphs created from: "Long term data" => "Graphics". but not from the normal graph in the main dashboard, even if they are covering the same period.
what I did was to make a long-term query covering the last 24 hours (the same as the main graph in the dashboard) and in comparison, the one is indeed reporting wrong numbers than the main graph.

Last 24 hours graph:

Long-term query graph covering the same period:

I've also generated a debug log but don't know how to share the debug token privatly with you?

The debug token is public, the information it points to is private.

1 Like

so sould I post it here so you can have a look?

here you go: 1eo24prqex

and thanks in dvance :slight_smile:

What is the output of these commands?

echo /var/log/pihole.log | pihole tricorder
echo /var/log/pihole.log.1 | pihole tricorder

first command: p51ji6x4ql
second command: yf2gsym9sn

Ah, sorry, I told you the wrong command. That command just printed /var/log/pihole.log and /var/log/pihole.log.1 to tricorder! Use these instead:

cat /var/log/pihole.log | pihole tricorder
cat /var/log/pihole.log.1 | pihole tricorder

No worries :slight_smile:
here they are:
first command: mfi3vako3b
second command: lcsajubmvx

Can reproduce locally, investigating...

The data was labeled incorrectly (and inconsistently compared to the dashboard). The green curve showed the permitted queries instead of all queries.

Previously:

Fixed:

Compare this with the dashboard graph:

1 Like