Bonkers since V 6.x

Help Community Help
1

The issue I am facing:

Pihole seems lost. It will not load in a reasonable amount of time, does not consistently respond on Web ports via any browser over LAN or Chromium on the Pi box.

Does not respond to DNS queries consistently.

Edit: I see SWAP is 100% of 100 MB.

Details about my system:

Pi 3
Linux raspberrypi 5.15.84-v8+ #1613 SMP PREEMPT Thu Jan 5 12:03:08 GMT 2023 aarch64

What I have changed since installing Pi-hole:

Did pihole -up to today's V6.x, accepted removal of lighttpd but it was still there, so removed manually per FAQ.

2

Please upload a debug log and post just the token URL that is generated after the log is uploaded by running the following command from the Pi-hole host terminal:

pihole -d

or if you run your Pi-hole as a Docker container:

docker exec -it <pihole-container-name-or-id> pihole -d

where you substitute <pihole-container-name-or-id> as required.

3

uploaded.

Token is https://tricorder.pi-hole.net/hKCYXhTB/

4

From your debug log:

*** [ DIAGNOSING ]: Pi-hole diagnosis messages
   count   last timestamp       type                  message 
   ------  -------------------  --------------------  ------------------------------------------------------------  
   1       2025-04-11 20:17:20  DNSMASQ_WARN          interface eth0 does not currently exist

Likely unrelated, but you have added as an adlist a file that is not an adlist (id 2).

*** [ DIAGNOSING ]: Adlists
   id     enabled  group_ids     address                                                                                               date_added           date_modified        comment                                           
   -----  -------  ------------  ----------------------------------------------------------------------------------------------------  -------------------  -------------------  --------------------------------------------------
   1            1  0             https://raw.githubusercontent.com/StevenBlack/hosts/master/hosts                                      2022-06-21 21:45:17  2022-06-21 21:45:17  Migrated from /etc/pihole/adlists.list            
   2            1  0             /etc/pihole/adlists.list                                                                              2023-04-19 18:22:01  2023-04-19 18:22:01
5

I noticed the eth0 thing, as it scrolled on by, but, it is working as I just connected via the web interface. However, the dashboard is just "churning".

I don't currently have a clue as to how to proceed. It all worked fine before the upgrade.

Don't recall ever adding an adlist. I presume it is the second entry that is bogus?

When the web interface responds I will look at it.

6

Yes.

7

Still having the issue. Below is what I see when Pihole stops responding. Just a minute ago saw a fully populated and active screen with eth0 being the port in use.

image
image1318×715 40 KB

8

I have doubled swap size rebooted. Takes quite a while for the web function to come up. Still seem to be same issues.

Ran another diagnostic, with this token: https://tricorder.pi-hole.net/gSIXde2T/

Thanks for any assistance.

10

Since you have a relatively huge query database (5.7 GB), let's make a new one and save the old one for posterity. This should speed things up.

sudo service pihole-FTL stop

sudo mv /etc/pihole/pihole-FTL.db /etc/pihole/pihole-FTL-old.db

sudo service pihole-FTL start
11

That does seem to have improved things. In my efforts I built a new Pi-hole raspberry. Transporter worked well.

Both showed great improvement with the new DB. How long should I keep the old one? It's of no consequence now, it is?

12

If you don't want to look at any historical query data, delete the old database.

13

Would it be likely there is a built in way to periodically start a new log and archive logs?

Log maintenance is not something that crossed my mind in this context. Is it in the docs?

14

With V6 we reduced the retention duration of the historic data from 365 days to 90 days. But this might still be to much for low-end hardware receiving a lot of queries. You can manually reduce the retention duration even further.

15

For the most part, that is already handled automatically.

Log files under /var/log/pihole are rotated regularly.

The long-term database retention period can be configured via database.maxDBdays, and Pi-hole will automatically mark entries older than database.maxDBdays as deleted.

However, that would not directly impact the size of the database file.

The database engine would grow file size if it cannot hold additional entries, but it would reuse space from deleted entries rather than shrink the file. Typically, that means that database size would stabilise around the size large enough to hold your queries for database.maxDBdays.

As yubiuser mentioned, v6 reduced that to a new 91 days default, from 365 previously.

This would have meant that Pi-hole's database maintenance would have been busy marking 274 days of DNS queries as deleted upon first starting Pi-hole v6, which may have kept your CPU quite busy, especially with larger long-term databases on low power hardware.

But the database file still would be as large as before.

If you'd wanted to reduce it in size, you could create a clean copy of Pi-hole's long-term query database to a new file with the following commands:

sudo pihole-FTL sqlite3 /etc/pihole/pihole-FTL.db "VACUUM INTO '/path/to/pihole-FTL.backup.db'"

You'd then have to sudo systemctl stop pihole-FTL and move the /path/to/pihole-FTL.backup.db over the original /etc/pihole/pihole-FTL.db before you start Pi-hole again.