Graph empty but pihole works just fine

Help
1

Pihole v5.1.2
FTL v5.2
Web interface v5.1.1
I've been using pihole with unbound for quite a while now. Yesterday I wanted to add a new blocklist but the database was corrupted (Database disk image is malformed error (Pihole 5.1.1)). So I created a new gravity database. Since then pihole works again just fine but the graph is empty and I only see the current hour. https://i.imgur.com/FTYXTsO.png

Expected Behaviour:

see the graph as usual.

Actual Behaviour:

no updated graph even though pihole is working fine.
Things I tried to resolve the issue:

  • clear browser cache
  • check if date in terminal is the same as pihole

Debug Token:

aplvqdo3v3

Any idea on how to resolve the issue?
Thank you!

2

It looks like you have recently changed the time on the Pi (power failure?):

[2020-11-17 20:35:00.063 765M] WARN: getOverTimeID(1605641700): 1899 is too large: 1604502300

The time shown in this log appears off - you should not have Pi-hole logs from 5 November (they are only kept for 5 days).

*** [ DIAGNOSING ]: Pi-hole log
-rw-r--r-- 1 pihole pihole 21126036 Nov 17 20:35 /var/log/pihole.log
   -----head of pihole.log------
   Nov  5 00:00:17 dnsmasq[631]: query[A] ocsp.int-x3.letsencrypt.org from 192.168.178.63
   Nov  5 00:00:17 dnsmasq[631]: cached ocsp.int-x3.letsencrypt.org is <CNAME>
3

Thanks for looking into this and your quick response.
I did some more digging and it seems it's indeed a time issue as you suggested as logs at /var/log/pihole-FTL.log indicate it. I didn't experience any power failure or the like recently. My entire home network is also connected to a UPS, so I'm not sure what could have caused this issue. Anyway I checked a couple of threads here with a similar issue but for some reason have not been able to resolve it.

  1. I checked timedatectl and it looks fine.
  2. I checked sudo raspi-config and timezone is correct.
  3. When I check the logs at /var/log/pihole-FTL.log I receive the following:

[2020-11-18 10:35:44.659 764M] WARN: getOverTimeID(1605692100): 1983 is too large: 1604502300.

date -d @1605690300 shows Nov 4 16:05. How can that be?!
4. When I look at journalctl -u systemd-timesyncd.service the pi syncs time right after reboot:

Nov 18 10:08:40 raspberrypi systemd-timesyncd[300]: Synchronized to time server 193.145.15.15:123 (2.debian.pool.ntp.org).

So is pihole unable to get the correct time from the system or what is the problem here? How can I solve this?
Thank you

4

No, because it logs the correct time, see the beginning of the line:

So we're looking for a different error. Have you tried flushing the logs and restarting the whole system? Maybe the time having been off once (maybe it was a network glitch and the time was synced wrongly for some time and this was fixed by the system later?) and this will not happen again but something is really confused right now.

5

thanks for your reply Coro.
I flushed the logs and restarted the pi. Problem persists.
Looking at cat /var/log/pihole-FTL.log I get the following error again:

[2020-11-18 16:37:01.750 718M] WARN: getOverTimeID(1605713700): 2019 is too large: 1604502300

date -d @1604502300 gives me:

Wed Nov 4 16:05:00 CET 2020

which is completely off.

Edit:
It really seems that something is broken as all logs such as /var/log/syslog show that there is a gap between 11.05. and 11.18. There are no logs at all in between those dates. Could that mean that my SD card is broken? The raspberry and definitely pihole have been running since 11.05. without any issues.

6

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.