Closed ashr123 closed 2 years ago
Are you using the Docker image?
Are you using the Docker image?
No, it runs directly on the OS
What do you see in "Long-term data / Graphics", when you select "Today"?
Note: if nothing appears, select "Last 7 days".
What do you see in "Long-term data / Graphics", when you select "Today"?
Note: if nothing appears, select "Last 7 days".
@rdwebdesign
Do you see any errors in the "developer tools / console" window (usually F12 key)?
You can try other simple tests:
Do you see any errors in the "developer tools / console" window (usually F12 key)?
You can try other simple tests:
- access the dashboard using a different browser or a different device;
- clear your browser cache and try again;
- disable all browser plugins/extensions and try again.
@rdwebdesign When I'm looking at Developer Tools I see the following error:
Unchecked runtime.lastError: Could not establish connection. Receiving end does not exist. index.php:1
When I open the page in Safari (I have MacBook), I see the same effect (all previous images are from Chrome):
I have a similar problem that might be related. I've noticed it yesterday after a php package update but I'm not sure if it was there before.
Doesn't matter which browser I'm using. No errors in web dev console or php-fpm log.
If I try to login in it's working, but I can see an error 500 for an ajax request to /admin/api_db.php?status
without any content. But I'm still able to login.
In addition, I can see the following error on settings.php.
I've already tried the repair option and cleared my cache and cookies.
Error 500 and no server logs doesn't really fit together, unless server logs have been disabled intentionally. Could you also check /var/log/lighttpd
?
Ok, that was quite a bad false positive relation I guess :D
I took a look at nginx erro.log (I'm using nginx instead of lighttp) and saw, that there were problems connecting to the sqllite database.
PHP message: PHP Fatal error: Uncaught Error: Undefined constant "SQLITE3_OPEN_READONLY" in /var/www/html/admin/scripts/pi-hole/php/database.php:61
And I also noticed, that pihole is running on php 8.0 and not 7.4 as I falsely remembered
thrown in /var/www/html/admin/scripts/pi-hole/php/database.php on line 61" while reading response header from upstream, client: 192.168.0.10, server: _, request: "GET /admin/api_db.php?status HTTP/1.1", upstream: "fastcgi://unix:/run/php/php8.0-fpm.sock:", host: "x.x.x.x:8080", referrer: "http://x.x.x.x:8080/admin/db_graph.php"
So I was also remembering that aptitude removed the php8.0-sqllite package yesterday...
Switched to php8.1 with sqllite package and everything's working fine again.
@s644 Your problem is unrelated (I'm glad you got it resolved).
@ashr123
Unchecked runtime.lastError: Could not establish connection. Receiving end does not exist. index.php:1
Have you tried disable ALL browser plugins/extensions? This error message doesn't say much, but sometimes this is related to browser extensions.
@ashr123
Unchecked runtime.lastError: Could not establish connection. Receiving end does not exist. index.php:1
Have you tried disable ALL browser plugins/extensions?
This error message doesn't say much, but sometimes this is related to browser extensions.
@rdwebdesign I'll try it, but I must say it will be weird if it is the problem.. cause I didn't change anything lately. I'll also try open it in Firefox, directly from the raspberry pi, if I remember correctly, I didn't install any plugin there
I believe you, but no one else is seeing/reporting this error.
Please, try it. If you provide more information we may find a solution.
I believe you, but no one else is seeing/reporting this error.
Please, try it. If you provide more information we may find a solution.
It's also happening in Firefox my Raspberry Pi :-/..
And also in My iPhone's Safari:
My PHP version is 8.0.8 by the way
Can you check in /var/log/pihole-FTL.log
, see if anything looks suspicious in there
@ashr123
Just a question: Did you restarted your pi-hole after the update?
@ashr123
Just a question:
Did you restarted your pi-hole after the update?
I think so.. 4 days ago
I think I've reproduced this on a Ubuntu 21.04 droplet... But I don't yet know the cause. Will try to investigate a bit more closely later on after work
Tell a lie... it has now started showing something on the graph. Which I was sort of hoping would be the case (logic being that it is a new install)
Error 500 and no server logs doesn't really fit together, unless server logs have been disabled intentionally. Could you also check
/var/log/lighttpd
?
@DL6ER if you meant to me, this is what I got from /var/log/lighttpd/error.log
:
2022-01-23 00:00:13: server.c.1787) logfiles cycled UID = 0 PID = 150890 2022-01-25 13:28:59: server.c.1976) server stopped by UID = 0 PID = 1 2022-01-10 06:08:08: server.c.1513) server started (lighttpd/1.4.59)
From /var/log/lighttpd/error.log.1
:
2022-01-16 00:00:37: server.c.1787) logfiles cycled UID = 0 PID = 285426 2022-01-17 00:00:38: server.c.1787) logfiles cycled UID = 0 PID = 330143 2022-01-19 16:55:04: server.c.1976) server stopped by UID = 0 PID = 1 2022-01-10 06:08:06: server.c.1513) server started (lighttpd/1.4.59)
Hope it helps
@ashr123 Can you browse to http://46.101.89.132/admin/index.php
please and tell me what you see on the dashboard there?
@ashr123 Can you browse to
http://46.101.89.132/admin/index.php
please and tell me what you see on the dashboard there?
@PromoFaux
Thanks. OK, that confirms that it is not your browser, at least... But I am at a loss as to what is going on on your install. Your debug log looks fine, I can't see anything that would make me think it was broken in there!
Please can I see more of the log at /var/log/pihole-FTL.log
(you can deliver it securely via cat /var/log/pihole-FTL.log | pihole tricorder
if you like)
@PromoFaux I don't see anything security issue with the output, so I'll just put it here:
[2022-01-28 03:59:39.432 1455M] Resizing "FTL-dns-cache" from 176128 to (11264 16) == 180224 (/dev/shm: 4.1MB used, 4.1GB total, FTL uses 4.1MB) [2022-01-28 11:06:56.002 1455M] Resizing "FTL-dns-cache" from 180224 to (11520 16) == 184320 (/dev/shm: 4.1MB used, 4.1GB total, FTL uses 4.1MB)
I might be barking up the wrong tree.
Do you get any output if you browse to:
http://pi.hole/admin/api.php?overTimeData10mins
http://pi.hole/admin/api.php?overTimeDataClients&getClientNames
I might be barking up the wrong tree.
Do you get any output if you browse to:
http://pi.hole/admin/api.php?overTimeData10mins
http://pi.hole/admin/api.php?overTimeDataClients&getClientNames
@PromoFaux
{"domains_over_time":{"1641790500":25346},"ads_over_time":{"1641790500":1181}}
[]
And on the command line, what is the output of:
echo ">ClientsoverTime" | nc 127.0.0.1 4711
(A small request: please don't quote the whole message you are replying to - it just causes clutter.. you also don't need to @ tag people on each response, we see them all 🙂)
I'm getting
1641790500 5590 8418 927 5964 1654 130 838 3468 535 1055 102 ---EOM---
And nc
doesn't exit normally, I need to press ⌃C for it to exit (maybe it is OK..)
^C is normal to exit, are you only seeing one line - or more?
Just one line with numbers?
yep, just it
I notice one strange thing:
The out put of your last command is showing data from "Monday, 10 January 2022 04:55:00" timestamp = 1641790500
.
At the same time, the logs you post are showing records apparently out of order:
2022-01-16 00:00:37: server.c.1787) logfiles cycled UID = 0 PID = 285426
2022-01-17 00:00:38: server.c.1787) logfiles cycled UID = 0 PID = 330143
2022-01-19 16:55:04: server.c.1976) server stopped by UID = 0 PID = 1
2022-01-10 06:08:06: server.c.1513) server started (lighttpd/1.4.59)
The last line is from 2022-01-10, but the previous one is from the 2022-01-19.
@ashr123 What is the current time of your system?
This is probably because weird bug: every time I restart the 𝜋, the clock resets to 10.1 for some reason.. to fix it I temporarily replace the DNS of the connection from the static IP of the 𝜋 to Cloudflare's until the OS pulls the right date and time
I'm just guessing, but maybe this "other weird bug" is changing the current time of your system and the records are getting lost/mixed/scrambled.
I never saw this behavior before (time resetting to a specific date). Do you have any idea what is causing it?
I wish I knew.... I don't have a clue, I even don't know how to tart to figure it out
Raspberry Pis don't have a Real Time Clock (RTC) hardware.
The time is always reset at start up. Usually it resets to the last recorded time before shutdown or to 1970-1-1 (timestamp=0), depending on your system's config.
Take a look at: https://forums.raspberrypi.com/viewtopic.php?t=254744
Do you use fake-hwclock
program? or do you have any hat with RTC hardware?
fake-hwclock
wasn't present on my raspberry pi, so I've installed it now.
I don't know what went wrong, but when I've installed my OS (Ubuntu 21.04, 64 bit) everything was OK, I do not remember when or why the phenomenon started to appear...
I'm not sure this will fix the issue (I don't even know if Ubuntu uses fake-hwclock
).
I'm just trying to indicate a possible solution to the "time reset" problem.
Yes, I know 🙂 and I appreciate it
Hi,
I was also having the exact same issue, with the same versions. I was going to open a ticket but found out this one.
The issue was that my Raspberry Pi was in the wrong timezone. Because of an improper shutdown, the timezone was set to UTC instead of Asia/Kolkata.
Changed the time timezone and restated the Pi-Hole and the DNS resolver from setting and the graphs showed up.
Maybe in the next update you could show the time the Pi-hole is running in the status.
@sahiluchil with the advice @rdwebdesign gave me (about fake-hwclock
) it seems that it solved my problem!
Now my Raspberry Pi starts correctly.
@rdwebdesign, maybe there is a bug in the way Long-term data > Graphics
and the main display displays the last 24 hours, I would recommend that the main graphics will use the same query as the Long-term data > Graphics
does (just with hard coded query)
@ashr123 Is your graphic problem solved?
Yes, thank you all for the help and the support!
I solved this issue by removing logs from /var/log/, I don't know but I faced the issue many times when freshly installed pi-hole on my raspberry pi. I hope it helps :)
sudo rm -rf /var/log/pihole-FTL sudo rm -rf /var/log/pihole.log sudo systemctl restart pihole-FTL
Thanks, in my case I have pi-hole running on a Raspberry Pi on Raspbian, and was having the same issue. It was the time that was set to the wrong time-zone (just 1 hour behind my actual time), so I changed the time-zone (timedatectl list-timezones
and then sudo timedatectl set-timezone <time zone>
) and after a restart of the Pi (not before the restart), the charts on the dashboard page are now showing properly. Thanks!
Thank you everyone! This issue happened to me as well. Perfect fix was for me the TimeZone settings. Somehow they got reset, and once I set to correct TZ, rebooted, the graphs for 'Total Queries over last 24 hours' appeared. Great fix, thank you! My environment on 9/18/2023 RaspberryPi 4, running Version 11 - Bullseye PiHole v5.17.1 FTL v5.23 Web Interface v5.20.1
Versions
Platform
Expected behavior
See the graph as usual
Actual behavior / bug
Can be seen in the attached photo
Steps to reproduce
Steps to reproduce the behavior:
Debug Token
Screenshots
Additional context
It may look like I've just installed the app, but it isn't the case... it is running for months