DNS Server Stops Responding

The issue I am facing:
Pi-hole stops responding every approx. 24 hours (not 100% sure on timing but I need to resolve it once per day in general. Rebooting the Debian box, or running pihole restartdns resolves the issue for a while.

Details about my system:
Pi-hole v5.1.2 Web Interface v5.1.1 FTL v5.2
Debug Upload https://tricorder.pi-hole.net/h2jep0a644 (issue was not present when this log was taken, will update with a new log when it next happens)
Debian 9 on Windows Server 2019 Virtual Machine

What I have changed since installing Pi-hole:
Installed DNSCrypt-Proxy for upstream DNS

I think it has to do with

SQLite3 message: database disk image is malformed in "SELECT VALUE FROM ftl WHERE id = 0;" (11)

You could try to recover the data, following this topic:

Note: usually this errors happend when either your SD Card is end of life and has errors or your power supply does not deliver enough power. You could check for the latter with

zgrep -ni voltage /var/log/syslog.*

Or tail the kernel ring buffer/log live for " under-voltage " messages while performing some tasks on the Pi:

dmesg -wT | grep -i voltage

2 Likes

Thanks for taking the time to respond! I've run the commands in the linked topic (the first few failed with a database disk image is malformed error similar to in the topic, but the second one to dump and rebuild the DB seems to have worked, and the reindex and vacuum commands no longer fail.

I'll leave it to run for a while and report back. The issue certainly won't be with voltage or a failing SD card, as the Pihole is running within a virtual machine on hard disks on a Windows Server, but we have had one or two power interruptions lately, so that's very possible.

Issue occurred again today unfortunately. Debug log is at https://tricorder.pi-hole.net/gnmqg0jprq

Please post the ouput of

cat /var/log/pihole-FTL.log

At the moment there's just one entry in there:
[2020-10-14 07:04:21.536 81053M] Resizing "/FTL-strings" from 180224 to 184320

Ok, and what's about

cat /var/log/pihole-FTL.log.1

Not a great deal more in there unfortunately:

[2020-10-14 07:04:21.536 81053M] Resizing "/FTL-strings" from 180224 to 184320
[2020-10-14 10:58:07.901 81053M] Resizing "/FTL-strings" from 184320 to 188416
[2020-10-14 10:58:08.447 81053M] Resizing "/FTL-dns-cache" from 114688 to 118784
[2020-10-14 15:07:36.183 81053M] Resizing "/FTL-strings" from 188416 to 192512
[2020-10-14 15:25:08.435 81053M] Resizing "/FTL-dns-cache" from 118784 to 122880
[2020-10-14 18:00:56.403 81053M] Resizing "/FTL-strings" from 192512 to 196608
[2020-10-14 18:22:08.265 81053M] Resizing "/FTL-dns-cache" from 122880 to 126976
[2020-10-14 18:42:09.654 81053M] Resizing "/FTL-strings" from 196608 to 200704
[2020-10-14 19:53:46.235 81053M] Resizing "/FTL-dns-cache" from 126976 to 131072
[2020-10-14 20:28:50.701 81053M] Resizing "/FTL-strings" from 200704 to 204800
[2020-10-14 20:55:47.620 81053M] Resizing "/FTL-dns-cache" from 131072 to 135168
[2020-10-14 21:16:40.577 81053M] Resizing "/FTL-dns-cache" from 135168 to 139264
[2020-10-14 21:37:57.035 81053M] Resizing "/FTL-strings" from 204800 to 208896
[2020-10-14 21:45:49.946 81053M] Resizing "/FTL-dns-cache" from 139264 to 143360
[2020-10-14 22:38:01.301 81053M] Resizing "/FTL-strings" from 208896 to 212992
[2020-10-14 23:20:31.800 81053M] Resizing "/FTL-strings" from 212992 to 217088

That doesn't give a lot of useful information why your pihole stops. I'm out of ideas....

No worries - the frequency of the issue seems to be far less, so I'll give it a few days and see if it happens again. If it does, I'll gather those same logs before running pihole restartdns and see if there's anything of interest in there.

Thanks for your help so far!

Haven't had the issue for almost 3 weeks now, so am assuming it was the SQLite issue you highlighted. Marked solved, thank you!

1 Like

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