Sending survey data failed

Creating a bug report/issue

I have searched the existing open and closed issues

Required Information

  • DietPi version | cat /boot/dietpi/.version
  • Distro version | echo $G_DISTRO_NAME $G_RASPBIAN
  • Kernel version | uname --all
  • Architecture | dpkg --print-architecture
  • SBC model | echo $G_HW_MODEL_NAME or (EG: RPi3)
  • Power supply used | (EG: 5V 1A RAVpower)
  • SD card used | (EG: SanDisk ultra)

Additional Information (if applicable)

  • Software title | (EG: Nextcloud)
  • Was the software title installed freshly or updated/migrated?
  • Can this issue be replicated on a fresh installation of DietPi?
    ← If you sent a “dietpi-bugreport”, please paste the ID here →
  • Bug report ID | echo $G_HW_UUID

Steps to reproduce

Expected behaviour

Actual behaviour

Extra details

Summary

This text will be hidden

Please provide more info, you just created an empty post.
What error do you get? And what about some system info?

Sorry, was too fast…

Dietpi on OrangePiZero3. Updated from v9.18 to v9.19 and got “sending survey data failed”.

Now tried to execute dietpi-survey and the result is:

[  OK  ] DietPi-Survey | Desired setting in /boot/dietpi.txt was already set: SURVEY_OPTED_IN=1
[..    ]cat: /tmp/G_EXEC_LOG: No such file or directory
[FAILED] DietPi-Survey | Sending survey data
 - Command: curl -m 20 -sT eb93452b-7436-4a73-82bd-1263c8b80674.txt sftp://dietpi-survey:upload2dietpi@ssh.dietpi.com:29248/survey/

This is freshly installed DietPi without any modifications. So, assume survey should work fine.

/tmp is empty

Thx

hmm for me survey is working without issues

root@DietPiProd:~# dietpi-survey
[  OK  ] DietPi-Survey | Setting in /boot/dietpi.txt adjusted: SURVEY_OPTED_IN=0
[  OK  ] DietPi-Survey | Purging survey data
root@DietPiProd:~#

So looks like the acutal sending process is failing, maybe an temporray issue.

Can you just run again

dietpi-survey 1

If it still fails: dietpi-survey is intentionally not verbose on errors, to not annoy users with details especially when they did not even opt in. So to get more curl output, try this:

cd /tmp
echo > eb93452b-7436-4a73-82bd-1263c8b80674.txt
curl -m 20 -sST 'eb93452b-7436-4a73-82bd-1263c8b80674.txt' 'sftp://dietpi-survey:upload2dietpi@ssh.dietpi.com:29248/survey/'

Result of Jappe request:

dietpi-survey 1
[FAILED] DietPi-Survey | Sending survey data
 - Command: curl -m 20 -sT eb93452b-7436-4a73-82bd-1263c8b80674.txt sftp://dietpi-survey:upload2dietpi@ssh.dietpi.com:29248/survey/

Result of MichaIng request:

cd /tmp etc...
curl: (28) Connection timed out after 20002 milliseconds

FYI, I use PiHole but I can see it allows ssh.dietpi.com for A and AAAA requests. Not other requests from DietPi.

Hmm, a connection timeout. Seems like DNS resolution via Pi-hole worked, but it could not build up the connection within 20 seconds for some reason :thinking:.

URL etc is all correct.

There is no such file on our server, so seems the upload never succeeded from your system. Does a ping get an answer?

ping -c 4 ssh.dietpi.com

In case does IPv4 vs IPv6 make a difference?

ping -4 -c 4 ssh.dietpi.com

Pings seem OK:

root@DietPi2:\~# ping -c 4 ssh.dietpi.com
PING ssh.dietpi.com (148.251.76.252) 56(84) bytes of data.
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=1 ttl=43 time=349 ms
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=2 ttl=43 time=409 ms
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=3 ttl=43 time=433 ms
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=4 ttl=43 time=558 ms

ssh.dietpi.com ping statistics —

4 packets transmitted, 4 received, 0% packet loss, time 3031ms
rtt min/avg/max/mdev = 349.286/437.463/558.162/76.095 ms

and:

root@DietPi2:\~# ping -4 -c 4 ssh.dietpi.com
PING ssh.dietpi.com (148.251.76.252) 56(84) bytes of data.
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=1 ttl=43 time=498 ms
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=2 ttl=43 time=521 ms
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=3 ttl=43 time=544 ms
64 bytes from mail.dietpi.com (148.251.76.252): icmp_seq=4 ttl=43 time=567 ms

ssh.dietpi.com ping statistics —

4 packets transmitted, 4 received, 0% packet loss, time 3003ms
rtt min/avg/max/mdev = 497.859/532.574/567.233/25.888 ms

Hmmm…

Ah so you have IPv6 disabled, right?

Hmm pings go through, just the SSH connection times out. Is there a chance that port 29248 is blocked somewhere in your network or by your ISP?

Nice, there is a tcptraceroute for investigating such:

sudo apt install tcptraceroute
tcptraceroute ssh.dietpi.com 29248

Yes, IPv6 is disabled on my router.

Here is tcptraceroute result:

root@DietPi2:\~# tcptraceroute ssh.dietpi.com 29248
Selected device wlan0, address 192.168.1.208, port 49455 for outgoing packets
Tracing the path to ssh.dietpi.com (148.251.76.252) on TCP port 29248, 30 hops max
1  192.168.1.1  1.340 ms  1.113 ms  2.768 ms
2  192.168.2.1  5.108 ms  2.844 ms  2.129 ms
3  \* \* \*
4  \* \* \*
5  \* \* \*
6  \* \* \*
7  \* \* \*
8  \* \* \*
9  \* \* \*
10  \* \* \*
11  \* \* \*
12  \* \* \*
13  \* \* \*
14  \* \* \*
15  \* \* \*
16  \* \* \*
17  \* \* \*
18  \* \* \*
19  \* \* \*
20  \* \* \*
21  \* \* \*
22  \* \* \*
23  \* \* \*
24  \* \* \*
25  \* \* \*
26  \* \* \*
27  \* \* \*
28  \* \* \*
29  \* \* \*
30  \* \* \*
Destination not reached

192.168.1.1 is my home router

192.168.2.1 is temporary, ISP delivered router, 5G. The one that shoud be returned when I terminate contract with them. I keep my home network intact, just connect ISP router in front.

So it looks like the packets never pass the 5G router. Maybe check whether it has some firewall or filter for certain ports. Maybe outgoing traffic for the non-privileged ports above 1023 is blocked, or on ports with more than 4 digits.

Actually can be also tested:

tcptraceroute ssh.dietpi.com 9999
tcptraceroute ssh.dietpi.com 1023

I like this tool: The protocol and whether the port on the destination is open or not does not matter.

And now I see MTR supports this as well:

apt install mtr-tiny
mtr --tcp -P 9999 ssh.dietpi.com

Never knew that :slightly_smiling_face:.

OK, so both of the tests

tcptraceroute ssh.dietpi.com 9999
tcptraceroute ssh.dietpi.com 1023

show same as before, packets never pass the 5G router.

I went inside 5G router to see what I can do and I found

Firewall security level

High: Traffic denied inbound and minimally permit common service outbound.

Low: All outbound traffic and pinhole-defined inbound traffic is allowed.

Off: All inbound and outbound traffic is allowed.

It was “High”, but when I changed to “Low”

root@DietPi2:\~# tcptraceroute ssh.dietpi.com 29248
Selected device wlan0, address 192.168.1.208, port 46925 for outgoing packets
Tracing the path to ssh.dietpi.com (148.251.76.252) on TCP port 29248, 30 hops max
1  192.168.1.1  1.105 ms  1.261 ms  4.082 ms
2  192.168.2.1  4.724 ms  2.115 ms  2.196 ms
3  10.163.80.2  153.637 ms  28.143 ms  54.070 ms
4  10.246.210.29  29.788 ms  27.667 ms  28.412 ms
5  101.115.191.250  31.136 ms  21.054 ms  16.091 ms
6  \* bri-apt-wic-wgw1-be-50.tpg.com.au (203.219.106.141) 158.332 ms \*
7  \* \* \*
8  \* \* \*
9  \* \* \*
10  \* \* \*
11  \* \* \*
12  \* \* dls-bb1-link.ip.twelve99.net (62.115.140.237) 428.834 ms
13  \* \* \*
14  \* \* \*
15  \* rest-bb1-link.ip.twelve99.net (62.115.138.70) 394.978 ms \*
16  \* \* \*
17  \* \* \*
18  \* \* \*
19  \* hbg-b2-link.ip.twelve99.net (62.115.120.71) 349.241 ms \*
20  \* \* \*
21  \* \* \*
22  ex9k2.dc11.fsn1.hetzner.com (213.239.229.10)  487.789 ms \* \*
23  \* \* \*
24  \* \* \*
25  \* \* \*
26  \* \* \*
27  \* \* mail.dietpi.com (148.251.76.252) \[open\] 364.395 ms

and then finally

root@DietPi2:\~# dietpi-survey
\[  OK  \] DietPi-Survey | Desired setting in /boot/dietpi.txt was already set: SURVEY_OPTED_IN=1
\[  OK  \] DietPi-Survey | Sending survey data

So, 5G firewall was preventing survey to work.

Thx for help…

2 Likes

Great. I guess port 22 would have worked then, but as said, we use a non-standard port intentionally to reduce spam in our system logs and network load.

The “Low” profile looks like what I know as common NAT standard: all outbound traffic allowed, inbound only for explicitly defined forwarded ports. If there are concerns about malicious software doing unwanted outbound connections via non-standard ports, probably there is also a way to explicitly define outbound ports to additionally allow with “High” profile.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.