Issue Connecting to Mecheye NANO Ultra-GL

twoeste · January 14, 2026, 2:19pm

Hi I am using the Mecheye Nano ultra-gl connected to a linux computer running ubuntu 20.04. I have installed your SDK v2.5.4. I am able to see the camera using the IP configure tool in ubuntu and I can choose the camera index 0 when I run the sample ConnectToCamera project on Github but I get a segfault and an error message that the camera failed to connect to the Ethernet port of the computer. I have pasted the full error message when I run the program at the end of this message. I also get a segfault when I am using the ip_configurator tool trying to change the IP address on ubuntu. I have to connect the camera to a windows computer to change the IP address for testing.

Steps so far I have done to debug this issue:

Confirmed that my machines Ip address is on the same subnet as the camera for the Ethernet port the camera is connected to. I can ping the camera via the ping tool in the terminal. Camera IP: 192.168.1.101 Subnet mask: 255.255.255.0, Ethernet IP: 192.168.1.100 Subnet mask: 255.255.255.0.
Tried another very different set of IP addresses Camera IP: 169.254.4.29 Subnet mask: 255.255.255.0, Ethernet IP: 169.254.4.44 Subnet mask: 255.255.255.0.
Tired downgrading to SDK v2.5.1 but that segfaults in a lot of the programs because I think the FW on my mecheye does not support that SDK. I have not found a way to also downgrade the FW yet so this may still be something we can try if I can get the FW to match the older SDK?
I have connected the mecheye camera to a windows computer and I am able to connect to the camera in the Mech-Eye Viewer software and take pictures.
I have confirmed that the MTU size is correct and I have not enabled large packets on the ethernet.
I have disabled all other network interfaces and disabled any VPNs on my machine. During this I am still able to ping the cameras IPaddress.

twoeste@3460db4:~/projects/mecheye_cpp_samples/area_scan_3d_camera/Basic/ConnectToCamera/build$ ./ConnectToCamera
Discovering all available cameras...
Camera index: 0
.............................
Model:                        Mech-Eye NANO ULTRA-GL
Device name:                  
Serial number:                RUM35254A510EG20
IP address:                   192.168.1.101
Subnet mask:                  255.255.255.0
IP address assignment method: Static
Hardware version:             V5.2.0
Firmware version:             V2.5.4
Support status:               Supported
.............................

Please enter the index of the camera that you want to connect: 0
Failed to obtain the IP address of the computer Ethernet port connected to the device. Please try disabling the Ethernet interface cards not connected to the device.
ping: option requires an argument -- 'I'

Usage
  ping [options] <destination>

Options:
  <destination>      dns name or ip address
  -a                 use audible ping
  -A                 use adaptive ping
  -B                 sticky source address
  -c <count>         stop after <count> replies
  -D                 print timestamps
  -d                 use SO_DEBUG socket option
  -f                 flood ping
  -h                 print help and exit
  -I <interface>     either interface name or address
  -i <interval>      seconds between sending each packet
  -L                 suppress loopback of multicast packets
  -l <preload>       send <preload> number of packages while waiting replies
  -m <mark>          tag the packets going out
  -M <pmtud opt>     define mtu discovery, can be one of <do|dont|want>
  -n                 no dns name resolution
  -O                 report outstanding replies
  -p <pattern>       contents of padding byte
  -q                 quiet output
  -Q <tclass>        use quality of service <tclass> bits
  -s <size>          use <size> as number of data bytes to be sent
  -S <size>          use <size> as SO_SNDBUF socket option value
  -t <ttl>           define time to live
  -U                 print user-to-user latency
  -v                 verbose output
  -V                 print version and exit
  -w <deadline>      reply wait <deadline> in seconds
  -W <timeout>       time to wait for response

IPv4 options:
  -4                 use IPv4
  -b                 allow pinging broadcast
  -R                 record route
  -T <timestamp>     define timestamp, can be one of <tsonly|tsandaddr|tsprespec>

IPv6 options:
  -6                 use IPv6
  -F <flowlabel>     define flow label, default is random
  -N <nodeinfo opt>  use icmp6 node info query, try <help> as argument

For more details see ping(8).
Failed to obtain the IP address of the computer Ethernet port connected to the device. Please try disabling the Ethernet interface cards not connected to the device.
Connected to the camera successfully.
Disconnected from the camera successfully.
Segmentation fault (core dumped)

Thank you for any support in this issue!

cycqaq · January 15, 2026, 9:35pm

I believe the issue is a network interface discovery failure within the Mech-Eye SDK. Could you try running the following command?

sudo apt update && sudo apt install net-tools

Please let me know if that works. Thanks!

twoeste · January 16, 2026, 2:08am

I already had net-tools installed but I re-ran it just to make sure and the program has the same error.

cycqaq · January 16, 2026, 3:36pm

The ping: option requires an argument -- 'I' error and subsequent segfault confirm the SDK cannot resolve the network interface name. It is passing an empty string to the system command, causing the crash.

Since net-tools is installed, this is likely a routing conflict. Please try these steps:

Isolate the Network: If WiFi or Docker uses the same 192.168.1.x subnet, the SDK fails to select the correct path. Temporarily disable them to test: sudo nmcli radio wifi off sudo ifconfig docker0 down Then retry ./ConnectToCamera.
Verify the Route: Run ip route get 192.168.1.101. The output must show your physical Ethernet interface (e.g., dev eth0 or eno1). If it shows nothing, you need to add a static route in your OS settings.
Disable Firewall: sudo ufw disable

If the problem persists, please let me know and I will reach out to the corresponding department for a solution.

twoeste · January 16, 2026, 3:48pm

I tried all 3 options and the issue still persists. I can see the camera a no other networks are active while testing.

twoeste · January 16, 2026, 7:40pm

UPDATE Jan 16,

I was using Claude to help debug this and it changed the example program ConnectToCamera to connect directly via IP and that was actually able to connect and capture images via the other tools if we changed the program to just take in the input via cli. Here was the diff of what was changed in the example code. Also it still has the problem that the ping command fails but it still seems to go ahead under the hood.

Index: area_scan_3d_camera/Basic/ConnectToCamera/ConnectToCamera.cpp
IDEA additional info:
Subsystem: com.intellij.openapi.diff.impl.patch.CharsetEP
<+>UTF-8
===================================================================
diff --git a/area_scan_3d_camera/Basic/ConnectToCamera/ConnectToCamera.cpp b/area_scan_3d_camera/Basic/ConnectToCamera/ConnectToCamera.cpp
--- a/area_scan_3d_camera/Basic/ConnectToCamera/ConnectToCamera.cpp	(revision 394f6bb415e8a1ffc951d5d2b4d5f0817db3407d)
+++ b/area_scan_3d_camera/Basic/ConnectToCamera/ConnectToCamera.cpp	(date 1768591585226)
@@ -38,41 +38,50 @@
 #include "area_scan_3d_camera/Camera.h"
 #include "area_scan_3d_camera/api_util.h"
 
-int main()
+int main(int argc, char* argv[])
 {
-    std::cout << "Discovering all available cameras..." << std::endl;
-    std::vector<mmind::eye::CameraInfo> cameraInfoList = mmind::eye::Camera::discoverCameras();
+    mmind::eye::Camera camera;
+    const std::string successMessage = "Connected to the camera successfully.";
+
+    // If an IP address is provided as argument, connect directly
+    if (argc > 1) {
+        std::string ipAddress = argv[1];
+        std::cout << "Connecting to camera at IP: " << ipAddress << std::endl;
+        showError(camera.connect(ipAddress), successMessage);
+    } else {
+        std::cout << "Discovering all available cameras..." << std::endl;
+        std::vector<mmind::eye::CameraInfo> cameraInfoList = mmind::eye::Camera::discoverCameras();
 
-    if (cameraInfoList.empty()) {
-        std::cout << "No cameras found." << std::endl;
-        return -1;
-    }
+        if (cameraInfoList.empty()) {
+            std::cout << "No cameras found." << std::endl;
+            std::cout << "Tip: You can connect directly by IP: ./ConnectToCamera <ip_address>" << std::endl;
+            return -1;
+        }
 
-    // Display the information of all available cameras.
-    for (int i = 0; i < cameraInfoList.size(); i++) {
-        std::cout << "Camera index: " << i << std::endl;
-        printCameraInfo(cameraInfoList[i]);
-    }
+        // Display the information of all available cameras.
+        for (int i = 0; i < cameraInfoList.size(); i++) {
+            std::cout << "Camera index: " << i << std::endl;
+            printCameraInfo(cameraInfoList[i]);
+        }
 
-    std::cout << "Please enter the index of the camera that you want to connect: ";
-    unsigned inputIndex;
+        std::cout << "Please enter the index of the camera that you want to connect: ";
+        unsigned inputIndex;
 
-    // Enter the index of the camera to be connected and check if the index is valid.
-    while (1) {
-        std::string str;
-        std::cin >> str;
-        if (std::regex_match(str.begin(), str.end(), std::regex{"[0-9]+"}) &&
-            atoi(str.c_str()) < cameraInfoList.size()) {
-            inputIndex = atoi(str.c_str());
-            break;
-        }
-        std::cout
-            << "Input invalid! Please enter the index of the camera that you want to connect: ";
-    }
+        // Enter the index of the camera to be connected and check if the index is valid.
+        while (1) {
+            std::string str;
+            std::cin >> str;
+            if (std::regex_match(str.begin(), str.end(), std::regex{"[0-9]+"}) &&
+                atoi(str.c_str()) < cameraInfoList.size()) {
+                inputIndex = atoi(str.c_str());
+                break;
+            }
+            std::cout
+                << "Input invalid! Please enter the index of the camera that you want to connect: ";
+        }
 
-    mmind::eye::Camera camera;
-    const std::string successMessage = "Connected to the camera successfully.";
-    showError(camera.connect(cameraInfoList[inputIndex]), successMessage);
+        showError(camera.connect(cameraInfoList[inputIndex]), successMessage);
+    }
 
     camera.disconnect();
     std::cout << "Disconnected from the camera successfully." << std::endl;

cycqaq · January 20, 2026, 4:25pm

Thanks for the update! This aligns with the feedback from our internal team, who also suggested connecting directly via IP to bypass potential network restrictions. I’ll record these findings and share them with our development team for further review.

twoeste · January 20, 2026, 4:37pm

Was there any guidance from them on how what the Ping error output was in the SDK or how to configure the network correctly so I do not need to know the IP before hand?

cycqaq · January 21, 2026, 4:22pm

Here’s the reply from our internal team. Hope this helps.

The issue of being unable to obtain the IP address of the industrial computer’s network port is only a prompt and will not affect the normal use of the camera. This prompt can be ignored.
The high probability of this occurring is due to the presence of other network cards on the industrial computer, which causes the acquisition to fail. You can try configuring an independent network segment and temporarily disabling the other network cards, leaving only the camera network card enabled.

twoeste · January 21, 2026, 5:32pm

Thanks for the info I will continue using the IP address to connect. But I would think your sdk should be able to work with other networks connected. Also I have tried disabling the other networks and the issue still occurs.

cycqaq · January 22, 2026, 4:17pm

Hi Thomas, thanks for your patience. I’ve shared your feedback with our team. While the IP connection is a valid workaround, I understand that the SDK should ideally discover the camera automatically, especially since it works on Windows.

Our team suggests that the ‘Ping error’ you see is a secondary prompt caused by network interference and shouldn’t stop the camera from functioning once connected. However, I am still pushing for more specific guidance on the Discovery issue in multi-NIC environments. In the meantime, please refer to the network troubleshooting documentation for further investigation: FAQ: Using Mech-Eye API

twoeste · January 23, 2026, 6:34pm

Hi,

I have tried again every step in the attached guide. Any update on how to suppress the ping command output in the sdk?

Thanks,
Thomas

Mech-Mind_Dshuai · January 27, 2026, 6:06am

Hello,

When you mentioned the ping error, do you mean the error indicating that the industrial PC’s IP address cannot be found? That message can be ignored and does not affect normal usage.

If you would like to avoid that warning, the network connection needs to be properly configured and reachable. If possible, could you please share the on-site network topology, and let us know whether there is any VPN, virtual machine, or other special network setup in the internal network?

In addition, the camera’s factory default IP address is usually set to 192.168.23.X. During the initial configuration, it can be modified to match the same subnet as the host PC.

twoeste · January 27, 2026, 1:20pm

Here is my network structure with the IP addresses redacted for privacy.

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000                                                    
      link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00                                                                                      
      inet 127.0.0.1/8 scope host lo                                                                                                             
         valid_lft forever preferred_lft forever                                                                                                 
      inet6 ::1/128 scope host                                                                                                                   
         valid_lft forever preferred_lft forever                                                                                                 
  2: enp3s0f0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000                                        
      link/ether XX:XX:XX:XX:XX:XX brd ff:ff:ff:ff:ff:ff                                                                                         
      inet 192.168.1.XXX/24 brd 192.168.1.255 scope global noprefixroute enp3s0f0                                                                
         valid_lft forever preferred_lft forever                                                                                                 
      inet6 fe80::XXXX:XXXX:XXXX:XXXX/64 scope link noprefixroute                                                                                
         valid_lft forever preferred_lft forever                                                                                                 
  3: enp2s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000                                                
      link/ether XX:XX:XX:XX:XX:XX brd ff:ff:ff:ff:ff:ff                                                                                         
      inet 192.168.0.XXX/24 brd 192.168.0.255 scope global dynamic noprefixroute enp2s0                                                          
         valid_lft 6275sec preferred_lft 6275sec                                                                                                 
      inet6 XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX/64 scope global temporary dynamic                                                            
         valid_lft 1673sec preferred_lft 1673sec                                                                                                 
      inet6 XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX/64 scope global temporary deprecated dynamic                                                 
         valid_lft 1673sec preferred_lft 0sec                                                                                                    
      inet6 XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX/64 scope global dynamic mngtmpaddr noprefixroute                                             
         valid_lft 1673sec preferred_lft 1673sec                                                                                                 
      inet6 fe80::XXXX:XXXX:XXXX:XXXX/64 scope link noprefixroute                                                                                
         valid_lft forever preferred_lft forever                                                                                                 
  5: enxXXXXXXXXXXXX: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel state DOWN group default qlen 1000                             
      link/ether XX:XX:XX:XX:XX:XX brd ff:ff:ff:ff:ff:ff                                                                                         
  7: docker0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default                                                
      link/ether XX:XX:XX:XX:XX:XX brd ff:ff:ff:ff:ff:ff                                                                                         
      inet 172.100.0.1/24 brd 172.100.0.255 scope global docker0                                                                                 
         valid_lft forever preferred_lft forever                                                                                                 
  8: tailscale0: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1280 qdisc fq_codel state UNKNOWN group default qlen 500                          
      link/none                                                                                                                                  
      inet6 fe80::XXXX:XXXX:XXXX:XXXX/64 scope link stable-privacy                                                                               
         valid_lft forever preferred_lft forever

The camera is set to 192.168.1.101 so I can ping it from enp3s0f0.

Mech-Mind_Dshuai · January 28, 2026, 3:30am

Thank you for your detailed investigation and patience — your findings are correct, and I’d like to clarify what’s happening and suggest the fastest workaround.

Root cause (Linux SDK only):
On Linux, the Mech-Eye SDK attempts to automatically determine which host Ethernet interface is connected to the camera. Internally, this involves enumerating network interfaces and performing a connectivity check.
In multi-NIC environments (for example, systems with multiple Ethernet ports, Docker, or VPN-related virtual interfaces), this detection step may fail to resolve a valid interface name. When that happens, an internal connectivity check command is constructed with an empty interface parameter, which leads to the warning you observed and may result in a segmentation fault during cleanup.

Importantly:

This issue affects only the interface discovery logic, not the actual camera communication.

The camera connection and image acquisition pipeline itself is functioning correctly, as you have already verified.

Fastest and recommended workaround (Linux):
For now, please connect to the camera directly by IP address, instead of using camera discovery or index-based connection. This bypasses the interface-detection step entirely and is the most stable approach on Linux systems with multiple network interfaces.

This method is safe, fully supported, and already proven to work reliably in your tests.

About the warning message:
The message about failing to obtain the host Ethernet port IP is a secondary diagnostic warning and does not indicate a camera-side failure. When connecting by IP, it can be safely ignored.

We have shared your findings internally, and they are very helpful for improving robustness in multi-interface Linux environments in future SDK updates.

Thank you again for the clear reproduction and analysis. Please let us know if you need confirmation for production use.

wjahn · January 29, 2026, 4:09am

Same here with PRO M series.

I’ve confirmed that this happens even when bypassing discovery and connecting directly via IP.

It’s functional but the continuous log warnings are an eyesore.

It looks like an internal SDK issue that can’t be handled on the user side.

I’d appreciate it if the dev team could resolve this in the upcoming version.