# INTCollector issues **URL:** https://forum.p4.org/t/intcollector-issues/905 **Category:** P4 Dev **Created:** [July 27, 2023, 6:51am UTC](https://forum.p4.org/t/intcollector-issues/905 "2023-07-27T06:51:03Z") **Posts on this page:** 4 **Page:** 1
### Author: ![msaueressig](https://avatars.discourse-cdn.com/v4/letter/m/edb3f5/32.png) [@msaueressig](https://forum.p4.org/u/msaueressig) #### Post date: [July 27, 2023, 6:51am UTC](https://forum.p4.org/t/intcollector-issues/905/1 "2023-07-27T06:51:03Z")
Hello. i’m trying to use INTCollector( [tunv\_ebpf / INTCollector · GitLab](https://gitlab.com/tunv_ebpf/BPFCollector)) to collect INT data from a testbed. I’m really struggling to do so, first I couldn’t install ONOS properly, so I installed a VM from a tutorial( [Basic ONOS Tutorial - ONOS - Wiki (onosproject.org)](https://wiki.onosproject.org/display/ONOS/Basic+ONOS+Tutorial)) to have a implementation of ONOS in my VM. After that I installed a P4 environment using the .sh script in the p4-guide Github repository. Now I’m just trying to run it and I’m failing to do so. I’m receiving the following log when trying to run a pytest with InfluxDB: ============================= test session starts ============================== platform linux – Python 3.8.10, pytest-7.4.0, pluggy-1.2.0 – /usr/bin/python3 cachedir: .pytest\_cache rootdir: /home/sdn/BPFCollector collecting … collected 2 items test\_e2e.py::test\_e2e\_indb\_threshold\_v10 FAILED [50%] test\_e2e.py::test\_e2e\_indb\_interval\_v10 FAILED [100%] =================================== FAILURES =================================== \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ test\_e2e\_indb\_threshold\_v10 \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ setup\_veth = None ``` def test_e2e_indb_threshold_v10(setup_veth): cmd = [python_ver, "InDBClient.py", "-t", "-D", test_db, outif] end_to_end_influxdb_v10(cmd) ``` test\_e2e.py:154: * * * test\_e2e.py:51: in end\_to\_end\_influxdb\_v10 start\_collector(cmd) * * * cmd = [‘python3’, ‘InDBClient.py’, ‘-t’, ‘-D’, ‘INT\_test\_database’, ‘int\_veth\_1’] ``` def start_collector(cmd): global subp try: subp = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, universal_newlines=True) except Exception as e: print(e) return None p = subp.stdout.readline() assert (subp.poll() == None), "Error loading XDP program" ``` E AssertionError: Error loading XDP program E assert 1 == None E + where 1 = \\>() E + where \\> = \.poll test\_e2e.py:31: AssertionError \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ test\_e2e\_indb\_interval\_v10 \_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_\_ setup\_veth = None ``` def test_e2e_indb_interval_v10(setup_veth): cmd = [python_ver, "InDBClient.py", "-t", "-e", "INTERVAL", "-D", test_db, outif] end_to_end_influxdb_v10(cmd) ``` test\_e2e.py:158: * * * test\_e2e.py:51: in end\_to\_end\_influxdb\_v10 start\_collector(cmd) * * * cmd = [‘python3’, ‘InDBClient.py’, ‘-t’, ‘-e’, ‘INTERVAL’, ‘-D’, …] ``` def start_collector(cmd): global subp try: subp = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, universal_newlines=True) except Exception as e: print(e) return None p = subp.stdout.readline() assert (subp.poll() == None), "Error loading XDP program" ``` E AssertionError: Error loading XDP program E assert 1 == None E + where 1 = \\>() E + where \\> = \.poll test\_e2e.py:31: AssertionError =========================== short test summary info ============================ FAILED test\_e2e.py::test\_e2e\_indb\_threshold\_v10 - AssertionError: Error loadi… FAILED test\_e2e.py::test\_e2e\_indb\_interval\_v10 - AssertionError: Error loadin… ============================== 2 failed in 10.00s ============================== I’m not sure if the problem is the configuration of InfluxDB, since i Had a problem with it last time I ran a test. Where can I see if it is connected to localhost(Both InfluxDB and INTCollector)? Did anyone use INTCollector before? If there is any VM .ova that has a fully functioning INTCollector or any INT collector API, I would appreciate. Thanks ---
### Author: ![msaueressig](https://avatars.discourse-cdn.com/v4/letter/m/edb3f5/32.png) [@msaueressig](https://forum.p4.org/u/msaueressig) #### Post date: [July 27, 2023, 8:07pm UTC](https://forum.p4.org/t/intcollector-issues/905/2 "2023-07-27T20:07:53Z")
UDPATE: It seems the problem lies in the bcc package necessary for the application. I’m struggling to make it work. I’m using a Ubuntu 20.04.6 LTS with llvm 10.0.0(eventhought it says that my version is in /usr/lib/llvm-60/include 6.0.1). I’ve tried the latest release and now I tried the 0.24 release. It helped a little to build the package from source, however it’s still not working ---
### Author: ![msaueressig](https://avatars.discourse-cdn.com/v4/letter/m/edb3f5/32.png) [@msaueressig](https://forum.p4.org/u/msaueressig) #### Post date: [July 27, 2023, 9:08pm UTC](https://forum.p4.org/t/intcollector-issues/905/3 "2023-07-27T21:08:22Z")
I’ve managed to run the BCC, however I still not able to run INTCollector. When i tried to start InfluxDB the following occurs: bpf: Argument list too long. Program too large (1400 insns), at most 4096 insns Traceback (most recent call last): File “InDBClient.py”, line 49, in collector = InDBCollector.InDBCollector(int\_dst\_port=args.int\_port, File “InDBCollector.pyx”, line 61, in InDBCollector.InDBCollector. **init** self.fn\_collector = self.bpf\_collector.load\_func(“collector”, BPF.XDP) File “/usr/lib/python3/dist-packages/bcc/ **init**.py”, line 522, in load\_func raise Exception(“Failed to load BPF program %s: %s” % Exception: Failed to load BPF program b’collector’: Argument list too long ---
### Author: ![Xiaoxin](https://avatars.discourse-cdn.com/v4/letter/x/3be4f8/32.png) [@Xiaoxin](https://forum.p4.org/u/Xiaoxin) #### Post date: [March 7, 2024, 7:19am UTC](https://forum.p4.org/t/intcollector-issues/905/4 "2024-03-07T07:19:58Z")
I also encountered such a problem. Could you tell me how you solved it. How do you run the BCC?