Unknown SDN controller type #594
Comments
|
It looks like it's not picking up your environment variables for some reason, can you do |
|
All variables are correct in env, As I am running poseidon and faucet on the same host I did not exported the URL, USER and PASS as suggested. 2018-05-24T11:09:24+00:00 172.17.0.1 plugin[21167]: 2018-05-24 06:09:24,432 - DEBUG - poseidonMonitor:400 - Monitor:config:{'config': 'True', 'logging_file': 'logging.json', 'logger_level': 'INFO', 'reinvestigation_frequency': '900', 'max_concurrent_reinvestigations': '1', 'scan_frequency': '5', 'rabbit_server': 'RABBIT_SERVER', 'rabbit_port': '5672', 'collector_nic': 'ens8', 'collector_interval': '900', 'vent_ip': 'vent_ip', 'vent_port': '8080', 'FA_RABBIT_ENABLED': '', 'FA_RABBIT_HOST': '', 'FA_RABBIT_PORT': '', 'FA_RABBIT_EXCHANGE': '', 'FA_RABBIT_EXCHANGE_TYPE': 'topic', 'FA_RABBIT_ROUTING_KEY': ''} I think it is because of uri? Do I need to export it as well? For making any change what steps should I follow? to make it up and running? |
|
Looks like that's an unexpected bug: That means the local file is not working is trying to connect remotely, except it's not specified. I'll open a new issue for that. In the meantime you should be able to get around that by exporting the vars for URL, USER and PASS even though they are just for the local host. I think I've only been testing remotely, so that's a good catch. Let me know if adding those 3 vars fixes the issue. For making changes, you should be able to just remove the containers and run |
|
After exporting the vars for URL, USER and PAS, still getting the same error.
Moreover, the log file is as follows: Moreover, I am adding further details of my setup, may be the error is caused by a problem because of these anything wrong with these configurations. output of env is as follows: Faucet configurations are as follows: network interfaces of the client are as follows: |
|
@FahadMaqbool thanks for the additional details, that is helpful. I'm working on tracking this down for you - will update soon. |
|
Ok, tracked it down and pushed up a fix. It also fixes #600 so you don't need the URI/USER/PASS anymore either if both FAUCET and Poseidon are running on the same box. Let me know if that helps. |
|
@cglewis Thanks for the update, but now as I exported the collector_nic = ens8, (as you can see below the results of |
|
apologies , I had used the wrong word in the helper script when I updated it to properly handle missing environment variables. #607 should fix it. |
|
Seems like that basic issues with poseidon are now fix. will you please help me in getting traffic at the mirro port as my Faucet configurations are as follows: As I ping from client 1 to client 2 (and vice versa) that are defined as port1.0.1 and port1.0.2 in faucet yaml file. but faucet is not picking it up. I am using an allied telesis switch. |
|
Can you provide the logs from You also need to have your mirror port (3) set as Are you starting Faucet with the Rabbit event adapter, or not? |
|
## Can you provide the logs from faucet.log as well? faucet.log is as follows. # You also need to have your mirror port (3) set as output_only: true and not be on a native_vlan. Do I need to set output_only: true in faucet.yaml? ## Are you starting Faucet with the Rabbit event adapter, or not? I dont know exactly, I have Installed the faucet and made it run on boot. |
|
The logs look good - it's learning the L3 addresses. For the mirror port, yeah setting it to So the rabbit event adapter is not set for Faucet by default (and technically not required, but Poseidon by default is expecting that Faucet is run using it. I start Faucet with I add these environment variables: Then run: If you don't plan to use the event sock, then you need to set: |
|
any ideas, guide or tutorial for running Poseidon in IXP (Internert Exchange Point) setup? |
|
@FahadMaqbool : I'd be concerned that Poseidon might not perform well in that scenario, particularly as there are (most likely) going to be a lot of NAT'ed and "overloaded" addresses crossing the wire. It really wasn't designed to handle that deployment model and while it technically could work, the ML is going to deliver crap results, IMO. (i.e. It should collect, spin up containers, process data just fine - but the current ML models aren't trained for that vantage point) Question: what's the function you'd like to most use it for in/at an IXP? I ask because maybe we could work on a specific use-case, depending on what it is you wanted to accomplish. |
|
@gregs5 Thank you for your response, A very useful function to have in an IXP will be to have a detection and blocking mechanism of DDOS attack using advanced blackholing. http://conferences.sigcomm.org/sosr/2017/papers/sosr17-demo-blackholing.pdf |
|
@FahadMaqbool per the original question in this issue - has that been resolved? If so, I'd like to close this out and start a new issue that discusses IXP as a use case. |
|
yeah sure. |
ERROR - UpdateSwitchState:136 - Unknown SDN controller type
I understand this error message, but i am running the export commands to provide this information as:
bitsym@faucetall-fahad:/$ cd poseidon/
bitsym@faucetall-fahad:/poseidon$ export controller_type=faucet
bitsym@faucetall-fahad:/poseidon$ export collector_nic=ens4
bitsym@faucetall-fahad:/poseidon$ export controller_mirror_ports=’{“allied-telesis”:3}’
bitsym@faucetall-fahad:/poseidon$ export max_concurrent_reinvestigations=1
bitsym@faucetall-fahad:/poseidon$ ./helpers/run
but still getting the error in the log file.
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Handle_FullConfig configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: 2018-05-18 04:13:39,365 - INFO - Monitor_Helper_Base:55 - Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Update_Switch_State configure()
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: 2018-05-18 04:13:39,365 - ERROR - UpdateSwitchState:136 - Unknown SDN controller type
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: Traceback (most recent call last):
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: File "/tmp/poseidonWork/poseidon/poseidonMonitor/poseidonMonitor.py", line 820, in
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: main(skip_rabbit=False)
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: File "/tmp/poseidonWork/poseidon/poseidonMonitor/poseidonMonitor.py", line 716, in main
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: pmain = Monitor(skip_rabbit=skip_rabbit)
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: File "/tmp/poseidonWork/poseidon/poseidonMonitor/poseidonMonitor.py", line 343, in init
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: self.mod_configuration['max_concurrent_reinvestigations'])
2018-05-18T09:13:39+00:00 172.17.0.1 plugin[1062]: ValueError: invalid literal for int() with base 10: ''
iv_timer_register: called with timer still on the heap
[2018-05-18T04:13:43.740590] WARNING: Configuration file format is too old, syslog-ng is running in compatibility mode Please update it to use the syslog-ng 3.9 format at your time of convenience, compatibility mode can operate less efficiently in some cases. To upgrade the configuration, please review the warnings about incompatible changes printed by syslog-ng, and once completed change the @Version header at the top of the configuration file.;
The text was updated successfully, but these errors were encountered: