Not sufficient rights to launch a simulation

Gaurav_Kumar · 2017-10-11 21:40:33 UTC

I removed my previous installation of NRP because it ran into some problems and installed a fresh copy.
During installation: deploy-models-nrp.sh script was not found in gzweb directory.
Still I went ahead with rest of the steps. Now, cle-start produces these errors.

2017-10-11 23:29:08 GMT+0200 Failed to execute request http://localhost:8080/experiment/ExDXMLExample-with-sm/preview. ERROR: Error: Error: ESOCKETTIMEDOUT
2017-10-11 23:29:08 GMT+0200 Failed to execute request http://localhost:8080/experiment/ExDBraitenbergMouse/preview. ERROR: Error: Error: ESOCKETTIMEDOUT
2017-10-11 23:29:08 GMT+0200 Failed to execute request http://localhost:8080/experiment/ExDRetinaICubTrackingRG/preview. ERROR: Error: Error: ESOCKETTIMEDOUT
2017-10-11 23:29:08 GMT+0200 Failed to load experiment Image: Error: Error: ESOCKETTIMEDOUT
at Request._callback (/home/gauku/NRP/nrpBackendProxy/proxy/serversProxy.js:88:21)
at self.callback (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:186:22)
at emitOne (events.js:96:13)
at Request.emit (events.js:191:7)
at ClientRequest. (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:781:16)
at Object.onceWrapper (events.js:293:19)
at emitNone (events.js:86:13)
at ClientRequest.emit (events.js:188:7)
at Socket.emitTimeout (_http_client.js:679:10)
at Object.onceWrapper (events.js:293:19)
2017-10-11 23:29:08 GMT+0200 Failed to load experiment Image: Error: Error: ESOCKETTIMEDOUT
at Request._callback (/home/gauku/NRP/nrpBackendProxy/proxy/serversProxy.js:88:21)
at self.callback (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:186:22)
at emitOne (events.js:96:13)
at Request.emit (events.js:191:7)
at ClientRequest. (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:781:16)
at Object.onceWrapper (events.js:293:19)
at emitNone (events.js:86:13)
at ClientRequest.emit (events.js:188:7)
at Socket.emitTimeout (_http_client.js:679:10)
at Object.onceWrapper (events.js:293:19)
2017-10-11 23:29:08 GMT+0200 Failed to load experiment Image: Error: Error: ESOCKETTIMEDOUT
at Request._callback (/home/gauku/NRP/nrpBackendProxy/proxy/serversProxy.js:88:21)
at self.callback (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:186:22)
at emitOne (events.js:96:13)
at Request.emit (events.js:191:7)
at ClientRequest. (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:781:16)
at Object.onceWrapper (events.js:293:19)
at emitNone (events.js:86:13)
at ClientRequest.emit (events.js:188:7)
at Socket.emitTimeout (_http_client.js:679:10)
at Object.onceWrapper (events.js:293:19)
2017-10-11 23:29:08 GMT+0200 Failed to execute request http://localhost:8080/experiment/ExDMouseCDP1/preview. ERROR: Error: Error: ESOCKETTIMEDOUT
2017-10-11 23:29:08 GMT+0200 Failed to load experiment Image: Error: Error: ESOCKETTIMEDOUT
at Request._callback (/home/gauku/NRP/nrpBackendProxy/proxy/serversProxy.js:88:21)
at self.callback (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:186:22)
at emitOne (events.js:96:13)
at Request.emit (events.js:191:7)
at ClientRequest. (/home/gauku/NRP/nrpBackendProxy/node_modules/request/request.js:781:16)
at Object.onceWrapper (events.js:293:19)
at emitNone (events.js:86:13)
at ClientRequest.emit (events.js:188:7)
at Socket.emitTimeout (_http_client.js:679:10)
at Object.onceWrapper (events.js:293:19)

and in browser, I can’t launch any experiment. The tooltip over launch button says that I don’t have sufficient rights to launch a simulation.

What should I do to get NRP to work?

lguyot · 2017-10-12 06:59:57 UTC

Dear Gaurav,

The script deploy-models-nrp.sh has been removed from neurorobotics/gzweb on the 9th of October 2017 and is not needed anymore, provided that $HBP/Models/create-symlinks.sh is run after the gzweb installation.
I adjusted our wiki documentation few minutes ago: https://bbpteam.epfl.ch/project/spaces/display/HSP10/Install+Gzweb

Regarding your error, I have no clear explanation at the moment. But some suggestions:
cle-kill
cd $HBP/user-scripts
./user-scripts update_nrp rebase all
./user-scripts update_nrp build all # this will take a while
./configure_nrp # yes to all prompts
cle-start
cle-frontend

You may also clean your browser cache or open a private navigation window in your browser.
If the problem persists, could you please copy the full console log and explain how it relates to the title of your question?

Best regards,
Luc

ManosAngelidis · 2017-10-12 07:10:45 UTC

Hello Gaurav,

On top of Luc’s comments you can also try to cle-nginx before cle-start.

On the sufficient rights, check if you have the variable $STORAGE_PATH in you environment. If you do, make sure that the directory it points to exists. By default it should be ~/.opt/nrpStorage. If the configure_nrp script went ok you should have inside the $STORAGE_PATH a folder called FS_DB and inside a file called users.

Before you launch an experiment you have to log in to your storage with credentials you will find inside the users file (I am not writing it here for security reasons). If you are not prompted to login when you go to localhost:9000/#/esv-web, try to go to localhost:9000/#/esv-private . There you should get an authentication page and after you log in you should have enough rights to launch a simulation. I hope it helps.

Regards,
Manos

vonarnim · 2017-10-12 07:49:36 UTC

Bitbucket install doc updated too. Thanks for reporting.
Axel

Gaurav_Kumar · 2017-10-12 12:19:47 UTC

Thanks for your reply.
I followed your suggestions and $STORAGE_PATH environment variable was defined, but the directory didn’t. After manually creating the ~/.opt/nrpStorage directory, I can see the login screen but again, the username and password found in users file under ~/.opt/FS_db directory do not work. What should I try next?

ManosAngelidis · 2017-10-12 12:32:16 UTC

Hello Gaurav,

Can you paste the contents of $NRP/nrpBackendProxy/config.json here? I have a feeling that you are in the EPFL login page, thus your credentials are not working.

Regards,
Manos

Gaurav_Kumar · 2017-10-12 13:02:19 UTC

Here it is.

{
  "refreshInterval": 5000,
  "auth": {
    "renewInternal": 600000,
    "clientId": "0dcfb392-32c7-480c-ae18-cbaf29e8a6b1",
    "clientSecret": "WfjTCn30aTlcmbL-wWVhmsAUgobaX9o5xR-6WQ9jQEDFAodNVozzjy283qaOjUYkl28DfG-L6vVnQGwlBvzikA",
    "url": "https://services.humanbrainproject.eu/oidc",
    "deactivate": true
  },
  "port": 8000,
  "modelsPath": "$HBP/Models",
  "servers": {
    "localhost": {
      "gzweb": {
        "assets": "http://localhost:8080/assets",
        "nrp-services": "http://localhost:8080",
        "videoStreaming": "http://localhost:8080/webstream/",
        "websocket": "ws://localhost:8080/gzbridge"
      },
      "rosbridge": {
        "websocket": "ws://localhost:8080/rosbridge"
      },
      "serverJobLocation": "local"
    }
  },
  "storage": {
    "type": "FS"
  }
}

ManosAngelidis · 2017-10-12 13:27:12 UTC

Could you try to start a private session in your browser?

And then go to esv-private?

Gaurav_Kumar · 2017-10-12 15:00:36 UTC

That didn’t work either. Still the same “invalid username or password” error. Also, just to clarify, the username and password I am trying are very generic and found in “users” file in ~/.opt/FS_db directory while the “users” file in ~/.opt/nrpStorage/FS_db is empty.

ManosAngelidis · 2017-10-12 15:20:45 UTC

Hi Gaurav,

Normally there should be nothing in the ~/.opt/FS_DB but in ~/.opt/nrpStorage/FS_DB . Try to remove everything relevant to the database under the ~/.opt folder and run configure_nrp again. Then you should get a new user in the users file ,and you can try again. Alternatively copy the users file from ~/.opt/FS_DB to ~/.opt/nrpStorage/FS_DB

Regards,
Manos

Gaurav_Kumar · 2017-10-12 15:25:53 UTC

./update_nrp build all failed with these errors:

Making all in pynest
make[1]: Entering directory '/home/gauku/NRP/nest-simulator/pynest'
cd ../pynest; /usr/bin/python -c \
"from Cython.Build import cythonize; \
cythonize('pynestkernel.pyx', verbose=1)"
Traceback (most recent call last):
  File "<string>", line 1, in <module>
ImportError: No module named Cython.Build
Makefile:974: recipe for target 'pynestkernel.cpp' failed
make[1]: *** [pynestkernel.cpp] Error 1
make[1]: Leaving directory '/home/gauku/NRP/nest-simulator/pynest'
Makefile:551: recipe for target 'all-recursive' failed
make: *** [all-recursive] Error 1
ERROR

Gaurav_Kumar · 2017-10-12 15:44:04 UTC

I was able to login after moving the users file from ~/.opt/FS_DB to ~/.opt/nrpStorage/FS_DB. Before doing that I also tried ./update_nrp build all which failed (errors posted in the reply to @Iguyot ). Anyhow, when I launch an experiment now, it loads all assets without issues (Connection accepted. message on cle-start terminal) but upon starting the simulation, it doesn’t start anything (Terminal output-

Thu Oct 12 2017 17:40:14 GMT+0200 (CEST) Connection accepted.
2017-10-12 17:40:28,597 [uWSGIWorker1] [hbp_nrp_back] [INFO]  starting State Machines...
2017-10-12 17:40:28,598 [/ros_cle_sim] [transitions.] [INFO]  Initiating transition from state paused to state started...
2017-10-12 17:40:28,600 [/ros_cle_sim] [transitions.] [INFO]  Executing callback 'start' before transition.
2017-10-12 17:40:28,601 [/ros_cle_sim] [transitions.] [INFO]  Exited state paused
2017-10-12 17:40:28,601 [/ros_cle_sim] [transitions.] [INFO]  Entered state started
2017-10-12 17:40:28,602 [/ros_cle_sim] [transitions.] [INFO]  Executed callback '<transitions.extensions.locking.LockedMethod instance at 0x7f733dbbd710>' after transition.
[ INFO] [1507822828.627603538, 0.021000000]: waitForService: Service [/gazebo/set_physics_properties] is now available.
python: ../nestkernel/scheduler.h:774: librandom::RngPtr nest::Scheduler::get_rng(nest::thread) const: Assertion `thrd < static_cast< thread >( rng_.size() )' failed.
python: ../nestkernel/scheduler.h:774: librandom::RngPtr nest::Scheduler::get_rng(nest::thread) const: Assertion `thrd < static_cast< thread >( rng_.size() )' failed.
[ INFO] [1507822828.705850582, 0.021000000]: Physics dynamic reconfigure ready.
bash: line 2: 14087 Aborted                 (core dumped) python $HBP/ExDBackend/hbp_nrp_cleserver/hbp_nrp_cleserver/server/ROSCLESimulationFactory.py

)

vonarnim · 2017-10-12 15:45:02 UTC

Did you run the

pip install Cython==0.23.4 mpi4py==2.0.0

step as in documentation in the Install Nest section?

Gaurav_Kumar · 2017-10-12 16:05:27 UTC

Yes I did. Reverified this from .bash_history just now.

vonarnim · 2017-10-12 16:29:03 UTC

then maybe you missed the virtual env stuff?

Gaurav_Kumar · 2017-10-12 16:57:13 UTC

As far as I can remember, I didn’t skip any step. But in the case I did miss it, reinstalling NEST should help, right? And which steps would I need to repeat after installing NEST properly?

vonarnim · 2017-10-13 07:14:25 UTC

Nest is independant from the rest. Just remove the folder completely, and repeat the install nest section.
Then, redo MUSIC, and finally user-script/update_nrp build