After creating a compute instance under Microsoft Azure Machine Learning and select Jupyter to open Jupyter Notebook, I am selecting a New terminal from the menu. However, I am getting the following errors:
terminado.js:4 WebSocket connection to 'wss://xalapa.eastus2.instances.azureml.ms/terminals/websocket/2' failed: Error during WebSocket handshake: Unexpected response code: 426
make_terminal # terminado.js:4
index.js:5 Uncaught TypeError: Cannot read property 'parentElement' of undefined
at proposeGeometry (index.js:5)
at fit (index.js:30)
at Terminal.terminalConstructor.fit (index.js:44)
at window.onresize (main.js:54)
I have recreated the compute instance with no luck.
Do you know what is the issue related?
Thanks
Looks like you might have your websockets blocked. Here is the troubleshooting doc: https://learn.microsoft.com/en-us/azure/machine-learning/how-to-run-jupyter-notebooks#troubleshooting
If you can't connect to a notebook, ensure that web socket
communication is not disabled. For compute instance Jupyter
functionality to work, web socket communication must be enabled.
Please ensure your network allows websocket connections to
*.instances.azureml.net and *.instances.azureml.ms.
Looks like you cannot connect to your compute instance, and 426 means the server refuses to connect.
Could you try to open Jupyter / JupyterLab from the same compute instance?
Related
When I am trying to load a query into a tabular dateset (from a devops docker image) I will get the following error:
raise DatasetValidationError(error_message + '\n' + str(e), e)
azureml.data.dataset_error_handling.DatasetValidationError: Cannot load any data from the datastore using the SQL query "<azureml.data.datapath.DataPath object at 0x>". Please make sure the datastore and query is correct.
Error Code: ScriptExecution.DatabaseConnection.Unexpected
Failed Step: 9ad57100-4870-49d2-a32f-1c9c15c244e0
Error Message: ScriptExecutionException was caused by DatabaseConnectionException.
DatabaseConnectionException was caused by UnexpectedException.
'MSSQL' encountered unexpected exception of type 'InvalidOperationException' with HResult 'x80131509' while opening connection.
Internal connection fatal error.
I believe that I have allowed the connection in firewall (I might not have done it quite right).
I don't get the error when I am running it from the notebook (on the compute instance).
Is the code that is running on the compute instance the exact same as the docker image? If so, it's likely that the firewall is causing connectivity issues. One way of checking this is to (be cautious) disable the firewall and see if the connectivity issue is gone.
If you're comfortable sharing the firewall/network configurations, that will also help troubleshoot this.
I need to connect a NodeJS to AS/400 server. In order to do that, I installed NodeJS in AS/400 with IBM documentation and tryed (succesfully) to send and receive data with Class iDataQueues (https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/IBM%20i%20Technology%20Updates/page/Toolkit%20for%20i%20APIs?section=Class%20iDataQueue) between AS/400 server and my develope computer. My next step was to run a program (Class iPgm) to received some data.
I'm running my test with
/path/to/ibm/node/installation/node /home/test/app.js"
and I'm getting a "node[537]: pthread_create: Resource temporarily unavailable" error, but I have no idea what is the problem:
Server has enough resources
There are not any node running in server
There are not any changes in server config between iDataQueue and iPgm tests
I have root privileges
I think my code is not the problem because I dont' even run it ¬¬'
This is caused by running node in QSH. By default QSH does not support running threads, so any attempt to create a thread will result in that "Resource temporarily unavailable" message. This can fixed by setting the environment variable QIBM_MULTI_THREADED=Y prior to running QSH, using QP2TERM, or using SSH.
I was trying Azure Machine Learning Services following this tutorial (Link). After successfully creating the Azure Machine Learning services accounts, I successfully installed the Workbench on my Windows 10 Laptop (Behind Proxy; Proxy has been configured at the WorkBench). Next, I was trying to create project following this section (Link). Once I click on the Create button, it goes to "Creating" state and stays there for ever. The errors displayed at Errors.log is the following. Any suggestion will be appreciated.
[2018-07-09 09:47:08.437] [ERROR] HttpService - {"event":"HttpService","task":"Failed","data":{"url":"http://localhost:54240/projects/v1.0/create/template","status":500,"statusText":"INKApi Error","jsonError":null,"requestId":null,"sessionType":"Workbench"},"sid":"365395c0-832b-11e8-b4ce-e5d7046c6143"}
[2018-07-09 09:47:08.960] [ERROR] CreateProjectForm - {"event":"CreateProject","task":"Error","data":{"_body":null,"status":500,"ok":false,"statusText":"INKApi Error","headers":{"Date":["Mon"," 09 Jul 2018 04:17:06 GMT"],"Via":["1.1 localhost.localdomain"],"Proxy-Connection":["close"],"Content-Length":["0"],"Content-Type":["text/html"]},"type":2,"url":"http://localhost:54240/projects/v1.0/create/template"},"sid":"365395c0-832b-11e8-b4ce-e5d7046c6143"}
[2018-07-09 09:47:08.963] [FATAL] ExceptionLogger - {"event":"exception","task":"","data":{"message":"Cannot read property 'error' of null","name":"TypeError","stack":"TypeError: Cannot read property 'error' of null\n at SafeSubscriber._error (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:61476:58)\n at SafeSubscriber.__tryOrUnsub (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:212279:20)\n at SafeSubscriber.error (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:212241:30)\n at Subscriber._error (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:212172:30)\n at Subscriber.error (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:212146:22)\n at MergeMapSubscriber.OuterSubscriber.notifyError (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:210968:30)\n at InnerSubscriber._error (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:211072:25)\n at InnerSubscriber.Subscriber.error (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:212146:22)\n at DeferSubscriber.OuterSubscriber.notifyError (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:210968:30)\n at InnerSubscriber._error (file:///C:/Users/MyUser/AppData/Local/AmlWorkbench/resources/app.asar/src/App/main.bundle.js:211072:25)"},"sid":"365395c0-832b-11e8-b4ce-e5d7046c6143"}
It was happening because of the Proxy (although I have configured the Proxy on the Workbench). When I am connected to internet directly, everything works fine (Able to create project, train, compare models etc). However the Workbench should return meaningful error instead of hanging or simply waiting while creating the project.
We have a redis cache hosted on Azure and we use the following as a service in our .net core application:
services.AddSingleton<IConnectionMultiplexer>(provider =>
ConnectionMultiplexer.Connect(Configuration.GetConnectionString("RedisConnection"))
);
Our connection string is similar to the following:
ourserver.redis.cache.windows.net:6380,password=******=,ssl=True,abortConnect=False,syncTimeout=4000
in getting from cache we don't face problems but if we add to cache we sometimes get the following error:
"StackExchange.Redis.RedisConnectionException: It was not possible to
connect to the redis server(s); to create a disconnected multiplexer,
disable AbortOnConnectFail. ConnectTimeout"
I have read solutions like setting abortconnect to false and increasing the timeout but I have implemented it like the connection string above.
If you have suggestions or you think that we have a problem with the connection string kindly advise.
Update: we have implemented a solution to solve connectivity issue but now we get this:
StackExchange.Redis.RedisConnectionException: No connection is available to service this operation
What is your connectTimeout set to? Can you try increasing that to see if it helps?
I'm trying my best to learn Google's Cloud Platform. They have a CloudSQL offering, which I'm learning via this NodeJS tutorial. Everything worked great until I deployed to their appspot server, at which point I got the following error:
connect ECONNREFUSED 127.0.0.1:3306
I've looked all through the NodeJS project and don't see anything in it or the Cloud Console that is referencing localhost or 127.0.0.1. Googling the error hasn't helped thus far. Any ideas?
I couldn't get this thing fixed when running on server but using this files I was able to read/write from local and production, now im using this connection strings in my own app
https://cloud.google.com/appengine/docs/flexible/nodejs/using-cloud-sql
https://github.com/GoogleCloudPlatform/nodejs-docs-samples/tree/master/appengine/cloudsql
I had a similar issue when deploying the nodejs sample app 2-structured-data
The reason why the error occurred is that the NODE_ENV environment variable was not passed to the config file that is used to check if node should use a socket for connecting to mysql
You can fix it by adding 'NODE_ENV' in the file config.js :
.env([
...
'NODE_ENV'])