Investigative information
Please provide the following:
- Timestamp:
started around may 10 20:43 UTC
- Function App name:
prd-modelo-preditivo(staging)
hml-modelo-preditivo(staging)
Repro steps
Provide the steps required to reproduce the problem:
No repro at the moment, but basically since the push out of :
PYTHON_ENABLE_INIT_INDEXING
We have seen an issue that has led to the exceptions during init leading to ten second timeout between host and language worker prevents logging of the exception causing the problem during this step. Which leads to confusion on why a failure may be occuring during startup of the language worker.
Indexing times using old indexing
percentile_AverageTime_30 | percentile_AverageTime_50 | percentile_AverageTime_75 | percentile_AverageTime_90 | percentile_AverageTime_95 | percentile_AverageTime_99
-- | -- | -- | -- | -- | --
16.5 | 16.5 | 16.6 | 21.406 | 21.406 | 21.406
Expected behavior
Metadata call during Indexing should occur while if a timeout happens, it will return the exception that occurred during this time that leads to the timeout and log that exception as well.
Actual behavior
Only a timeout exception is shown
System.TimeoutException : The operation has timed out. at Microsoft.Azure.WebJobs.Script.Grpc.GrpcWorkerChannel.PendingItem.OnTimeout() at /src/azure-functions-host/src/WebJobs.Script.Grpc/Channel/GrpcWorkerChannel.cs : 1764
While the captured exception isn't logged out during indexing.
Known workarounds
Provide a description of any known workarounds.
PYTHON_ENABLE_INIT_INDEXING = 0 in app settings will allow you to see the real exception of the failure
Contents of the requirements.txt file:
Provide the requirements.txt file to help us find out module related issues.
n/a
Related information
Provide any related information
(https://github.com/Azure/azure-functions-python-worker/blob/5bc5d39efffe86da25cacdf7d612a2bc03b32ec3/azure_functions_worker/constants.py#L68C32-L68C59)
ICM # 503543832
Source
-->
Investigative information
Please provide the following:
started around may 10 20:43 UTC
prd-modelo-preditivo(staging)
hml-modelo-preditivo(staging)
Repro steps
Provide the steps required to reproduce the problem:
No repro at the moment, but basically since the push out of :
PYTHON_ENABLE_INIT_INDEXING
We have seen an issue that has led to the exceptions during init leading to ten second timeout between host and language worker prevents logging of the exception causing the problem during this step. Which leads to confusion on why a failure may be occuring during startup of the language worker.
Indexing times using old indexing
percentile_AverageTime_30 | percentile_AverageTime_50 | percentile_AverageTime_75 | percentile_AverageTime_90 | percentile_AverageTime_95 | percentile_AverageTime_99 -- | -- | -- | -- | -- | -- 16.5 | 16.5 | 16.6 | 21.406 | 21.406 | 21.406Expected behavior
Metadata call during Indexing should occur while if a timeout happens, it will return the exception that occurred during this time that leads to the timeout and log that exception as well.
Actual behavior
Only a timeout exception is shown
System.TimeoutException : The operation has timed out. at Microsoft.Azure.WebJobs.Script.Grpc.GrpcWorkerChannel.PendingItem.OnTimeout() at /src/azure-functions-host/src/WebJobs.Script.Grpc/Channel/GrpcWorkerChannel.cs : 1764
While the captured exception isn't logged out during indexing.
Known workarounds
Provide a description of any known workarounds.
PYTHON_ENABLE_INIT_INDEXING = 0 in app settings will allow you to see the real exception of the failure
Contents of the requirements.txt file:
Provide the requirements.txt file to help us find out module related issues.
n/a
Related information
Provide any related information
(https://github.com/Azure/azure-functions-python-worker/blob/5bc5d39efffe86da25cacdf7d612a2bc03b32ec3/azure_functions_worker/constants.py#L68C32-L68C59)
ICM # 503543832
Source