Azurestorageexplorer: Child process not starting in a timely manner

[tehlen]

Storage Explorer Version: 1.2.0
Platform/OS Version: Windows 10
Architecture: x64
Regression From: 1.1.0

Steps to Reproduce:

1. Open a Blob Container

Expected Experience:

Blob Container browser shows contents

Actual Experience:

Browser window closes and error message "The child process failed to start in a timely manner."

[craxal]

If you open the developer tools (F12), do you get any error outputs? If you're working with other Azure storage services (files, tables, etc.), do they encounter the same issue?

[JakeRadMSFT]

We haven't been able to reproduce the issue ... so I don't believe it's version specific but likely something environmental.

If you could also try restarting the machine or making sure all StorageExplorer processes are killed and try again, that might resolve the delayed process start time.

How much available memory does the machine have? Is it a Virtual Machine?

[tehlen]

I'll try to debug more, is there a way to get the old version to unblock the dev team?

[tehlen]

Sorry I missed a previous email with the old version disregard my request. I will do some more debugging and get back to you

[JakeRadMSFT]

Please report back if the previous version works.

Cheers,
Jake

[tehlen]

Enclosed is the log with the errors…

It looks like this happens for all of the types and seems to happen for all people in Chevron, they have rolled back to 1.0.

[tehlen]

The previous version works.

[nicomi]

Exact same problem here with v1.2.0, reproduced on two machines (physical x64 Win 10, and virtual x64 Win2008 R2). Tables also have the problem.
Probably related, Storage Explorer takes a scaringly long time to start (over 1 minute), on a Core i7 with not much else going on (1.5 GB free RAM).

When I click on a blob container, developer tools show ProcessManager: Starting child process with ID 5528, and I see that process in Task Manager. After a few seconds the process disappears and the console shows

Uncaught (in promise) Error: The child process failed to start in a timely manner.
    at AppInsightsAsyncCorrelatedErrorWrapper.ZoneAwareError (C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\zone.js\dist\zone-node.js:811:33)
    at new AppInsightsAsyncCorrelatedErrorWrapper (C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\applicationinsights\out\AutoCollection\CorrelationContextManager.js:172:18)
    at C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\se-process-manager\dist\Host\ForkedProcess.js:47:49
    at ZoneDelegate.invokeTask (C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\zone.js\dist\zone-node.js:275:35)
    at Zone.runTask (C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\zone.js\dist\zone-node.js:151:47)
    at ZoneTask.invoke (C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\zone.js\dist\zone-node.js:345:33)
    at data.args.(anonymous function) (C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\zone.js\dist\zone-node.js:1359:25)

[MRayermannMSFT]

Hello everyone! If you are running into this issue, please attempt the following steps to mitigate:

1. Open "C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\se-process-manager\dist\Host\ForkedProcess.js" in your text editor of choice (if you are using an editor other than VSCode, make sure to run the editor as admin before opening the file)
2. Find ForkedProcess._startupTimeoutInSeconds = 20;
3. Change the 20 to a larger value, something around 60 or even 120 is probably a safe place to start.

If the issue is mitigated, awesome! If not, you can continue trying to up the value until the process has enough time to start. And if you do end up doing that, please comment on this issue with what value you ended up using. We will be increasing the value in our next release, so if you needed a very high value, we would love to know.

Thanks!

[MRayermannMSFT]

Merged in to rel/1.3.0, will be shipped with 1.3.0. We went with a 90 second timeout. Hopefully that will be enough.

[JoaoRaimundo14]

Still get the same error

[MRayermannMSFT]

@JoaoRaimundo14 , can you try the instructions on setting the timeout period, and do a longer value? I'm curious as to how long your machine needs. I did not expect anyone to need 90 seconds (or longer).

[JoaoRaimundo14]

@MRayermannMSFT I've tested 90/120/150 and i still get the same error

[donkz36]

Have tried up to 240 and am still getting the same error.

[MRayermannMSFT]

If y'all open task manager and watch the group of Storage Explorer processes as you open the tab, do you at least see a new process appear? You'll probably need to sort by name:

Additionally, what other software do y'all have running on your machine? Any anti-virus or monitoring software?

[gitmichalk]

Hi all,
I have this issue but my version is 1.4.1

So far nothing help.

Any advice?

[Helen2011]

Just installed 1.4.1, increased ForkedProcess._startupTimeoutInSeconds, but no luck, any advice?

[PandaShoots]

Okay I've still got this issue in 1.4.2 on Windows 10. It looks like the child process is timing out after 20 seconds even though the default ForkedProcess._startupTimeoutInSeconds is set to 90 seconds in the config file and bumping it up to say 300 makes no difference.

[jLynx]

I'm having this issue on Windows 10 Enterprise

[mafer09]

Issue has not been resolved. Still encountering this issue while using version 1.4.2. Increased ForkedProcess._startupTimeoutInSeconds and nothing

[gitmichalk]

Hello all victims :)

For workaround I suggest start using Cloud Explorer in Visual Studio (like I did - if they don't know how to resolve this problem). When you login into yours account on Azure you should have access for all Storage Accounts where you can open Blobs or tables you want without any problems.

references:
https://docs.microsoft.com/en-us/azure/vs-azure-tools-resources-managing-with-cloud-explorer

Kind Regards,
Michal

[craxal]

Hey folks! Something that might help us narrow this down: try launching Storage Explorer from the command line and observe the output. If the child processes are working correctly, you should see some output from them in the command line. It might get messy, but you're looking for words like "child", "process", and "message". You should also see process IDs.

[Caldina777]

I can't believe we are up to version 1.5 of Storage explorer and this issue is STILL happening, it has been happening since version 0.9,7 (I thankfully kept an old install of 0.9.6).

[jinglouMSFT]

@Caldina777 There are many root causes to run into this issue. We haven't been able to repro this issue in house. Have you tried with these steps to mitigate the issue?

• Open "C:\Program Files (x86)\Microsoft Azure Storage Explorer\resources\app\node_modules\se-process-manager\dist\Host\ForkedProcess.js" in your text editor of choice (if you are using an editor other than VSCode, make sure to run the editor as admin before opening the file)
• Find ForkedProcess._startupTimeoutInSeconds = 20;
• Change the 20 to a larger value, something around 60 or even 120 is probably a safe place to start.

This is to help extend the timeout.

Please let us know the findings. Thanks.

[gdamyanova-fourth]

Ver. 1.6.2 - still has it. ForkedProcess._startupTimeoutInSeconds = 240; doesn't help. What is the oldest version that doesn't manifest this behavior? Can I have a link, please? I think that it's really inconvenient that no older versions are available.

[jinglouMSFT]

We are working on capturing additional logs when this happens to help us further diagnose the root cause. If bumping up the timeout value doesn't help you, feel free to use a version that works for you. You can find our previous release here on Github under the "Releases" tab. We have all releases there since our GA.

[gdamyanova-fourth]

Thanks jinglou!
Downloaded ver. 1.0.0 and no issues so far.

[jaysheelkalgal]

Hi,

I am using version 1.6.2 on Mac on Mojave 10.14.3. I get the below error when trying to expand the blob container or file share.

[craxal]

@jaysheelkalgal, this doesn't appear to be related to the child process issue this thread is about. Have you followed our troubleshooting guide? If you need assistance, we invite you to submit a service request. If you believe this behavior is a bug, you many open a new issue.

[jaysheelkalgal]

I have a storage account name and key. I am trying to connect using the account name and key method. I dont have any proxy settings on my Mac. I am able to connect but when I try to expand the Blob container I get this error. Is there any other troubleshooting I should try ?

[jinglouMSFT]

Couple of things to verify: 1. Is this a primary name/key or secondary name/key? We don't work with secondary name/key. 2. Do you know what permissions do you have?

[jaysheelkalgal]

I would need to get those details from our Admin team. Probably in a day or two.

[MRayermannMSFT]

Closing due to lack of activity.