Qlik Sense April 2018 file corrupt error

[agigliotti]

Hello,

I just upgraded with success a Qlik Sense Site from February to April 2018 version.

Now I'm facing the attached error while opening the app.

below the engine log error:

35 20.4.2.0 20180521T105940.839+0200 ERROR UMBQLIK-S 9fb6321d-f254-4bd2-8eec-04ea4048027d 20180521T105940.834+0200 12.145.3.0 Command=Open app;Result=3002;ResultText=Error: File corrupted c2c77482-db0f-4fa1-8f4a-4f779ac34dfe 287aa595-f68b-4431-8a8a-6ac23fdc2061 1 UMBRA qlikadmin 59a029ba-0aba-4f67-8181-154df7b3f0be 435719e7-a55d-4312-82c5-dce8508a091e Not available Engine Not available Global::OpenApp Open app 3002 435719e7-a55d-4312-82c5-dce8508a091e 9fb6321d-f254-4bd2-8eec-04ea4048027d

Other apps are working with no error.

Where I should look for some error logs ?

What could be the reason for this issue and how can I fix it ?

tko‌

mto‌

Thanks in advance for your time.

Best Regards

Andrea

[Chanty4u]

is this desktop or server? 

try the same app in other system ?

or you have back up file replace the same and try

[agigliotti]

is qlik sense server.

With April version I noted a strange behaviour after restarting the server, "the Qlik Server Engine Service" was not started automatically, I must start it manually.

Also last friday I tried duplicating the published app and opening with no data.

After that I reloaded data and published it again.

Finally the published app was opening with no error.

Today I got corrupted file error.

the app size is about 275MB.

I have a scheduled task with the app that runs hourly.

[ToniKautto]

You log snippet indicates a file corruption and a missing engine, so your observations appear to be inline with the log entries. Do you still have a problem, are did you manage to resolve it?

I'm not sure why the service would not start, but a possibility could be that the Windows default timeout is not long enough. Have you looked at extending the default time out? See Microsoft KB as reference http://support.microsoft.com/kb/839803/

Opening the app without data and resolving the issue with a reload, indicate that the data portion of the app was corrupt. If this is only the case for one app, perhaps it is related to a non Qlik event corrupting the file?

The app files would be migrated during upgrade, so if you see the issue in multiple apps or can replicate it I would suggest reaching out to Qlik Support on that part also.

[agigliotti]

Regarding the Qlik Sense Engine service they told me it's a bug that has to be solved.

Actually I get this error only for one app.

I just opened a new case in Support Portal this morning.

I'm waiting for a feedback asap.

Many thanks for your collaboration.

Andrea

[mphowlett]

Hi Andrea,

did you get this resolved or get any update? I upgraded from February to April 2018 over the weekend and all seemed fine but now I have the same scenario as you where a published application has become corrupt. I actually followed your steps and was able to duplicate from the QMC and open without data and reload. When I published to the stream and overwrote the existing corrupt application it did not resolve the issue and I still get the error message. However, if I publish to the stream as a new application it works fine.

Unfortunately we have 10's of community sheets and objects in the published version of the application that will be lost if I cannot "overwrite" the corrupt application successfully. I thought I would check with you but I am going to raise a ticket tomorrow.

Thanks,

Matt

[agigliotti]

Hi Matt,

unfortunately Qlik Support did not solve the problem yet.

which windows server operating system are you using with Qlik Sense Server April 2018 ?

i'll let you know next updates.

Regards

Andrea

[mphowlett]

Hi Andrea,

I've just raised a case with Qlik - The version of the application that was working has now also become corrupt.

I was investigating last night and found that if I give myself root admin in the Hub context, I can open the original corrupt application without data if I add "/noData/true" to the URL of the published application:

<yourserverURL>/sense/app/<yourPublishedAppID>/noData/true

I could then reload the published application directly without issue. This fixed the corruption and maintained all the community and user sheets/objects.

However...since setting the hourly reload task to enabled again last night, it has become corrupt again, some time in the last 9 hours. I have done the same as above and can very quickly fix the published application again but it seems it will always become corrupted again when reloading with the scheduler.

Waiting on Qlik now.

Thanks,

Matt

[agigliotti]

I followed your advice but I can't reload the published app after loading it with no data as you said above.

could you let me know ?

Thanks

Andrea

[mphowlett]

Hi,

did you give yourself root admin in the Hub? We have security rule that we can quickly enable for the whole team as below:

I take it the URL worked but you couldn't edit the script? I just checked and we also have SecurityAdmin so not sure if that is a factor...

Windows Server 2012 R2 Standard

Thanks,

Matt