Rotate Log File failed: Could not find file 'C:\Users\_sa_da_PBIEgService\AppData\Local\Microsoft\o

[detlev]

Last two weeks I see new (never occurred before) errors in the eventlog of On-premises data gateway service.

The error seems to refer to a write or read accessissue to a new logfile. The error seems intermittent.  Some logfiles are created.

Gateway20180529.000000001.log is created.

Gateway20180529.000000000.log does not exist

Who knows what’s happening?

(We are keen of having our logfiles being updated due to other errors we are trying to fix. Restaring the windowsservice results in writing the logfiles again. But it is not really preferable to restart our productieopn gateway for this ;-))

"Rotate Log File failed:
Could not find file 'C:\Users\_sa_da_PBIEgService\AppData\Local\Microsoft\On-premises data gateway\Gateway20180529.000000000.log'. 
StackTrace:
at System.IO.__Error.WinIOError(Int32 errorCode, String maybeFullPath)
at System.IO.FileStream.Init(String path, FileMode mode, FileAccess access, Int32 rights, Boolean useRights, FileShare share, Int32 bufferSize, FileOptions options, SECURITY_ATTRIBUTES secAttrs, String msgPath, Boolean bFromProxy, Boolean useLongPath, Boolean checkHost)
at System.IO.FileStream..ctor(String path, FileMode mode, FileAccess access, FileShare share, Int32 bufferSize, FileOptions options, String msgPath, Boolean bFromProxy)
at System.IO.FileStream..ctor(String path, FileMode mode)
at Microsoft.PowerBI.DataMovement.Pipeline.Common.RotatableFileStream.RotateIfNeeded()"

[v-qiuyu-msft]

Hi @detlev,

I have sent a email to consult this issue internally, will update here once I get any information. 

Best Regards,
Qiuyun Yu

[v-qiuyu-msft]

Hi @detlev,

I got information below:

Event log is our backup storage for traces when log file rotation failed, and the next write after the failure will try to rotate again.

Rotation error message is still in the old gateway logs. So please ask customer to send us the gateway log after this issue occurs again.

Regarding customer’s concern about losing traces, we always write to eventlog in case file based tracing is down. So no trace will be lost.

Best Regards,
Qiuyun Yu

[detlev]

Hi Qiuyun Yu, many thanks for taking the time to look at our case.

Which part of the event log is used as backup storage?  I cant find them and a few days of logging is lost due to the rotation faillure.

In the meantime I deleted (and backuped) all old logs and restarted the service and the roptationissue did not occur again.

I kept every log which was logged, so wich part (timing) do you like to see to deep dive in this rotation faillure?

kind regards,

Detlev

[v-qiuyu-msft]

Hi @detlev,

I have consulted this followed questions internally, will update here once I get information. 

Best Regards,
Qiuyun Yu