Back on 6/19, one of our inputs for Microsoft Office 365 Reporting Add-on for Splunk stopped working. We actually have two - one of them is still working fine, but one of them no matter what we try/do it just won't resume indexing.
This is the only seemingly relevant error I can see in the _internal logs:
2019-07-02 14:09:02,846 DEBUG pid=107942 tid=MainThread file=connectionpool.py:_make_request:400 | https://reports.office365.com:443 "GET /ecp/reportingwebservice/reporting.svc/MessageTrace?$filter=StartDate%20eq%20datetime'2019-06-19T15:34:27.739334Z'%20and%20EndDate%20eq%20datetime'2019-06-19T20:34:27.739334Z' HTTP/1.1" 200 216
I found another thread that suggested this Error 400 can be if you have a start date set too far back - well I tried removing the start date all together and it still won't resume indexing anything.
Any suggestions here? I'm hesitant to blow up the app and start over because one of the Message Trace inputs is still working fine. We didn't make any changes to this app or anything at the time it stopped.
Have you tried restarting Splunk on the DCN and deleting the checkpoint file? I have had the issue and this is the only way I can recover. I am looking at migrating this to one of the methods listed here:
When you say disabled on the DCN, what does that mean? We got the input to start working again by creating it new from scratch and then deleting the old one. But it hung up again shortly after and is stuck on 7/1.
Where is the checkpoint file that you are deleting? Would be curious to see if we can try that instead of deleting and recreating from scratch again.
Would also like to know where this checkpoint file is located? Have been having ongoing issues, and would like to delete it. A restart or recreation of the inputs file doesn't work for me.
Unless its under $SPLUNK_HOME/var/lib/splunk/modinputs
Then I have nothing inside my dir.