shrivardhan92 opened a new issue #10652:
URL: https://github.com/apache/druid/issues/10652


   Segments present in historical local cache (Segment cache) but not 
queryable. 
   
   ### Affected Version 
   0.19.0
   
   ### Description
   
   - Cluster size: 2 routers, 4 brokers, 2 overlords, 2 coordinators, 6 
historicals, 20 middlemanagers (m5.2xlarge)
   Background: After a successful ingestion using index_parallel, I find that 
for a particular interval (2020-11-18T17:00:00.000Z_2020-11-18T18:00:00.000Z) 
the UI shows that the segments are used, published and available [Figure 
1](https://10281656745555498052.googlegroups.com/attach/22ac56a37f2d7/image.png?part=0.1&view=1&vt=ANaJVrGFndy3N-npfLDDqeE48xWdFLklbEyJPWpjnFddNFsDZTQGQugMrsA-Y4qW9zkbHbsWmBXMBK4SMrHjVJ3IusEHdpjZje9LJPX00c6toMKDSZwhyu4).
   
   Issue: Although the segments are available I am not able to query data in 
this interval and it returns empty results. The native segment metadata query 
for this interval also returns no data. I dug into the historical logs to find 
interesting events but found none. I'm curious as to what causes this behavior, 
as it is a potential data loss.
   
   Here's a log that shows 
[logs](https://gist.github.com/shrivardhan92/6205640fab3a12442d76532bbf254d66) 
on different druid processed for a particular segment that belongs to the 
interval in question.
   
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to