Details
-
Bug
-
Resolution: Won't Fix
-
Critical
-
2.0-beta
-
Security Level: Public
-
build 1808
Description
4 node cluster, 2 buckets...50 million items each (30% Active 13% Replica ratio)
Running 300 queries per bucket with active load of 5k op/sec.
UI alerts eventually pop up (attached): Write Commit Failure. Disk write failed for item in Bucket "default" on node 10.6.2.68.
Also attaching atop logs, seeing rsize of beam.smp to be 1.2GB (600s sample collection)