Details
-
Technical task
-
Resolution: Unresolved
-
Major
-
5.0.0
-
Security Level: Public
-
None
Description
As part of other disk failure/failover discussions, we agreed that it may be too hard to determine what "too high" means for a disk write queue because it varies so much from application to application.
This is meant to mitigate that by providing a lighter-weight alert when the disk write queue reaches over 1M for any node as usually indicative of an issue but not necessarily (much like our metadata warnings)