Slow lun access with high latency from Network subsystem
Applies to
- ONTAP 9
- Fibre Channel LUNs
Issue
- Multiple ESXI Datastore are failing and going offline, either they show as inaccessible, or they are extremely slow when data access.
- High network latency reported on the storage during the issue
- End-to-End connectivity was checked and no issues reported on the Fabric end.
- Multiple low memory errors seen in EMS during the time frame of the issue:
Sat Jan 27 09:31:02 +1000 [cluster02: mpcleaner_thread1: wafl.memory.statusVeryLowMemory:alert]: WAFL is running very low on memory, with 2171MB remaining.
Sat Jan27 09:36:02 +1000 [cluster02: wafl_exempt04: wafl.memory.statusVeryLowMemory:alert]: WAFL is running very low on memory, with 2111MB remaining.
Sat Jan 27 09:41:03 +1000 [cluster02: wafl_exempt10: wafl.memory.statusVeryLowMemory:alert]: WAFL is running very low on memory, with 2219MB remaining.