MetroCluster Issues
MetroCluster Issues
Highest rated
(rating)- CONTAPEXT-895: ATTO 7600 FibreBridge reboots unexpectedly with error HW Code: L4 Watchdog 0 Warm
- CONTAPEXT-618: ATTO 7500N/7600N FibreBridge might unexpectedly reboot
- CONTAP-444832: Unexpected node reboot during a switch reboot in MetroCluster IP
- CONTAP-554458: ONTAP cluster node reboots unexpectedly in a MetroCluster IP
- CONTAP-644033: RFE: Enforce deletion of sync-destination SVM and all owned objects upon deletion of sync-source SVM even if degraded
- CONTAP-607421: DR CONFIG REPLICATION FAILED: Apply failed for Object: vserverPeer_permission_api_byuuid Method: baseline. Reason: A cluster that is peered with the MetroCluster partner cluster is not peered with the local cluster
- CONTAP-616115: Downgrade from 9.17.1P2 using ANDU fails with waiting for reservations to clear
- CONTAP-469157: Unexpected node reboot during back-end switch firmware upgrade
- CONTAP-514817: Page fault due to dereferencing a NULL pointer
- CONTAP-586103: Cisco Nexus 9336C-FX2 switches logging UNEXPECTED_MCAST_FRAMES for qos-group 5
- CONTAP-111612: HA partner information removed for new platforms
- CONTAPEXT-2854: Node unexpectedly reboots when ATTO Bridge is rebooted
- CONTAP-633165: metrocluster check ends with warning when schedule exists in vserver consists of two characters
- CONTAP-586842: Check for disk-pool-allocation not shown on MCC-IP with no SAS shelves
- CONTAP-44211: LIF Revert is Failing because information for that Vserver is temporarily unavailable on that node. Cluster looks healthy.
Recently updated
(date updated)- CONTAP-634381: NVRAM Log Synchronization May Fail During MetroCluster Interface Changes
- CONTAP-625861: Sync-destination SVMs stopped after takeover whilst in switchover
- CONTAP-659523: ONTAP System Manager does not show Current Node, Current Home and Is Home details for FC LIFS from MetroCluster DR SVMs
- CONTAP-569877: Aggregate renamed after switchover
- CONTAP-811360: WAFL hung on CP after giveback amid concurrent disk firmware flash & shelf firmware flash
- CONTAP-806664: Automatic Switchover May Not Occur During a MetroCluster IP Site Failure
- CONTAP-741301: Flapping on single ISL causing unexpectedly node reboot in MetroCluster IP
- CONTAP-776986: ONTAP MetroCluster IP loses remote disk path
- CONTAP-419128: MCC-IP node changes can cause a system panic during RDMA connection cleanup
- CONTAP-604368: Repeated IncompleteSnmpDataRefreshAlert Call Homes With 64 Gbps Fibre Channel Switch Ports
- CONTAP-366074: Storage Controller Disruption During MetroCluster IP Recovery
- CONTAP-44211: LIF Revert is Failing because information for that Vserver is temporarily unavailable on that node. Cluster looks healthy.
- CONTAP-98045: WAFL hung due to device disk errors
- CONTAP-514944: After an ONTAP upgrade to 9.13.1P17, 9.15.1P13, 9.15.1P14 or 9.16.1P6, MetroCluster DR/NVRAM goes offline
- CONTAP-325108: Unexpected MGWD restart in MCCIP cluster
Recently added
(date created)- CONTAP-811360: WAFL hung on CP after giveback amid concurrent disk firmware flash & shelf firmware flash
- CONTAP-776986: ONTAP MetroCluster IP loses remote disk path
- CONTAP-634381: NVRAM Log Synchronization May Fail During MetroCluster Interface Changes
- CONTAP-806664: Automatic Switchover May Not Occur During a MetroCluster IP Site Failure
- CONTAP-604368: Repeated IncompleteSnmpDataRefreshAlert Call Homes With 64 Gbps Fibre Channel Switch Ports
- CONTAP-366074: Storage Controller Disruption During MetroCluster IP Recovery
- CONTAP-419128: MCC-IP node changes can cause a system panic during RDMA connection cleanup
- CONTAP-805272: MetroCluster: Remote-site volumes still listed after switchback
- CONTAP-581697: MetroCluster IP Statistics Event Describes an Incorrect Collection Frequency
- CONTAP-757733: MGWD crashes whilst backing out of FC to IP transition
- CONTAP-764421: Switchback/Switchover fails due to configuration error
- CONTAP-720454: Node hang during back-end switch maintenance/replacement in MetroCluster IP
- CONTAP-741301: Flapping on single ISL causing unexpectedly node reboot in MetroCluster IP
- CONTAP-635042: Switchover hung and wafl hung panic with mediator disks IOs failing.
- CONTAP-72373: Possible interconnect slowness causing wafl_exempt hung waiting for fs_lock to free the bkt
For Translation
