1630 errors when using XIV storage system with release V6.3

Resolving The Problem

Abstract

IBM SAN Volume Controller or Storwize V7000 for Lenovo systems running V6.3, and IBM Storwize V7000 Unified systems running 1.3 may experience Cluster Error 1630 when restarting a node or a node canister, including during software upgrades.

Symptom

There is a timing window that exists during a node (or node canister) restart, where some of the fibre channel logins between the node (or node canister) and the backend storage system may not be configured correctly. Therefore, these ports are not available for I/O. When this error is detected, the software logs a Cluster Error 1630.

This problem should not cause any loss of access from host systems. However, this problem will cause the system to report the XIV Storage System as degraded.

This issue can be encountered only during a node restart. Discovery of Fibre Channel logins when the node or node canister is online in the system is not affected by this issue.

IBM SAN Volume Controller and Storwize V7000 for Lenovo systems running V6.2 and earlier levels are not affected by this issue.

Workaround

If this issue is encountered, customers should disable and re-enable the XIV ports one at a time, to restart the discovery process, leaving at least 60 seconds between each port. Wherever possible, perform this action plan on the least busy XIV port first, and the busiest XIV port last. The fix procedure for the 1630 error should then be run to clear any remaining 1630 errors in the event log.

Solution

This issue was addressed by APAR IC81571 in the V6.3.0.2 PTF release of the IBM SAN Volume Controller and Storwize V7000 for Lenovo software. This issue will be addressed in a future release of the IBM Storwize V7000 Unified software.

Affected configurations

Cross-reference information