XtremIO X2: Storage Controller Unexpectedly Reboots with XIOS S/W version 6.3.0-62

A Storage Controller kernel panic condition may occur resulting in a sudden Storage Controller reboot with XtremIO X2 clusters running XIOS S/W version 6.3.0-62.

This issue may impact cluster performance due to the Storage Controller reboot. In extreme cases, if two (or more) Storage Controllers of the cluster are simultaneously impacted by this issue, then cluster service may be impacted as well.

This issue is currently under investigation by XtremIO Engineering. This article will be updated once further details are available on possible triggers for this issue and steps to avoid it.

A race condition in the InfiniBand (IB) driver running on the Storage Controller may lead to a kernel panic of the affected Storage Controller.

This race condition is not applicable with XIOS versions earlier than 6.3.0-62.

While it is known that this race condition can occur without a specific trigger, it is known that an IPMI password change is a possible trigger for this issue.

An investigation is in progress for other possible triggers for this issue.

Resolution:
A permanent fix for this issue is in XIOS version 6.3.1-5 (or later). To avoid this issue if using (or considering to use) XIOS version 6.3.0-62, it is highly recommended to upgrade to version 6.3.1-5 (or later) instead. Contact Dell Global Tech Support to schedule a cluster software upgrade (NDU) to this version. Specify Dell KB# 541398 (this article) to expedite the processing of this request.

Dell is soon taking proactive measures to upgrade affected clusters to XtremIO XIOS 6.3.1-5 (or later).

XMS version 6.3.0-62 is still recommended for customers that need the new features, or have experienced an issue which is resolved in this XMS release.