What’s the “sfcb-smx” error?

This error is related to HPE smx-providers for the Smart Array controller MRA and any HPE ProLiant G8 or G9 server will be affected if VMware ESXi has been deployed on the server by using HPE VMware ESXi 6.5 (Nov2016) ISO image.

NOTE: As result of this issue, HPE Insight Management WBEM Providers are restarted by the sfcbd when they fail; therefore, the inventory information will be available. However, when a provider crashes and is re-started, the sfcbd may not resend all the indication subscriptions and the clients listening will not get any indications if the Smart Array hardware has issues.

You may also faced with the below lines in vmkernel.log file:

sfcb-smx: wantCoreDump:sfcb-smx signal:11 exitCode:0 coredump:enabled
sfcb-smx: Dumping cartel 684037 (from world 684042) to file /var/core/sfcb-smx-zdump.000 …
sfcb-smx: Userworld(sfcb-smx) coredump complete.

What’s Solution?

There is just a solution and administrators should upgrade their ESXi hosts by HPE VMware ESXi 6.5 (July 2017) custom ISO image.

To fix the impact of not getting indications after a provider crashes, restart the HPE Insight Management WBEM Providers, by performing any of the following procedures:

• Run the command “/etc/init.d/sfcbd-watchdog restart”.
• Use the console to restart the management agents.
• Use vCenter to restart the management agents.

References:

https://www.hpe.com/us/en/servers/hpe-esxi.html

https://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2148526