Cisco Bug: CSCun45306 - DN2:P2:Running SPECJVM stress test, system got rebooted
Mar 08, 2018
- Cisco Unified Computing System
Known Affected Releases
Symptom: Unexpected system reboot with RHEL 6.5 due to frequent bmc-watchdog driver timeouts. Conditions: This issue is caused by RHEL 6.5 due to the faulty IPMI driver. The bmc-watchdog daemon in RHEL 6.5 will be enabled when OS boot. This daemon does not detect OS IPMI driver since it failed to acess "/dev/ipmi0",so it will create system IO port (ca2/ca3) to communicate with BMC, however, the action creating system IO port will conflict with built-in OS ipmi driver cause built-in OS ipmi driver is already created this IO port to communicate with BMC. Therefore, we found this bmc-watchdog will sometimes receive the wrong data that should response to OS ipmi driver, and this bmc-watchdog will stop running itself because the data is not correct. And then, the timer of BMC watchdog will count down to 0 and the corresponding hard reset action and event log will be done by BMC.
Bug details contain sensitive information and therefore require a Cisco.com account to be viewed.
Bug Details Include
- Full Description (including symptoms, conditions and workarounds)
- Known Fixed Releases
- Related Community Discussions
- Number of Related Support Cases