Ticket #9049 (closed defect: fixed)
|Reported by:||joe42||Owned by:|
During a normal working day, suddenly all our VMs appeared to spontaneously restart. It turned out that VBoxSvc had crashed and that all the VBoxHeadless processes had therefore powered off.
VBoxSvc left behind a core dump. A 'pstack core' reports this interesting trace:
----------------- lwp# 22 / thread# 22 -------------------- fffffd7ffe748dfb _smalloc () + 3b fffffd7ffe74906d _malloc_unlocked () + 1ad fffffd7ffe748e8d malloc () + 2d fffffd7ffe9ac8bb _Znwm () + 2b fffffd7ffe3227fc _Z15IPC_SendMessagejRK4nsIDPKhj () + ec fffffd7ffe32af39 _ZN18ipcDConnectService7OnSetupEjPK13DConnectSetupj () + 8e7 fffffd7ffe32a637 _ZN18ipcDConnectService17OnIncomingRequestEjPK10DConnectOpj () + 2d fffffd7ffe328d91 _ZN14DConnectWorker3RunEv () + 211 fffffd7ffeb6334e _ZN8nsThread4MainEPv () + 2e fffffd7ffeb942ea ???????? () fffffd7ffeb943af ???????? () fffffd7ffed7332c rtThreadMain () + 2c fffffd7ffedbcf91 ???????? () fffffd7ffe7abfbb _thr_setup () + 5b fffffd7ffe7ac1e0 _lwp_start ()
All other threads are in lwp_park() so I assume they are not terribly interesting.
Attached is a log file from one of the VBoxHeadless processes where this happened - the log only shows that VBoxSvc appeared to disappear.
Also attached is the core file from VBoxSvc.