Here is a crash we got from a customer. The subhandle obj_ops point to NULLs causing the function call "entry->sub_handle->obj_ops.close(entry->sub_handle)" crash. Appreciate if this kind of defect is fixed or known to be in Ganesha2.5 release.

Here is the backtrace:

(gdb) bt

#0 0x00007f8d8a2a154b in raise (sig=11) at ../nptl/sysdeps/unix/sysv/linux/pt-raise.c:37

#1 0x00000000004548cc in crash_handler (signo=11, info=0x7f8d86f04eb0, ctx=0x7f8d86f04d80)

at /usr/src/debug/nfs-ganesha-2.5.3-ibm021.00-0.1.1-Source/MainNFSD/nfs_init.c:225

#2 <signal handler called>

#3 0x0000000000000000 in ?? ()

#4 0x0000000000540fd4 in mdcache_close (obj_hdl=0x7f8a6c075718)

at /usr/src/debug/nfs-ganesha-2.5.3-ibm021.00-0.1.1-Source/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_file.c:441

#5 0x00000000005307c4 in fsal_close (obj_hdl=0x7f8a6c075718)

at /usr/src/debug/nfs-ganesha-2.5.3-ibm021.00-0.1.1-Source/include/fsal.h:432

#6 0x0000000000533fbf in lru_run_lane (lane=87, totalclosed=0x7f8d86f05460)

at /usr/src/debug/nfs-ganesha-2.5.3-ibm021.00-0.1.1-Source/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_lru.c:1146

#7 0x00000000005348ac in lru_run (ctx=0x145d5b0)

at /usr/src/debug/nfs-ganesha-2.5.3-ibm021.00-0.1.1-Source/FSAL/Stackable_FSALs/FSAL_MDCACHE/mdcache_lru.c:1328

#8 0x000000000050a5d0 in fridgethr_start_routine (arg=0x145d5b0)

at /usr/src/debug/nfs-ganesha-2.5.3-ibm021.00-0.1.1-Source/support/fridgethr.c:550

#9 0x00007f8d8a299dd5 in start_thread (arg=0x7f8d86f06700) at pthread_create.c:308

#10 0x00007f8d89954b3d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:113