Customer replaced hardware of tape drive.
Then he ran 'rmdev' for unloading tape device, system went crash.
OS support team analyzed the dump and found it caused by function from SEOS_syscall.
OS Support provide following crash information:
SYSTEM_CONFIGURATION:
CHRP_SMP_PCI POWER_PC POWER_8 machine with 8 available CPU(s) (64-bit registers)
SYSTEM STATUS:
sysname... AIX
nodename.. HOSTX
release... 2
version... 7
build date Mar 2 2018
build time 13:02:46
...
CRASH INFORMATION:
CPU 0 CSA F00000002FF47600 at time of crash, error code for LEDs: 30000000
pvthread+091800 STACK:
[F1000000C0701D74]SEOS_syscall:seos_sysconfig_pre_sc+0003B4 (F00000002FF47510, F00000002FF475C0,
F1000B01501FDBD0, F00000002FF47488)
[0091B18C]ld_intercept_sc+00044C (0000000000000004, 000000002FF21C20,
000000000000000C, 000000000000F032, 0000000000003B9C, 0000000000000000,
000000005056BC00, 0000000000000000)
[00003930]syscall+000228 ()
[kdb_get_virtual_memory] no real storage @ 2FF21BE8
[D08D0274]D08D0274 ()
[kdb_read_mem] no real storage @ FFFFFFFFFFF8E20
OS: AIX 7.1/7.2
Prod: CA Privileged Access Manager Server Control r14.0/14.1 for Endpoint
The root cause of this issue is a logical problem in sysconfig system call interception;
the data structure used to store all incoming module load and unload request had an issue while building and retrieving information.