vxconfigd core dumped due to disk.info file corruption

Article: 100018508
Last Published: 2023-10-18
Ratings: 0 0
Product(s): InfoScale & Storage Foundation

Problem

vxconfigd core dumped due to disk.info file corruption

 

Error Message

Error: VxVM vxconfigd ERROR V-5-1-0 Segmentation violation - core dumped

 

Solution

Troubleshooting steps:

1.) Debug vxconfigd:

# vxconfigd -x 6 -k -x log -x logfile=/tmp/vxconfigd.out

The last few lines of the debug output should look like this:

VxVM vxconfigd DEBUG V-5-1-0 assign_disk_unique_native_name: path not found with such devno
VxVM vxconfigd DEBUG V-5-1-0 assign_disk_unique_native_name: looking for path with devno 0x50fc98
VxVM vxconfigd DEBUG V-5-1-0 assign_disk_unique_native_name: path found with such devno
VxVM vxconfigd ERROR V-5-1-0 Segmentation violation - core dumped


2.) Run pstack on the core:

# pstack ./<vxconfigd.0.1.25982.pssna045.core>
core'./vxconfigd.0.1.25982.pssna045.core' of 25982: 

vxconfigd-k
----------------- lwp# 1 /thread# 1 --------------------
ff14f904 strrchr (0, ffbeec14, 0, 22b430, 50c5e0, ffffffff) +4
0010c614 assign_disk_unique_native_name (0, ccd7, 3335c, 21b51c, 30b590, 0) +2c4
0010b79c assign_disk_native_name (4422e0, 22b3e8, 4, 1, 21b51c, 5126e0) +268
0010c294 assign_disk_local_name (50003, 22b400, 505f78, 0, 4, 45c0010) +804
00108068 dmp_process_buf (5126a0, 3e65f0, ffbeeca4, 50000, 22b0d8, 21b51c) +6e4
0011a0f4 dmp_make_mplist (2569b0, 50003, 50000, 239358, 21b51c, 512650) + 10c 0011af90dmp_reconfig_db (21b51c, 0, 22274c, 0, 502f40, 4f) + 9c 000fb340ddl_find_devices_in_system (1f6958, 4e, 21b51c, 0, 0, ffbeedc8) +350
000bbc24 find_devices_in_system (2226c8, ffbef278, 0, 2, 22884c, 0) +38
000bb214 mode_set (2, 220c00,234778, 50003, 1400, 228810) + 18c
000b43d4 setup_mode (2, 1e3944, 5, 1e39a4, 20000, 0) + 1c 000b4330startup (220800, 220800, 657d, ffbef650, 1e3590, 1e357c) +27c
000b3d34 main (256800, ffbef5ac, 1e357c, 22874c, 0, ffbef6cc) +c08
0002d920 _start (0, 0, 0,0, 0, 0) + b8

If the output from the above troubleshooting appears similar to the examples above, and the "assign_disk_unique_native_name" is present, this indicates there is a problem with the /etc/vx/disk.info file. 

The solution for this problem is to move /etc/vx/disk.info and restart vxconfigd. Below are the commands that need to be run to correct this situation:

# cd /etc/vx/
# mv disk.info save.disk.info
# vxconfigd -k

 

Was this content helpful?