starting a volume failed with VxVM vxvol ERROR V-5-1-10128 Unexpected kernel error in configuration update

Article:TECH181697  |  Created: 2012-02-17  |  Updated: 2012-07-28  |  Article URL http://www.symantec.com/docs/TECH181697
Article Type
Technical Solution


Environment

Issue



Cannot start a volume


Error



starting a volume failed with VxVM vxvol ERROR V-5-1-10128 Unexpected kernel error in configuration update


Cause



# vxplex -g dg off plex-02
# vxvol -g dg start vol
failed with "no clean plex on site"
siteconsitent is on for the diskgroup
# vxdg list dg |grep site
# vxdg -g dg set siteconsistent=off
# vxvo -g dg start vol
# fsck -t vxfs /dev/vx/rdsk/dg/vol
log reply all good
test mount
# mount -t vxfs /dev/vx/dsk/dg/vol /tmnt
all good
restarted servive group
all mounted successfully
recreate dco and restart mirroring
# vxsnap -g dg prepare vol
# vxplex -g dg on plex-02
# vxrecover -g dg -sb volname
# vxtask -l list
/.....smartmove was used by default


Solution



use vxtune
# vxtune volpagemod_max_memsz 1024M
# vxtune |grep volpa

Tunable Value Description
----------------- -------- ---------------
volpagemod_max_memsz 1048576 Cache Size for FMR Metadata (KBytes)


I’m now getting the following error when starting the volume.

VxVM vxio V-0-0-0 Memory allocated for paging module is 1048576
VxVM vxio V-5-3-456 Error whileaccessing FMR metadata of volume DB-billingdb01datavol

different but when i called in the
volpagemod_max_memsz 286576 Cache Size for FMR Metadata (KBytes) (256M)

chaning the tunable between 1g 2g 4g failed to make any difference but now the error was always
[root@greenzobdb01 ~]# vxvol -g DB-billingdb01dg start DB-billingdb01datavol
VxVM vxvol ERROR V-5-1-10128 Unexpected kernel error in configuration update

# effect dg and volume
vxprint -ht g dg

sr R1 ACTIVE
sr R3 ACTIVE

dm DB-billing01dg01 tagmastore-usp0_4710 auto 65536 2147310624 -
dm DB-billing01dg02 tagmastore-usp1_4710 auto 65536 2147310624 -

\v DB-billingdb01datavol - DISABLED ACTIVE 2143289344 SITEREAD - fsgen

checking the regionsz we have
# vxprint -g DB-billingdb01dg -F%regionsz DB-billingdb01datavol_dco
128
volume szie = 2143289344 sectors
using this formulafor tunable we have
volpagemod_max_memsz >= max(vol_size/region_size) * 3 / 8 * 2
6132M
to large something is wrong'
decided the dco was corrupt and to recreate dco
# vxsnap -g dg -f unprepare volname
line one plex




Article URL http://www.symantec.com/docs/TECH181697


Terms of use for this information are found in Legal Notices