If you see a message in AIX like this:
LABEL: DMPCHK_TOOSMALL
IDENTIFIER: E87EF1BE
Date/Time: Sun Jun 1 12:00:00 GMT+00:00 2012
Sequence Number: 182200
Machine Id: XXXXXXXXXXXX
Node Id: db03
Class: O
Type: PEND
WPAR: Global
Resource Name: dumpcheck
Description
The largest dump device is too small.
Then you need to resize your dump devices. First check the devices which are used as dump devices:
# sysdumpdev -l
primary /dev/lg_dumplv
secondary /dev/lg_dumplv2
copy directory /var/adm/ras
forced copy flag TRUE
always allow dump FALSE
dump compression ON
type of dump traditional
The above output shows /dev/lg_dumplv and /dev/lg_dumplv2 as dump devices. Next check the estimated dump size which is needed:
# sysdumpdev -e
0453-041 Estimated dump size in bytes: 9665773568
Dividing 9665773568 bytes three time by 1024 shows the correct size in GB:
# echo 9665773568/1024/1024/1024 | bc -l
9.00195312500000000000
According to the above output 9GB is needed. To resize the dump devices you need the size of your PP's:
# lsvg rootvg
...
VG STATE: active PP SIZE: 512 megabyte(s)
...
512MB per PP, keep that value in mind. Now begin to resize the dump devices. First check the current size for your dump devices:
# lsvg -l rootvg
rootvg:
LV NAME TYPE LPs PPs PVs LV STATE MOUNT POINT
...
lg_dumplv sysdump 8 8 1 open/syncd N/A
lg_dumplv2 sysdump 8 8 1 open/syncd N/A
...
Each dump device is 8 PP's in size, that means 8PP's multiplied by 512MB equals 4096MB (or 4GB), but 9GB are needed. In this case 5GB are needed - or at least 10 PP's. To extend the dump devices use extendlv:
# extendlv lg_dumplv 12
# extendlv lg_dumplv2 12
I used 12 PP's instead of only 10 PP's to have a little space left. Now check the sizes of the dump devices again:
# lsvg -l rootvg | grep lg_dumplv
lg_dumplv sysdump 20 20 1 open/syncd N/A
lg_dumplv2 sysdump 20 20 2 open/syncd N/A
20PP's multiplied by 512MB equals 10GB which is enough for 9GB.
No comments:
Post a Comment