Your clue is in this line: > DISK CRITICAL - free space: /data02b 0 MB (0% inode=99%);| > /data02b=2147483647MB;10874142;11096064;0;11096064
That "2147483647" number should look suspicious to you (being exactly 2^^31 − 1). It's your evidence that check_disk encountered some sort of integer-overflow condition when sensing the disk space. Quite possibly, the underlying disk driver returned such a number when asked about the amount of space in the partition. Given that the two partitions in question both had 27873280 MB overall capacity (i.e., 28542238720 KB), and none of the other partitions approach anywhere near this size, I'd say that check_disk has trouble dealing with such a large partition. 64-bit arithmetic will be required throughout to handle this case. The only way to debug this for sure is to dig into the check_disk code and instrument it to print out what it gets back from the OS when asking for the disk information, and then trace the arithmetic operations in check_disk itself. And, look for evidence of 32-bit arithmetic within both results requested for disk sizing from the OS and for subsequent calculations within check_disk. On Tue, 3/31/15, Phan, Thao-Chi X <thao-chi.p...@allina.com> wrote: > Please explain to me why the /usr/local/groundwork/nagios/libexec/check_disk > doesn't report the correct disk space free on 2 mount points /data02b and > /data02a on a server. > > root@abc2/usr/local/groundwork/nagios/libexec# df -m > Filesystem MB blocks Free %Used Iused %Iused Mounted on > /dev/hd4 512.00 336.67 35% 5761 3% / > /dev/hd2 5120.00 2051.53 60% 44310 4% /usr > /dev/hd9var 512.00 228.29 56% 26189 20% /var > /dev/hd3 3072.00 2973.20 4% 101 1% /tmp > /dev/hd1 7168.00 4794.46 34% 1854 1% /home > /proc - - - - - /proc > /dev/hd10opt 1024.00 320.47 69% 9826 4% /opt > /dev/locallv 2048.00 1136.73 45% 720 1% /usr/local > /dev/hd11admin 128.00 123.93 4% 20 1% /admin > /dev/oralv 143360.00 100961.11 30% 203739 1% /oracle > /dev/data01lv 276224.00 124619.55 55% 86 1% /data01 > /dev/eplv 256.00 253.61 1% 6 1% /epic > /dev/clarlv 303104.00 89595.34 71% 67974 1% /epic/clarity > /dev/ftplv 512.00 507.60 1% 4 1% /ftpdir > /dev/dbkuplv 97792.00 67809.48 31% 3748 1% /dbkup > /dev/data03lv 215040.00 32499.46 85% 89 1% /data03 > /dev/ctmaglv 5120.00 4587.26 11% 3301 1% /ctmag > /dev/dump_cp 5120.00 5098.90 1% 4 1% /var/adm/ras/dump_cp > /dev/lv00 27873280.00 11204341.16 60% 880 1% /data02b > /dev/lv02 5120.00 5027.61 2% 12 1% /data04b > /dev/lv01 46080.00 16327.36 65% 14 1% /oraredob > /dev/lv04a 27873280.00 2550106.27 91% 880 1% /data02a > /dev/lv06a 5120.00 5027.61 2% 12 1% /data04a > /dev/lv05a 46080.00 16327.36 65% 14 1% /oraredoa > > -------INCORRECT --- 0 MB is not correct. 0% free is not correct. > root@abc2 /usr/local/groundwork/nagios/libexec# ./check_disk -w 2% -c 0% -p > /data02b > DISK CRITICAL - free space: /data02b 0 MB (0% inode=99%);| > /data02b=2147483647MB;10874142;11096064;0;11096064 > > -------INCORRECT ---- 22% free is not correct > root@abc2 /usr/local/groundwork/nagios/libexec# ./check_disk -w 2% -c 0% -p > /data02a > DISK OK - free space: /data02a 2550106 MB (22% inode=99%);| > /data02a=8545957MB;10874142;11096064;0;11096064 > > -------CORRECT > root@abc2 /usr/local/groundwork/nagios/libexec# ./check_disk -w 2% -c 0% -p > /data04a > DISK OK - free space: /data04a 5027 MB (98% inode=99%);| > /data04a=92MB;5017;5120;0;5120 > > -------CORRECT > root@abc2 /usr/local/groundwork/nagios/libexec# ./check_disk -w 2% -c 0% -p > /oracle > DISK OK - free space: /oracle 100961 MB (70% inode=99%);| > /oracle=42398MB;140492;143360;0;143360 > > Thanks a lot. > > Thao-Chi
