I apologize if this is the wrong place to report this.. it seemed more
like a developer issue than a user issue.
There appear to be some situations where nrpe doesn't send all of the
output it receives from the scripts it invokes to the network. I'm
invoking "sar"..
command[sar]=/usr/bin/sar -Brcquwv -n SOCK 1 1|grep Average
and I think that strace shows that it receives different information..
7271 read(4, "Average: proc/s\nAverage: 0.00\nAverage:
cswch/s\nAverage: 483.17\nAverage: CPU %user
%nice %system %iowait %steal %idle\nAverage: all
0.00 0.00 0.25 0.00 0.00 99.75\nAverage:
pgpgin/s pgpgout/s fault/s majflt/s pgfree/s pgscank/s pgscand/s
pgsteal/s %vmeff\nAverage: 0.00 0.00 12.87
0.00 18.81 0.00 0.00 0.00 0.00\nAverage:
kbmemfree kbmemused %memused kbbuffers kbcached kbswpfree kbswpused
%swpused kbswpcad\nAverage: 150172 3931940 96.32 24652
3660292 8189344 27884 0.34 48\nAverage:
dentunusd file-nr inode-nr pty-nr\nAverage: 75489
576 81846 2\nAverage: totsck tcpsck udpsck
rawsck ip-frag tcp-tw\nAverage: 41 18 4
0 0 5\nAverage: runq-sz plist-sz ldavg-1
ldavg-5 ldavg-15\nAverage: 1 182 0.00
0.00 0.00\n", 2047) = 1028
than it outputs...
7271 send(5, "\0\2\0\2\215&X\332\0\0Average: proc/s\nAverage:
0.00\nAverage: cswch/s\nAverage: 483.17\nAverage:
CPU %user %nice %system %iowait %steal
%idle\nAverage: all 0.00 0.00 0.25 0.00
0.00 99.75\nAverage: pgpgin/s pgpgout/s fault/s majflt/s
pgfree/s pgscank/s pgscand/s pgsteal/s %vmeff\nAverage: 0.00
0.00 12.87 0.00 18.81 0.00 0.00 0.00
0.00\nAverage: kbmemfree kbmemused %memused kbbuffers
kbcached kbswpfree kbswpused %swpused kbswpcad\nAverage: 150172
3931940 96.32 24652 3660292 8189344 27884 0.34
48\nAverage: dentunusd file-nr inode-nr
pty-nr\nAverage: 75489 576 81846 2\nAverage:
totsck tcpsck udpsck rawsck ip-frag tcp-tw\nAverage:
41 18 4 0 0
5\nAverage: runq-sz plist-sz ldavg-1 ldavg-5
ldavg-15\nAverage: 1 182 0.00 0.00
\0k:", 1036, 0) = 1036
If I issue the sysstat command on the command line, I see the output I
expect.. and looking at strace, I see that it's getting the correct
output from sar, but it's not passing on the output in it's entirety to
the server requesting it via check_nrpe. I don't think I'm running
into the size of MAX_INPUT_BUFFER.
I can provide the complete strace if that'd help. This is on an LFS
based linux box running 2.6.21.3 (i686), nrpe 2.10 (compiled with
3.4.5), and sysstat 8.0.2, running check_nrpe both remotely and locally.
Liam
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]