Everything works except Load Average after patch upgrade

[neilr]

Hi,

I downloaded and installed the following patches and addon:
wget http://www.cacti.net/downloads/patches/ ... zoom.patch
374 patch -p1 -N < graph_properties_zoom.patch
375 wget http://www.cacti.net/downloads/patches/ ... auth.patch
376 patch -p1 -N < script_server_snmp_auth.patch
377 wget http://www.cacti.net/downloads/patches/ ... ding.patch

Addon: thold2.1a_cacti-0.8.6g.tar.gz

Since this upgrade, all the Processor Load Average graphs have stopped working.

When I run the manual snmp walk it returns data fine. When I check the log file I see this:
tail -1000 cacti.log |more
958 /usr/bin/rrdtool update /var/www/html/cacti/rra/sandbox_load_15min_1120.rrd
959 /usr/bin/rrdtool update /var/www/html/cacti/rra/sandbox_load_15min_1120.rrd --template
960 /usr/bin/rrdtool update /var/www/html/cacti/rra/sandbox_load_15min_1120.rrd --template load_15min
961 /usr/bin/rrdtool update /var/www/html/cacti/rra/sandbox_load_15min_1120.rrd --template load_15min 1132686001:U
962 /usr/bin/rrdtool update /var/www/html/cacti/rra/sandbox_load_15min_1120.rrd --template load_15min 1132686002:U
963 /usr/bin/rrdtool update /var/www/html/cacti/rra/sandbox_load_15min_1120.rrd --template load_15min 1132686003:U
964 /usr/bin/rrdtool update /var/www/html/cacti/rra/sandbox_load_15min_1120.rrd --template load_15min 1132686004:U
965 tail -1000 cacti.log |more
966 /usr/bin/rrdtool graph - --imgformat=PNG --start=-86400 --end=-300 --title="sandbox
967 - Load Average" --rigid --base=1000 --height=120 --width=500 --alt-autoscale-max --lower-limit=0 --units-exponent=0 --vertical-label
968 ="processes in the run queue" DEF:a="/var/www/html/cacti/rra/sandbox_load_1min_1119.rrd":load_1min:AVERAGE DEF:b="/var/www/html/cacti/rra/s
969 andbox_load_5min_1121.rrd":load_5min:AVERAGE DEF:c="/var/www/html/cacti/rra/sandbox_load_15min_1120.rrd":load_15min:AVERAGE CDEF:cdefg=TIME
970 ,1132685867,GT,a,a,UN,0,a,IF,IF,TIME,1132685867,GT,b,b,UN,0,b,IF,IF,TIME,1132685867,GT,c,c,UN,0,c,IF,IF,+,+ AREA:a#EACC00:"1 Minute Average"
971 GPRINTLAST:" Current\:%8.2lf\n" STACK:b#EA8F00:"5 Minute Average" GPRINTLAST:" Current\:%8.2lf\n" STACK:c#FF0000:"15 Minute Av
972 erage" GPRINT:c:LAST:"Current\:%8.2lf\n" LINE1:cdefg#000000:"Total"
973 /usr/bin/rrdtool graph - --imgformat=PNG --start=-86400 --end=-300 --title="sandbox
974 - Load Average" --rigid --base=1000 --height=120 --width=500 --alt-autoscale-max --lower-limit=0 --units-exponent=0 --vertical-label
975 ="processes in the run queue" DEF:a="/var/www/html/cacti/rra/sandbox_load_1min_1119.rrd":load_1min:AVERAGE DEF:b="/var/www/html/cacti/rra/s
976 andbox_load_5min_1121.rrd":load_5min:AVERAGE DEF:c="/var/www/html/cacti/rra/sandbox_load_15min_1120.rrd":load_15min:AVERAGE CDEF:cdefg=TIME
977 ,1132685867,GT,a,a,UN,0,a,IF,IF,TIME,1132685867,GT,b,b,UN,0,b,IF,IF,TIME,1132685867,GT,c,c,UN,0,c,IF,IF,+,+ AREA:a#EACC00:"1 Minute Average"
978 GPRINTLAST:" Current\:%8.2lf\n" STACK:b#EA8F00:"5 Minute Average" GPRINTLAST:" Current\:%8.2lf\n" STACK:c#FF0000:"15 Minute Av
979 erage" GPRINT:c:LAST:"Current\:%8.2lf\n" LINE1:cdefg#000000:"Total"



All my graphs are showing "nan"

However when I do a snmpwalk manually I see the follwing:
snmpwalk -c xxxxx -v2c sandbox .1.3.6.1.4.1.2021.10.1.3.3
UCD-SNMP-MIB::laLoad.3 = STRING: 0.09

Any help or ideas would be greatly appreciated!

Thanks,
--neil

[gandalf]

replaced community string for sake of security
Reinhard

[flundy]

Same issue. Upgraded from .86c w/ cmd.php to .86f w/ cactid .86f. All graphs work except load avg.

My rrd's are being populated w/ nan:

Code: Select all

rrdtool fetch rra/<host>_load_1min_1566.rrd AVERAGE
                    load_1min

1133300400: nan
1133300700: nan
1133301000: nan
...

But I can snmpget the data:

Code: Select all

snmpget -c <community> -v 2c <host> .1.3.6.1.4.1.2021.10.1.3.1
UCD-SNMP-MIB::laLoad.1 = STRING: 0.11

I am also getting these in my DEBUG poller output:

Code: Select all

OK u:0.04 s:0.13 r:6.84
11/30/2005 04:55:08 PM - POLLER: Poller[0] CACTI2RRD: /usr/bin/rrdtool update /var/www/html/cacti/rra/<host>_traffic_in_1575.rrd --template traffic_in:traffic_out 1133387701:3651368011:2069308367
OK u:0.04 s:0.13 r:6.84
11/30/2005 04:55:08 PM - POLLER: Poller[0] CACTI2RRD: /usr/bin/rrdtool update /var/www/html/cacti/rra/<host>_load_1min_1566.rrd --template load_1min 1133387701:U
OK u:0.04 s:0.13 r:6.84

Note the "U" instead of a number for the load_1min rrd.

Any thoughts why load avg is special compared to the other graphs (cpu, disk, capacity, network...)?

Thanks

[flundy]

NET-SNMP version: 5.0.9
PHP 4.3.2
cacti 0.86f
RRDtool 1.0.41
Apache/2.0.46
mysqld Ver 4.1.7-standard for pc-linux on i686 (Official MySQL RPM)

[flundy]

I downloaded the cactid .86f the last week of November... My CACTI is all patched up w/ the official patches-- are there patches or updates for CACTID that I am missing? I would love to get this working again.

[neilr]

I downloaded the cactid .86f the last week of November... My CACTI is all patched up w/ the official patches-- are there patches or updates for CACTID that I am missing? I would love to get this working again.

Good question, no one seems to care I guess. Unless I posted this thread in the wrong place??

[flundy]

Switched back to .086f's cmd.php poller. Load numbers seem to have come back. Performance went south though-- before (cactid), 78 hosts-5 processes-25 threads took < 10 seconds. After (cmd.php) takes ~70 seconds.

On another note, an official cactid .86g appears to be on its way...
http://forums.cacti.net/viewtopic.php?t ... c&start=30