error after upgrade cacti 0.8.6.j to 0.8.7

[fernandakp]

Hi i have the following error in my log after upgrading cacti. I do have also graphs with a lot of brakes in the graph.


11/06/2007 11:35:03 AM - POLLER: Poller[0] WARNING: Poller[0] Host[1] SNMP Time was not numeric
11/06/2007 11:35:01 AM - POLLER: Poller[0] WARNING: Poller Output Table not Empty. Potential Data Source Issues for Data Sources: 5min_cpu(DS[972]), 5min_cpu(DS[976]), int_status(DS[979]), 5min_cpu(DS[980]), int_status(DS[983]), 5min_cpu(DS[984]), int_status(DS[987]), 5min_cpu(DS[988]), int_status(DS[991]), 5min_cpu(DS[992]), 5min_cpu(DS[1006]), 5min_cpu(DS[1021]), int_status(DS[1024]), 5min_cpu(DS[1025]), int_status(DS[1028]), 5min_cpu(DS[1029]), int_status(DS[1325]), 5min_cpu(DS[1326]), int_status(DS[1329]), 5min_cpu(DS[1330]), int_status(DS[1333]), 5min_cpu(DS[1334]), int_status(DS[1337]), 5min_cpu(DS[1338]), 5min_cpu(DS[1342]), int_status(DS[1345]), 5min_cpu(DS[1346]), int_status(DS[1349]), traffic_in(DS[1912]), traffic_out(DS[1912]), traffic_in(DS[1913]), traffic_out(DS[1913]), traffic_in(DS[1914]), traffic_out(DS[1914]), traffic_in(DS[1915]), traffic_out(DS[1915]), traffic_in(DS[1916]), traffic_out(DS[1916]), traffic_in(DS[1917]), traffic_out(DS[1917]), traffic_in(DS[1918]), traffic_out(DS[1918]), traffic_in(DS[1919]), traffic_out(DS[1919]), traffic_in(DS[1920]), traffic_out(DS[1920]), traffic_in(DS[1921]), traffic_out(DS[1921]), traffic_in(DS[1922]), traffic_out(DS[1922]), traffic_in(DS[1923]), traffic_out(DS[1923]), traffic_in(DS[1924]), traffic_out(DS[1924]), traffic_in(DS[1925]), traffic_out(DS[1925]), traffic_in(DS[1926]), traffic_out(DS[1926]), traffic_in(DS[1927]), traffic_out(DS[1927]), traffic_in(DS[1928]), traffic_out(DS[1928]), traffic_in(DS[1929]), traffic_out(DS[1929]), traffic_in(DS[1930]), traffic_out(DS[1930]), traffic_in(DS[1931]), traffic_out(DS[1931]), traffic_in(DS[1932]), traffic_out(DS[1932]), traffic_in(DS[1933]), traffic_out(DS[1933]), traffic_in(DS[1938]), traffic_out(DS[1938]), traffic_in(DS[1939]), traffic_out(DS[1939]), traffic_in(DS[1942]), traffic_out(DS[1942]), traffic_in(DS[1943]), traffic_out(DS[1943]), traffic_in(DS[1950]), traffic_out(DS[1950]), traffic_in(DS[1951]), traffic_out(DS[1951]), traffic_in(DS[1952]), traffic_out(DS[1952]), traffic_in(DS[1953]), traffic_out(DS[1953]), traffic_in(DS[1954]), traffic_out(DS[1954]), traffic_in(DS[1955]), traffic_out(DS[1955]), traffic_in(DS[1956]), traffic_out(DS[1956]), traffic_in(DS[1957]), traffic_out(DS[1957]), traffic_in(DS[1958]), traffic_out(DS[1958]), traffic_in(DS[1959]), traffic_out(DS[1959]), traffic_in(DS[1960]), traffic_out(DS[1960]), traffic_in(DS[1961]), traffic_out(DS[1961]), traffic_in(DS[2127]), traffic_out(DS[2127]), traffic_in(DS[2128]), traffic_out(DS[2128]), traffic_in(DS[2129]), traffic_out(DS[2129]), traffic_in(DS[2130]), traffic_out(DS[2130]), traffic_in(DS[2131]), traffic_out(DS[2131]), traffic_in(DS[2140]), traffic_out(DS[2140]), traffic_in(DS[2141]), traffic_out(DS[2141]), traffic_in(DS[2142]), traffic_out(DS[2142]), traffic_in(DS[2144]), traffic_out(DS[2144]), traffic_in(DS[2154]), traffic_out(DS[2154]), traffic_in(DS[2157]), traffic_out(DS[2157]), traffic_in(DS[2158]), traffic_in(DS[2159]), traffic_in(DS[2160]), traffic_in(DS[2161]), traffic_out(DS[2161]), traffic_in(DS[2162]), traffic_out(DS[2162]), traffic_in(DS[2401]), traffic_out(DS[2401]), traffic_in(DS[2402]), traffic_out(DS[2402]), traffic_in(DS[2412]), traffic_out(DS[2412]), traffic_in(DS[2507]), traffic_out(DS[2507]), traffic_in(DS[2508]), traffic_out(DS[2508]), traffic_in(DS[2509]), traffic_out(DS[2509])

[TheWitness]

Make sure your cont interval is 5 mnutes right now. This is a bug. Were you running with a cron interval of something less? It could also be a double cron issue as well.

TheWitness

[fernandakp]

i have cacti cont interval with 5 minuts, but i keep the fast_poller of monitor with 1 min.
my crontab is listed below

#
# check scripts in cron.hourly, cron.daily, cron.weekly, and cron.monthly
#
-*/15 * * * * root test -x /usr/lib/cron/run-crons && /usr/lib/cron/run-crons >/dev/null 2>&1
*/5 * * * * user_cacti php /srv/www/htdocs/cacti/poller.php > /dev/null 2>&1
*/1 * * * * php /srv/www/htdocs/cacti/plugins/monitor/fastpoller_old/fast_poller.php > /dev/null 2>&1


should I remove something? should i remove the fast_poller?

[fernandakp]

The fast_poller i have is the old one, that one we have for cacti 0.8.6.j. This because the new version of monitor for cacti 0.8.7 does not come with the fast_poller.php...

*/1 * * * * php /srv/www/htdocs/cacti/plugins/monitor/fastpoller_old/fast_poller.php > /dev/null 2>&1

[TheWitness]

Nor does the newest version require it, I think anyway. Send a PM to Jimmy.

TheWitness

[fernandakp]

I commented the line with fast_poller and the problem continues...

Nor does the newest version require it, I think anyway. Send a PM to Jimmy.

TheWitness

[fernandakp]

i receive the folowing warning in the log:


11/06/2007 03:45:08 PM - POLLER: Poller[0] WARNING: Cron is out of sync with the Poller Interval! The Poller Interval is '300' seconds, with a maximum of a '300' second Cron, but 3906 seconds have passed since the last poll!


what do i do?

[TheWitness]

Until 0.8.7a, edit poller.php and set the cron interval to 300 and whack anything that attempts to set it. This will be fixed by late Moday evening the 12th.

TheWitness

[fernandakp]

thanks a lot. It worked!!
And congratulations for the new version. I apreciate a lot the fix that you guys made for correcting the code to accept the interface status graph data with the spine.
I havent migrated to cactid just because it dont work with interface status graphs...Now, when i upgrade to cacti 0.8.7 and spine it work all right.
Thanks. Besides, the duration of the process is much lower then before.
Good work!!

[fernandakp]

Hi, is there a patch for this bug already?

[TheWitness]

Cacti 0.8.7a. Make sure you review the new setting in Settings->Poller called "Cron Interval".

TheWitness