Graphs not being created / updated all of a sudden

[yetdog]

All of a sudden, as of 3:40pm yesterday afternoon, after weeks of graphing - my graphs have stopped updating. Throwing CACTID into debug mode doesn't show any CACTI2RRD entries (other than WEBLOG initiated entries from viewing within the browser) - so for whatever reason, it seems that poller.php is no longer calling rrdtool during its polling process.

Cacti Version - 0.8.6i
Plugin Architecture - 1.0
Poller Type - Cactid v0.8.6i
Server Info - Linux 2.6.9-42.0.3.ELsmp
Web Server - Apache/2.0.52 (CentOS)
PHP - 4.3.9
PHP Extensions - yp, xml, wddx, tokenizer, sysvshm, sysvsem, standard, sockets, shmop, session, pspell, posix, pcre, overload, mime_magic, iconv, gmp, gettext, ftp, exif, dio, dbx, dba, curl, ctype, calendar, bz2, bcmath, zlib, openssl, apache2handler, gd, mysql, snmp, eAccelerator
MySQL - 4.1.20
RRDTool - 1.2.15
SNMP - 5.1.2
Plugins

• Thresholds (thold - v0.3.0)
  Network Tools (tools - v0.2)
  Syslog Monitoring (haloe - v0.4)
  mactrack
  RRD Cleaner (rrdclean - v1.1)
  Update Checker (update - v0.3)
  FlowView (flowview - v0.3)
  Host Info (hostinfo - v0.1)
  Error Images (errorimage - v0.1)
  Report Creator (reports - v0.1b)
  Dataquery (dataquery - v0.1.2a)

Anyone ever run into this before? I have made no system-level changes to the box. CactiEZ 0.2 install.

[rjones237]

Having the same problem since updating my Cacti to 0.8.6j last week. Doing this has got my rrds updating again(found in post http://forums.cacti.net/about8441-0-asc-15.html):

mysql -u root -p cacti
> select count(*) from poller_output;
> truncate table poller_output;
> quit

[yetdog]

Will try that, but now I have another problem since rebooting:

02/06/2007 10:30:01 AM - CACTID: Poller[0] WARNING: Falling back to UDP Ping due to not running asroot. Please use "chmod xxx0 /usr/bin/cactid" to resolve.
02/06/2007 10:30:01 AM - CACTID: Poller[0] DEBUG: The UDP Ping return_code was -1, errno was 111, total_time was 1073.1220
02/06/2007 10:30:01 AM - CACTID: Poller[0] Host[2] PING Result: UDP: Host is Alive
02/06/2007 10:30:01 AM - CACTID: Poller[0] Host[2] SNMP Result: SNMP not performed due to setting or ping result
02/06/2007 10:30:01 AM - CACTID: Poller[0] DEBUG: MySQL Insert ID '1': 'update host set status='3', status_event_count='0', status_fail_date='2007-02-06 09:55:00', status_rec_date='2007-01-12 09:42:02', status_last_error='Host did not respond to SNMP, UDP: Cannot connect', min_time='0.355000', max_time='2519.025000', cur_time='0.536560', avg_time='17.562300', total_polls='37322', failed_polls='318', availability='99.1480' where id='2''
02/06/2007 10:30:01 AM - CACTID: Poller[0] DEBUG: MySQL Insert ID '1': OK


For basically ALL of my hosts...oddly enough, my cactid path isn't even /usr/bin/cactid, and cacti knows the proper path so I don't know why it's referencing that path!

[yetdog]

Yes rjones, truncating that table DID indeed allow the poller to call CACTI2RRD again - and all my graphs are "updating" - I say that in quotes because I still believe they aren't being fed any data due to the bolded lines I'm showing in my previous post - but thank you for helping me solve that one issue!

I wonder why that is..

[yetdog]

Hmm - that error might be moot. My graphs have started "graphing" again! LOL rjones, thank you very much. Perhaps that cactid warning is not important?

[TheWitness]

Now that I know UNIX/Linux half ass, the chmod is not accurate.

TheWitness

Best way to avoid this problem is to keep PHP memory high, or update to 0.8.6j.

TheWitness