Missing Data in Graphs 1.1.38

[xq1xq1xq1]

I am running Cacti 1.1.38-3 on Centos 7.5

I have been having issues since pretty much the first installation of Cacti on this server with holes in my data.

This is an example of my graph data:

Code: Select all

2019-02-09 03:30:00	19,149	19,149	537	537	19,149	-536.7047
2019-02-09 03:35:00	19,149	19,149	537	537	19,149	-536.7047
2019-02-09 03:40:00	14,156	14,156	524	524	14,156	-523.699
2019-02-09 03:45:00	-	-	-	-	-	-
2019-02-09 03:50:00	-	-	-	-	-	-
2019-02-09 03:55:00	-	-	-	-	-	-
2019-02-09 04:00:00	-	-	-	-	-	-
2019-02-09 04:05:00	298,028	298,028	561	561	298,028	-561.1444
2019-02-09 04:10:00	196,816	196,816	556	556	196,816	-555.5956
2019-02-09 04:15:00	196,816	196,816	556	556	196,816	-555.5956
2019-02-09 04:20:00	-	-	-	-	-	-
2019-02-09 04:25:00	-	-	-	-	-	-
2019-02-09 04:30:00	-	-	-	-	-	-
2019-02-09 04:35:00	-	-	-	-	-	-
2019-02-09 04:40:00	-	-	-	-	-	-
2019-02-09 04:45:00	-	-	-	-	-	-
2019-02-09 04:50:00	-	-	-	-	-	-
2019-02-09 04:55:00	-	-	-	-	-	-
2019-02-09 05:00:00	-	-	-	-	-	-
2019-02-09 05:05:00	-	-	-	-	-	-
2019-02-09 05:10:00	-	-	-	-	-	-
2019-02-09 05:15:00	-	-	-	-	-	-
2019-02-09 05:20:00	-	-	-	-	-	-
2019-02-09 05:25:00	-	-	-	-	-	-
2019-02-09 05:30:00	-	-	-	-	-	-
2019-02-09 05:35:00	-	-	-	-	-	-
2019-02-09 05:40:00	-	-	-	-	-	-
2019-02-09 05:45:00	-	-	-	-	-	-
2019-02-09 05:50:00	-	-	-	-	-	-
2019-02-09 05:55:00	-	-	-	-	-	-
2019-02-09 06:00:00	-	-	-	-	-	-
2019-02-09 06:05:00	-	-	-	-	-	-
2019-02-09 06:10:00	11,237	11,237	545	545	11,237	-544.5216
2019-02-09 06:15:00	10,813	10,813	446	446	10,813	-445.8933
2019-02-09 06:20:00	8,758	8,758	528	528	8,758	-527.5672
2019-02-09 06:25:00	8,758	8,758	528	528	8,758	-527.5672
2019-02-09 06:30:00	4,828	4,828	525	525	4,828	-524.6663
2019-02-09 06:35:00	-	-	-	-	-	-
2019-02-09 06:40:00	-	-	-	-	-	-
2019-02-09 06:45:00	-	-	-	-	-	-
2019-02-09 06:50:00	10,169	10,169	570	570	10,169	-570.3995
2019-02-09 06:55:00	14,809	14,809	558	558	14,809	-557.7301
2019-02-09 07:00:00	179,672	179,672	476	476	179,672	-476.1035
2019-02-09 07:05:00	179,672	179,672	476	476	179,672	-476.1035
2019-02-09 07:10:00	-	-	-	-	-	-

In the logs it seems to work sometimes:

Code: Select all

2019/02/09 07:09:30 - POLLER: Poller[Main Poller] Device[12] Device[UBNT] RECACHE DQ[Net-SNMP - Get Device I/O] OID: .1.3.6.1.2.1.1.3.0, output: 158442636
2019/02/09 07:09:30 - POLLER: Poller[Main Poller] Device[12] Device[UBNT] RECACHE DQ[SNMP - Get Processor Information] OID: .1.3.6.1.2.1.1.3.0, output: 158442636
2019/02/09 07:09:30 - POLLER: Poller[Main Poller] Device[12] Device[UBNT] RECACHE DQ[SNMP - Get Mounted Partitions] OID: .1.3.6.1.2.1.1.3.0, output: 158442636
2019/02/09 07:09:30 - POLLER: Poller[Main Poller] Device[12] Device[UBNT] RECACHE DQ[Net-SNMP - Get Monitored Partitions] OID: .1.3.6.1.2.1.1.3.0, output: 158442635
2019/02/09 07:09:30 - POLLER: Poller[Main Poller] Device[12] Device[UBNT] RECACHE DQ[SNMP - Interface Statistics] OID: .1.3.6.1.2.1.1.3.0, output: 158442635
2019/02/09 07:09:30 - POLLER: Poller[Main Poller] Device[12] Device[UBNT RECACHE: Processing 5 items in the auto reindex cache for '172.23.23.6'.
2019/02/09 07:09:30 - POLLER: Poller[Main Poller] Device[12] Device[UBNT] STATUS: Device '172.6.4.6' is UP.

I am a loss as to how to try and debug this issue.

[xq1xq1xq1]

I searched Cacti.log for timeouts for the ip address of the this device and found no timeouts:

Code: Select all

grep SNMP cacti.log | grep -i time | grep 'x.x.x'

I should also clarify that all of my devices are missing data in their graphs

[xq1xq1xq1]

I also tried to run cmd.php directly for this one host with no data returned:

Code: Select all

# php -q /usr/share/cacti/cmd.php --first=12 --last=12
2019/02/09 07:59:44 - POLLER: Poller[1] NOTE: There are no items in your poller for this polling cycle!

[xq1xq1xq1]

I turned on Data Source Debug for the Host and Graph I am using as my example issue and I get what I expect:

Code: Select all

/bin/rrdtool create \
/usr/share/cacti/rra/ubnt_traffic_in_60.rrd \
--step 300  \
DS:traffic_in:COUNTER:600:0:10000000000000 \
DS:traffic_out:COUNTER:600:0:10000000000000 \
RRA:AVERAGE:0.5:1:600 \
RRA:AVERAGE:0.5:6:700 \
RRA:AVERAGE:0.5:24:775 \
RRA:AVERAGE:0.5:288:797 \
RRA:MIN:0.5:1:600 \
RRA:MIN:0.5:6:700 \
RRA:MIN:0.5:24:775 \
RRA:MIN:0.5:288:797 \
RRA:MAX:0.5:1:600 \
RRA:MAX:0.5:6:700 \
RRA:MAX:0.5:24:775 \
RRA:MAX:0.5:288:797 \
RRA:LAST:0.5:1:600 \
RRA:LAST:0.5:6:700 \
RRA:LAST:0.5:24:775 \
RRA:LAST:0.5:288:797 \

[xq1xq1xq1]

With absolutely NO response from cacti superusers as to rectify or furtherttroubleshoot this problem, am I to understand that I should use a different maintenance platform for my network as cacti doesn't work reliably ???

From what I have seen, cacti is in fact a very good platform for monitoring a network so I don't really believe the statement above.

I am very taken back that NOONE has any idea as to how to rectify this issue.

Guidance is required or I will be forced to look for a reliable platform.

Thank you in advance for your assistance.

[netniV]

There can sometimes be a lot of posts so things can get missed. Additionally, everyone who does respond does so voluntarily, so it may take a while before people have an answer if it's not known before.

If this is one graph that is missing data, use Enable the Device Debug to see if it is getting a value everytime. Increasing logging levels may give you more information as to the timings of things, see if it overruns at all. For 1.1.38, there is a plugin by Jimmy Conner called DSDebug which was integrated into Cacti 1.2.0 (currently at 1.2.2).

Personally, if you haven't been running this long, I would just move forward to 1.2.2 and see how things fair then. Still seeing the same issues use the builtin Troubleshooting -> Data source option to diagnose the specific data sources you are having an issue with.

[xq1xq1xq1]

Thank you for your suggestions.

I found this procedure

Code: Select all

https://www.askitmen.com/unix/centos/centos-7/how-to-install-cacti-1-2-on-centos7/

for installing 1.2 on CENTOS but it is a manual process and not in the CENTOS repository that I could find.

Code: Select all

yum --showduplicates list cacti
Installed Packages
cacti.noarch                                                                     1.1.38-3.el7                                                                      @epel
Available Packages
cacti.noarch                                                                     1.1.38-3.el7                                                                      epel

Is it not important to use the version in the repository to ensure that the packages gets updates?

When I turn on graph debug I get data every time I try it.

I am confused.

Thank you for your guidance.

[netniV]

1.2.2 is actually in the testing repo currently. If you search GitHub there is an issue that was opened for it. Search for EPEL and SELinux

[xq1xq1xq1]

thank you for the advice.

I enabled epel-testing and see version 1.22

Can I simply upgrade to 1.22 with yum and retain my data?

Any downside or chance of issues if I upgrade to 1.22?

I really appreciate your advice!

[xq1xq1xq1]

1.2.2 is now in epel

I upgraded to it.

On the second screen it says:

Code: Select all

/usr/share/cacti/log/  Not Writable

I applied

Code: Select all

chmod -R 777 /usr/share/cacti/log/

to force write

and

Code: Select all

chown -R root.root /usr/share/cacti/log/

as suggested on page

Still no joy

Ideas?

[netniV]

Permissions of the log folder need to be such that the polling user and website user are able to r/w at the folder level. Also, if SELinux is enabled, it needs http_sys_content_rw (I think it is). There's a GitHub Issue tracker for the SELinux part whcih describes the scenarios.

[xq1xq1xq1]

Seems to be working now but do not really have any root cause for the missing data.

Thanks for the assistance!