Hello team ,
I am observing strange issue where every time a DCI for ping on a particular devices is polled it result in two values (example attached). Checked in the database, the values there are the same:
| 261543 | 1509960003 | 100 |
| 261543 | 1509960003 | 0 |
| 261543 | 1509960103 | 0 |
| 261543 | 1509960103 | 100 |
| 261543 | 1509960203 | 0 |
| 261543 | 1509960203 | 100 |
| 261543 | 1509960303 | 0 |
| 261543 | 1509960303 | 100 |
Any idea what is causing this or how can be fixed ?
Hi,
seems to be very strange bug. Table should have DCI ID + timestamp as primary key, so those record theoretically should not appear. What database you are using? What NetXMS version? Could you check constraints on a table? How DCI is configured?
Best regards,
Victor
Hello Victor ,
Hope you are doing well.
The database is:
~# mysql -V
mysql Ver 14.14 Distrib 5.5.40, for debian-linux-gnu (x86_64) using readline 6.2
NetXMS server version is
:~# netxmsd -v
NetXMS Server Version 2.0.7 Build 8076 (2.0.7-RELEASE-5-g3f73792) (NON-UNICODE)
NXCP: 3.47.1.4 (AES-256, Blowfish-256, 3DES, AES-128, Blowfish-128)
NetXMS agent that collects the DCI is:
# nxagentd -v
NetXMS Core Agent Version 2.1-RC1 Build 9010 (UNKNOWN) (UNICODE)
The DCI is a simple Agent ping with source the NetXMS Agent server (screenshot attached).
Describe of the table attached as screenshot.
I haven't made any configuration changes on the DCI since i made this post, now i can see that the time stamps are not exactly the same as before. But the issue stays similar. Two values one 0 and one 100 in less interval than the configured (100 seconds).
Hi,
are you using cache mode for agent? It seems like corrupted data collection configuration on agent side, when agent sends both correct data and data for some other DCI. If yes, please check what is configured in table dc_config in agent local database. You may also try to delete agent local database (while agent is stopped) - it should reconfigure on first configuration poll.
Best regards,
Victor