Temporary failure in name resolution

influxdb
#1

Hello everyone,

We have been currently using the Influxdb and Telegraf combination in our project for the past four months.
The Telegraf listens to a mqtt broker and then writes the data to the InfluxDB in different databases.
This works perfectly.

Recently (from last week), we found the following Info message in the influxdb logs:

2019-04-25T23:59:40.007788198Z ts=2019-04-25T23:59:40.007677Z lvl=info msg="Post http://a2976570d9ba:9092/write?consistency=&db=db3&precision=ns&rp=autogen: dial tcp: lookup a2976570d9ba: Temporary failure in name resolution" log_id=0F1EDu60000 service=subscriber
2019-04-25T23:59:40.007804099Z ts=2019-04-25T23:59:40.007715Z lvl=info msg="Post http://a2976570d9ba:9092/write?consistency=&db=db2&precision=ns&rp=autogen: dial tcp: lookup a2976570d9ba: Temporary failure in name resolution" log_id=0F1EDu60000 service=subscriber
2019-04-25T23:59:40.007807899Z ts=2019-04-25T23:59:40.007730Z lvl=info msg="Post http://a2976570d9ba:9092/write?consistency=&db=db7&precision=ns&rp=autogen: dial tcp: lookup a2976570d9ba: Temporary failure in name resolution" log_id=0F1EDu60000 service=subscriber
2019-04-25T23:59:40.026503322Z [httpd] 172.18.0.6 - admin [25/Apr/2019:23:59:40 +0000] "POST /write?db=db3 HTTP/1.1" 204 0 "-" "Telegraf/1.10.3" 3016ba6c-67b6-11e9-9063-0242ac120003 24868
2019-04-25T23:59:40.026629623Z [httpd] 172.18.0.6 - admin [25/Apr/2019:23:59:40 +0000] "POST /write?db=db7 HTTP/1.1" 204 0 "-" "Telegraf/1.10.3" 3016ba74-67b6-11e9-9064-0242ac120003 24863
2019-04-25T23:59:40.036073836Z [httpd] 172.18.0.6 - admin [25/Apr/2019:23:59:40 +0000] "POST /write?db=db2 HTTP/1.1" 204 0 "-" "Telegraf/1.10.3" 3016b433-67b6-11e9-9062-0242ac120003 34637

The Telegraf writes to different databases and the Info message occurs for almost all the incoming data.
But still the InfluxDB contains all the data.

As we could not find any network or other issues, we tried running the same setup after deleting all the old data (340 MB). This solved the issue and we did not receive the Info message anymore.
Does this mean this is a data overload issue?

Thanks in advance for any suggesetions…

#2

Those info lines indicate one or more subscriptions are not able to deliver data to the subscribed service (usually a Kapacitor instance). If you are running Kapacitor, I recommend checking if it is still running and reachable by InfluxDB. Otherwise, if that Kapacitor instance is no longer running, you can remove the subscription.

Orphaned Kapacitor subscriptions can sometimes pop up when people connect Kapacitor running on their dev laptop to InfluxDB and then fail to do a clean shutdown of Kapacitor.

#3

Yes, there were a few kapacitor subscriptions hanging still.
After deleting them, the messages does not come any more.
Thank you for the solution.

1 Like