Couldn't write data for continuous query #1120
Comments
|
I updated to the version 0.8.6. Here is the updated error: |
|
@odolbeau Do you have other continuous queries that are working correctly? Can you provide any other details about the continuous queries themselves? |
|
We use InfluxDB to store all http queries made on our website. Here is the json used to create the DB. {
"spaces": [
{
"name": "everything",
"retentionPolicy": "6h",
"shardDuration": "3h",
"regex": "/.*/",
"replicationFactor": 1,
"split": 1
},
{
"name": "1h",
"retentionPolicy": "30d",
"shardDuration": "1d",
"regex": "/.*1h/",
"replicationFactor": 1,
"split": 1
},
{
"name": "1m",
"retentionPolicy": "7d",
"shardDuration": "1d",
"regex": "/.*1m/",
"replicationFactor": 1,
"split": 1
}
],
"continuousQueries": [
"select * from http.response into http.response.by_media.[media]",
"select * from http.response into http.response.by_locale.[locale]",
"select * from http.response into http.response.by_locale_and_media.[locale].[media]",
"select * from http.response into http.response.by_route.[route]",
"select count(duration), median(duration) as median, percentile(duration, 95) as percentile_95, percentile(duration, 99) as percentile_99 from /^http.response.by_.*/ group by time(1m) into agg.:series_name.duration.1m",
"select count(duration), median(duration) as median, percentile(duration, 95) as percentile_95, percentile(duration, 99) as percentile_99 from /^http.response.by_.*/ group by time(1h) into agg.:series_name.duration.1h"
]
}All associated logs: Maybe this problem is due to invalid data ? (a missing |
|
I faced the same problem today. It's a newly created database (on a clean setup, after crash mentioned in #1226). It is filled with just a couple hours of data (roughly 1M records), and when I create a continuous query, it does not work and log says: |
|
Well, the funny part is that this continuous query started working after re-creation. I dropped both continuous query and the series, and created it again: No more errors in log, it works as expected. Probably the reason for this is related to shard creation. When I was creating CQ first time, there were no shard existing suitable for this series. And the second time shard already existed (I did not deleted the shard, just series). |
beckettsean
added
obsoleted by v0.9.0
and removed
0 - Backlog
area/continuous queries
labels
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi!
I have a lot of errors with my continuous queries. Unfortunately, I cannot explain why I encounter this problem and I didn't find anyone who have the same issue.
Environment:
Our applications send events to an logstash server, configured with the influxdb output.
We receive (I think) between 100k & 150k events / minute.
Here is the error:
If needed I can give you more informations, just tell me. :)
The text was updated successfully, but these errors were encountered: