Closed Mitrofanov closed 4 years ago
Hi @Mitrofanov,
I haven't been able to reproduce the issue locally with the official ES docker images (though there doesn't seem to be a tag for 6.3.x - I tried the earliest 6.x versions available: 6.4.0 and 6.8.2). It's probably something specific with your setup and/or AWS's modifications.
Are you able to attach the output of a manual _nodes/stats
call (redacted as needed)? Feel free to send me an email if that's easier.
Thanks, Braedon
Hi, i think problem is with AWS managed ES. So looks like it blocks requests to system endpoints (like _nodes/stats
), that's why i get such error.
Just wonder if we can add switch to disable system metrics collection.
There are already arguments to disable the different system metrics collectors. e.g. --nodes-stats-disable
for _nodes/stats
.
If you can, could you still send me that output of the manual _nodes/stats
call, or the return value of the es_client.nodes.stats()
call in NodesStatsCollector.collect()
if the manual call doesn't get a response at all. It'd be good to work out why its getting past the failure checks.
AWS Managed Elastic Search 7.4
output
/_nodes/stats
outputThanks @jonny-wg2, I think I've found the issue.
In the fs.data
list for the nodes, it looks like AWS is dropping the path
field for each data directory entry. The parser expects the path to be there as it uses it to populate the path
label, identifying the metrics for each data directory.
AWS example:
...
"fs" : {
...
"data" : [
{
"type" : "ext4",
"total_in_bytes" : 105152176128,
"free_in_bytes" : 49135415296,
"available_in_bytes" : 43749928960
}
],
...
},
...
Non-AWS example:
...
"fs" : {
...
"data" : [
{
"path" : "/usr/share/elasticsearch/data/nodes/0",
"mount" : "/ (none)",
"type" : "aufs",
"total_in_bytes" : 91787288576,
"free_in_bytes" : 15394492416,
"available_in_bytes" : 10708357120
}
],
...
},
...
I'll put in a fallback to use the data directory's position in the fs.data
list as the path
label value e.g. path="0"
. I'm not sure if AWS ever uses more than one data directory, but I'd hope they keep the same order between calls if they do... Fingers crossed.
Thanks @braedon !
Hi,
I faced with errors in some queries. Errors are as follow (pls check that response code was 200):
I use AWS managed ES version 6.3. I've just upgraded es exporter to latest version (0.5.2), but issue still persists.
Please suggest.