Node timeout in clusters

CrateDB
1

Hi All,

We did a setup of cratedb. in which are getting below error.

[o.e.t.TransportService ] [Gametzalpenkopf] Received response for a request that has timed out, sent [5003ms] ago, timed out [2001ms] ago, action [internal:crate:sql/sys/nodes], node [{Hochkünzelspitze}{879XQnSHQTGY6r9je_jG8g}{spOq_XLFTdaoLiz8IKg_bw}{192.168.1.188}{192.168.1.188:4300}{http_address=192.168.1.188:4200}], id [143070680]

Server is VM in which we crated 3 nodes and given 35 GB heap size to Cratedb.
Any idea how to solve this issue.

Also the GUI look like below

image
image1352×607 25.6 KB

2

Hi,
This message may indicate that node Hochkünzelspitze is unresponsive or that there is a network connectivity issue between Gametzalpenkopf and Hochkünzelspitze.
I would advice you take a quick look to see if there is anything from the infrastructure side and if everything looks ok check on sys.jobs / sys.jobs_log what is running/has run. There may be some query taking too many CPU resources.

3

Hi,

Thanks for your response.
we kill that top CPU-consuming query.
but still, timeout is occurring

thank
Vinayak Katkar

4

Hi ,
Is there any parameter, that changes the node communication timeout?
means can we increase node-to-node timeout communication time to avoid spit brain.

5

Dear Vinayak,

thank you for writing in. I can see you are using CrateDB 4.8.4. While I can’t pinpoint any specific issue, would using a more recent version of CrateDB be possible, to see if those observed woes go away?

With kind regards,
Andreas.