I encountered This issue, which illustrates this and shows that this mainly occurs during heavy server load, I also noticed it.
I am looking for a solution to keep my server running, but to do it At this point, I need to know more about this behavior. Unfortunately, neither the CouchDB wiki nor the guide provides much information about this (only changes the parameters in the API).
Someone Can you provide more relevant information?
< p>My CouchDB (production) server is sometimes shut down, and I don’t want to do it. While checking the logs, I noticed that Erlang/CouchDB (one or the other) sends heartbeats and monitors the heartbeats. When a timeout occurs, The server was brutally killed (couchdb -k). This is almost always unwanted behavior, definitely on a production server.
I came across this issue, which illustrates this and It shows that this mainly occurs during heavy server load, and I have also noticed it.
I am looking for a solution to keep my server running, but to do this, I need to learn more about Information on this behavior. Unfortunately, neither the CouchDB wiki nor the guide provides much information about this (only the parameters in the API are changed).
Can anyone provide more information?
couchdb -k should kill the background process and then the couch should be restarted. However, by default, the restart feature is disabled. To enable it, you should use -R parameter with a value greater than 0. You can do this in the bin/couchdb script-change RESPAWN_TIMEOUT from 0 to for example 5.