wohali commented on issue #859: Various error messages and breakage URL: https://github.com/apache/couchdb/issues/859#issuecomment-333619692 Basically all of your errors are timeouts. Your system is overloaded. You can keep bumping up timeouts, but this only sweeps the problem under the rug. You should consider moving to a proper CouchDB 2.x 3-node cluster to start, with a load balancer in front. You may need to experiment to determine the correct `q` value for this database as well. By spreading the load across more nodes you should be able to scale this horizontally. You could consider running the nodes all on the same Docker host, but only you know what load you have and what that machine is capable of handling. For redundancy it would be better to use separate physical hosts, even if those machines are also running Docker containers. Another thing to consider is, if you are using filtered `_changes` feeds (could relate to your `couchjs` os process timeouts), to switch to [Mango-style selectors](http://docs.couchdb.org/en/2.1.0/api/database/changes.html?highlight=selector#selector) which are significantly better performance than JavaScript filters. None of the error messages above suggest you're hitting file handle limits (you'd see `emfile` or `enfile` errors). ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
With regards, Apache Git Services
