Oliver Eilhard
Oliver Eilhard
Thanks for reporting. Yep, there's something wrong. I probably know why this happens, and will fix in the next release.
Hmm... for the moment, I'm not sure where this is coming from. Need to investigate a bit more.
Which amount of memory usage would be acceptable for your application? That's 1.4 MB—not too uncommon these days.
Can you provide a reproducible test case that yields the error? I'm not really sure this is a driver issue.
Can you enable the trace log to see the communication between elastic and ES. That’s what I typically do for debugging these issues. See https://github.com/olivere/elastic/wiki/Logging
Thanks. The elastic.v5 branch is no longer maintained, but we can port back the changes to v6 and v7. It looks like this is basically what we have [here](https://github.com/olivere/elastic/blob/release-branch.v7/errors.go#L95-L109), right?...
Can you make a test similar to [this one](https://github.com/olivere/elastic/blob/release-branch.v7/backoff_test.go#L36-L79) that fails?
I agree, that seems to be wrong. It should be `-1` by default, or incremented after retrying.
This will need to be backported to 6.x and 5.x.
Thanks for your commit. I probably won't add anything like this to the core library. Last time I checked the cost of deserialization was neglectable compared to the I/O and...