ensure resp.Body is discarded to avoid broken tcp sockets #348
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Hi guys,
currently the Blackbox exporter produces server-side (Confirmed with Zuul, NGINX, Python 3 http.server, TCP-Resets in Wireshark) broken pipes when probing with a 'big' HTTP request.
Example:
small-garbage
big-garbage
In ProbeHTTP (https://github.com/prometheus/blackbox_exporter/blob/master/prober/http.go#L135) resp.Body is read/discarded only if the payload gets checked for certain contents in https://github.com/prometheus/blackbox_exporter/blob/master/prober/http.go#L42.
The resp.Body should always be read/discarded, no matter whether the body is needed or not.