Pass delivery abandoned does not qualify as an error

... so log it under the Debug tag.

FetchErrors should be actual errors which can be addressed. In this case,
nothing is wrong in any way, the fact that we abort a fetch if we don't
need the body is a varnish internal optimization (which makes sense, but
comes at the cost of closing a connection).

Merges #2450

Conflicts:
	bin/varnishd/cache/cache_fetch.c

bin/varnishd/cache/cache_fetch.c

@@ -495,8 +495,8 @@ vbf_fetch_body_helper(struct busyobj *bo)
 			 * objects to be created.
 			 */
 			AN(vfc->oc->flags & OC_F_PASS);
-			VSLb(vfc->wrk->vsl, SLT_FetchError,
-			    "Pass delivery abandoned");
+			VSLb(vfc->wrk->vsl, SLT_Debug,
+			    "Fetch: Pass delivery abandoned");
 			vfps = VFP_END;
 			bo->htc->doclose = SC_RX_BODY;
 			break;

doc/changes.rst

@@ -21,6 +21,7 @@ Bugs fixed
 
 * 1772_ - Honor first_byte_timeout for recycled backend connections
 * 2135_ - Limit Backend Connection retries to a single retry
+* 2450_ - Pass delivery abandoned does not qualify as an error
 * 2495_ - Avoid leaking an OH ref on reembark failure
 * 2502_ - objcore reference count leak
 * 2530_ - Varnish shouldn't test gunzip for range responses
@@ -29,6 +30,7 @@ Bugs fixed
 
 .. _1772: https://github.com/varnishcache/varnish-cache/issues/1772
 .. _2135: https://github.com/varnishcache/varnish-cache/pull/2135
+.. _2450: https://github.com/varnishcache/varnish-cache/pull/2450
 .. _2495: https://github.com/varnishcache/varnish-cache/issues/2495
 .. _2502: https://github.com/varnishcache/varnish-cache/issues/2502
 .. _2530: https://github.com/varnishcache/varnish-cache/issues/2530