diff --git a/gstlal-inspiral/bin/gstlal_inspiral b/gstlal-inspiral/bin/gstlal_inspiral
index c37672165a7e78553ff86f4d46234d42bc21d8ee..14f736e356fedae89994b105350fd0ecf9a3514e 100755
--- a/gstlal-inspiral/bin/gstlal_inspiral
+++ b/gstlal-inspiral/bin/gstlal_inspiral
@@ -780,7 +780,7 @@ for output_file_number, (svd_bank_url_dict, output_url, likelihood_url_namedtupl
 		rankingstat = rankingstat,
 		zerolag_rankingstatpdf_filename = zerolag_rankingstatpdf_filename,
 		rankingstatpdf_filename = options.marginalized_likelihood_file,
-		likelihood_url_namedtuple = likelihood_url_namedtuple,
+		likelihood_url_namedtuple = likelihood_url_namedtuple if not options.injections else None,
 		likelihood_snapshot_interval = options.likelihood_snapshot_interval,	# seconds
 		thinca_interval = options.thinca_interval,
 		min_log_L = options.min_log_L,
diff --git a/gstlal-inspiral/python/inspiral.py b/gstlal-inspiral/python/inspiral.py
index 9b37beaa423709fb0bf21506886744c1685517e2..933333e038fcc8402b7a19969f61bc98957e11a0 100644
--- a/gstlal-inspiral/python/inspiral.py
+++ b/gstlal-inspiral/python/inspiral.py
@@ -567,6 +567,22 @@ class Data(object):
 		# might be configured to disable ranking statistic output
 		# since they produce nonsense.
 		#
+		# FIXME:  gstlal_inspiral now sets the likelihood_url to
+		# None when doing injections to disable the collection of
+		# ranking statistic information in injection jobs.  this is
+		# now yet another way in which online analyses are broken
+		# following the ranking statistic rewrite and will need to
+		# be looked at.  the reason for doing this is that because
+		# injection jobs skip intervals of SNR reconstruction they
+		# create ranking statistic data that appear to be
+		# inconsistent with the triggers that are being produced
+		# and the inconsistency triggers assertion failures
+		# throughout the new ranking statistic code.  we have
+		# tried, in vain, to work around the problem but in the end
+		# we've had to simply stop injection jobs from even trying.
+		# the data they collect was never used anyway, exactly
+		# because of its inconsistencies.
+		#
 
 		if likelihood_url_namedtuple is not None:
 			self.reference_likelihood_url = likelihood_url_namedtuple.reference_likelihood_url