chore(seer grouping): Add debug logging for old excess frames check

[lobsterkatie]

As of #82414, we should no longer be using TooManyOnlySystemFramesException, but the fact that we're still getting did_call_seer metrics tagged with "over-threshold-frames" says that we still are, somehow. This is currently a blocker on merging #82434, which removes TooManyOnlySystemFramesException and the code associated with it, so it'd be nice to figure out why it's happening so we can fix it.

Try as I might, I can't replicate it, nor can I find the flaw in the logic, so this adds a temporary log to try to help figure out how in the heck it's happening. It also adds a record_metrics parameter to the has_too_many_contributing_frames helper, so that I can call it for purposes of the logging without messing up the metrics it collects. Once I figure it out what's happening, I'll back out all of these changes.

[wedamija]

Just checking that this isn't expensive to run here? I assume this log happens only on group creation

[lobsterkatie]

It runs on every event, but no, it's not expensive - all it does is check a few different pre-computed values (and collect metrics, if record_metrics is true, but in this case it's false, so that's a moot point).

sentry/src/sentry/seer/similarity/utils.py

Lines 323 to 382 in 520f8d9

	def has_too_many_contributing_frames(
	event: Event | GroupEvent,
	variants: dict[str, BaseVariant],
	referrer: ReferrerOptions,
	record_metrics: bool = True,
	) -> bool:
	platform = event.platform
	shared_tags = {"referrer": referrer.value, "platform": platform}
	contributing_variant, contributing_component = get_contributing_variant_and_component(variants)
	# Ideally we're calling this function after we already know the event both has a stacktrace and
	# is using it for grouping (in which case none of the below conditions should apply), but still
	# worth checking that we have enough information to answer the question just in case
	if (
	# Fingerprint, checksum, fallback variants
	not isinstance(contributing_variant, ComponentVariant)
	# Security violations, log-message-based grouping
	or contributing_variant.variant_name == "default"
	# Any ComponentVariant will have this, but this reassures mypy
	or not contributing_component
	# Exception-message-based grouping
	or not hasattr(contributing_component, "frame_counts")
	):
	# We don't bother to collect a metric on this outcome, because we shouldn't have called the
	# function in the first place
	return False
	# Certain platforms were backfilled before we added this filter, so to keep new events matching
	# with the existing data, we turn off the filter for them (instead their stacktraces will be
	# truncated)
	if platform in EVENT_PLATFORMS_BYPASSING_FRAME_COUNT_CHECK:
	if record_metrics:
	metrics.incr(
	"grouping.similarity.frame_count_filter",
	sample_rate=options.get("seer.similarity.metrics_sample_rate"),
	tags={**shared_tags, "outcome": "bypass"},
	)
	return False
	stacktrace_type = "in_app" if contributing_variant.variant_name == "app" else "system"
	key = f"{stacktrace_type}_contributing_frames"
	shared_tags["stacktrace_type"] = stacktrace_type
	if contributing_component.frame_counts[key] > MAX_FRAME_COUNT:
	if record_metrics:
	metrics.incr(
	"grouping.similarity.frame_count_filter",
	sample_rate=options.get("seer.similarity.metrics_sample_rate"),
	tags={**shared_tags, "outcome": "block"},
	)
	return True
	if record_metrics:
	metrics.incr(
	"grouping.similarity.frame_count_filter",
	sample_rate=options.get("seer.similarity.metrics_sample_rate"),
	tags={**shared_tags, "outcome": "pass"},
	)
	return False

If you still think it's an issue, I can move the call to where the log is, which would drastically cut down the number of times that it runs. I originally had it that way, but in the end, especially since it's hopefully only going to be in place for a day or two while I debug this, it just seemed easier to pass one metadata blob rather than all the stuff needed to compute said blob's values. LMK if you want me to change it.

[wedamija]

I think that as long as it won't cause a performance issue in prod I'm fine with it as is