def test_resolved_in_next_release(self):
self.login_as(user=self.user)
project = self.create_project()
project.flags.has_releases = True
project.save()
group = self.create_group(project=project)
Release.get_or_create(
version='abcd',
project=project,
)
url = '/api/0/issues/{}/'.format(group.id)
response = self.client.put(
url, data={
'status': 'resolvedInNextRelease',
}
)
assert response.status_code == 200, response.content
group = Group.objects.get(
id=group.id,
project=group.project.id,
)
assert group.status == GroupStatus.RESOLVED
assert GroupResolution.objects.filter(
group=group,
).exists()
def test_transfer_to_team_releases(self):
from_org = self.create_organization()
from_team = self.create_team(organization=from_org)
to_org = self.create_organization()
to_team = self.create_team(organization=to_org)
project = self.create_project(teams=[from_team])
environment = Environment.get_or_create(project, 'production')
release = Release.get_or_create(project=project, version='1.0')
ReleaseProjectEnvironment.objects.create(
project=project,
release=release,
environment=environment,
)
assert ReleaseProjectEnvironment.objects.filter(
project=project,
release=release,
environment=environment,
).exists()
assert ReleaseProject.objects.filter(
project=project,
release=release,
).exists()
project.transfer_to(team=to_team)
project = Project.objects.get(id=project.id)
assert project.teams.count() == 1
assert project.teams.first() == to_team
assert project.organization_id == to_org.id
assert not ReleaseProjectEnvironment.objects.filter(
project=project,
release=release,
environment=environment,
).exists()
assert not ReleaseProject.objects.filter(
project=project,
release=release,
).exists()
def get_release(self, create=False):
"""Convenient helper to return the release for the current data
and optionally creates the release if it's missing. In case there
is no release info it will return `None`.
"""
release = self.data.get('release')
if not release:
return None
if not create:
return Release.get(project=self.project, version=self.data['release'])
timestamp = self.data.get('timestamp')
if timestamp is not None:
date = datetime.fromtimestamp(timestamp).replace(tzinfo=timezone.utc)
else:
date = None
return Release.get_or_create(
project=self.project,
version=self.data['release'],
date_added=date,
)
def get_release(self, create=False):
"""Convenient helper to return the release for the current data
and optionally creates the release if it's missing. In case there
is no release info it will return `None`.
"""
release = self.data.get("release")
if not release:
return None
if not create:
return Release.get(project=self.project,
version=self.data["release"])
timestamp = self.data.get("timestamp")
if timestamp is not None:
date = datetime.fromtimestamp(timestamp).replace(
tzinfo=timezone.utc)
else:
date = None
return Release.get_or_create(project=self.project,
version=self.data["release"],
date_added=date)
def debounce_update_release_health_data(organization, project_ids):
"""This causes a flush of snuba health data to the postgres tables once
per minute for the given projects.
"""
# Figure out which projects need to get updates from the snuba.
should_update = {}
cache_keys = ["debounce-health:%d" % id for id in project_ids]
cache_data = cache.get_many(cache_keys)
for project_id, cache_key in izip(project_ids, cache_keys):
if cache_data.get(cache_key) is None:
should_update[project_id] = cache_key
if not should_update:
return
projects = {p.id: p for p in Project.objects.get_many_from_cache(should_update.keys())}
# This gives us updates for all release-projects which have seen new
# health data over the last 24 hours. It will miss releases where the last
# date is <24h ago. We need to aggregate the data for the totals per release
# manually here now. This does not take environments into account.
for project_id, version in get_changed_project_release_model_adoptions(should_update.keys()):
project = projects.get(project_id)
if project is None:
# should not happen
continue
# We might have never observed the release. This for instance can
# happen if the release only had health data so far. For these cases
# we want to create the release the first time we observed it on the
# health side.
release = Release.get_or_create(project=project, version=version)
# Make sure that the release knows about this project. Like we had before
# the project might not have been associated with this release yet.
release.add_project(project)
# Debounce updates for a minute
cache.set_many(dict(izip(should_update.values(), [True] * len(should_update))), 60)
def _get_or_create_release_many(jobs, projects):
jobs_with_releases = {}
release_date_added = {}
for job in jobs:
if not job["release"]:
continue
release_key = (job["project_id"], job["release"])
jobs_with_releases.setdefault(release_key, []).append(job)
new_datetime = job["event"].datetime
old_datetime = release_date_added.get(release_key)
if old_datetime is None or new_datetime > old_datetime:
release_date_added[release_key] = new_datetime
for (project_id,
version), jobs_to_update in six.iteritems(jobs_with_releases):
release = Release.get_or_create(
project=projects[project_id],
version=version,
date_added=release_date_added[(project_id, version)],
)
for job in jobs_to_update:
# dont allow a conflicting 'release' tag
data = job["data"]
pop_tag(data, "release")
set_tag(data, "sentry:release", release.version)
job["release"] = release
if job["dist"]:
job["dist"] = job["release"].add_dist(job["dist"],
job["event"].datetime)
# dont allow a conflicting 'dist' tag
pop_tag(job["data"], "dist")
set_tag(job["data"], "sentry:dist", job["dist"].name)
def save(self, project, raw=False):
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
message = data.pop('message')
level = data.pop('level')
culprit = data.pop('culprit', None)
time_spent = data.pop('time_spent', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
environment = data.pop('environment', None)
if not culprit:
culprit = generate_culprit(data)
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'message': message,
'platform': platform,
}
event = Event(project_id=project.id,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs)
tags = data.get('tags') or []
tags.append(('level', LOG_LEVELS[level]))
if logger_name:
tags.append(('logger', logger_name))
if server_name:
tags.append(('server_name', server_name))
if site:
tags.append(('site', site))
if release:
# TODO(dcramer): we should ensure we create Release objects
tags.append(('sentry:release', release))
if environment:
tags.append(('environment', environment))
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags,
event,
_with_transaction=False)
if added_tags:
tags.extend(added_tags)
event_user = self._get_event_user(project, data)
if event_user:
tags.append(('sentry:user', event_user.tag_value))
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
data['fingerprint'] = fingerprint or ['{{ default }}']
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = map(md5_from_hash,
get_hashes_from_fingerprint(event, fingerprint))
elif checksum:
hashes = [checksum]
else:
hashes = map(md5_from_hash, get_hashes_for_event(event))
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'time_spent_total': time_spent or 0,
'time_spent_count': time_spent and 1 or 0,
})
if release:
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
group_kwargs['first_release'] = release
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **group_kwargs)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project,
group=group,
event_id=event_id)
except IntegrityError:
self.logger.info('Duplicate EventMapping found for event_id=%s',
event_id,
exc_info=True)
return event
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info('Duplicate Event found for event_id=%s',
event_id,
exc_info=True)
return event
if event_user:
tsdb.record_multi((
(tsdb.models.users_affected_by_group, group.id,
(event_user.tag_value, )),
(tsdb.models.users_affected_by_project, project.id,
(event_user.tag_value, )),
),
timestamp=event.datetime)
if is_new and release:
buffer.incr(Release, {'new_groups': 1}, {
'id': release.id,
})
safe_execute(Group.objects.add_tags,
group,
tags,
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info(
'Raw event passed; skipping post process for event_id=%s',
event_id)
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def save(self, project_id, raw=False, assume_normalized=False):
# Normalize if needed
if not self._normalized:
if not assume_normalized:
self.normalize()
self._normalized = True
data = self._data
project = Project.objects.get_from_cache(id=project_id)
project._organization_cache = Organization.objects.get_from_cache(
id=project.organization_id)
# Check to make sure we're not about to do a bunch of work that's
# already been done if we've processed an event with this ID. (This
# isn't a perfect solution -- this doesn't handle ``EventMapping`` and
# there's a race condition between here and when the event is actually
# saved, but it's an improvement. See GH-7677.)
try:
event = Event.objects.get(
project_id=project.id,
event_id=data['event_id'],
)
except Event.DoesNotExist:
pass
else:
# Make sure we cache on the project before returning
event._project_cache = project
logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': data['event_id'],
'project_id': project.id,
'model': Event.__name__,
}
)
return event
# Pull out the culprit
culprit = self.get_culprit()
# Pull the toplevel data we're interested in
level = data.get('level')
# TODO(mitsuhiko): this code path should be gone by July 2018.
# This is going to be fine because no code actually still depends
# on integers here. When we need an integer it will be converted
# into one later. Old workers used to send integers here.
if level is not None and isinstance(level, six.integer_types):
level = LOG_LEVELS[level]
transaction_name = data.get('transaction')
logger_name = data.get('logger')
release = data.get('release')
dist = data.get('dist')
environment = data.get('environment')
recorded_timestamp = data.get('timestamp')
# We need to swap out the data with the one internal to the newly
# created event object
event = self._get_event_instance(project_id=project_id)
self._data = data = event.data.data
event._project_cache = project
date = event.datetime
platform = event.platform
event_id = event.event_id
if transaction_name:
transaction_name = force_text(transaction_name)
# Some of the data that are toplevel attributes are duplicated
# into tags (logger, level, environment, transaction). These are
# different from legacy attributes which are normalized into tags
# ahead of time (site, server_name).
setdefault_path(data, 'tags', value=[])
set_tag(data, 'level', level)
if logger_name:
set_tag(data, 'logger', logger_name)
if environment:
set_tag(data, 'environment', environment)
if transaction_name:
set_tag(data, 'transaction', transaction_name)
if release:
# dont allow a conflicting 'release' tag
pop_tag(data, 'release')
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
set_tag(data, 'sentry:release', release.version)
if dist and release:
dist = release.add_dist(dist, date)
# dont allow a conflicting 'dist' tag
pop_tag(data, 'dist')
set_tag(data, 'sentry:dist', dist.name)
else:
dist = None
event_user = self._get_event_user(project, data)
if event_user:
# dont allow a conflicting 'user' tag
pop_tag(data, 'user')
set_tag(data, 'sentry:user', event_user.tag_value)
# At this point we want to normalize the in_app values in case the
# clients did not set this appropriately so far.
grouping_config = load_grouping_config(
get_grouping_config_dict_for_event_data(data, project))
normalize_stacktraces_for_grouping(data, grouping_config)
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event, _with_transaction=False)
if added_tags:
# plugins should not override user provided tags
for key, value in added_tags:
if get_tag(data, key) is None:
set_tag(data, key, value)
for path, iface in six.iteritems(event.interfaces):
for k, v in iface.iter_tags():
set_tag(data, k, v)
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.path, None)
# The active grouping config was put into the event in the
# normalize step before. We now also make sure that the
# fingerprint was set to `'{{ default }}' just in case someone
# removed it from the payload. The call to get_hashes will then
# look at `grouping_config` to pick the right paramters.
data['fingerprint'] = data.get('fingerprint') or ['{{ default }}']
apply_server_fingerprinting(data, get_fingerprinting_config_for_project(project))
# Here we try to use the grouping config that was requested in the
# event. If that config has since been deleted (because it was an
# experimental grouping config) we fall back to the default.
try:
hashes = event.get_hashes()
except GroupingConfigNotFound:
data['grouping_config'] = get_grouping_config_dict_for_project(project)
hashes = event.get_hashes()
data['hashes'] = hashes
# we want to freeze not just the metadata and type in but also the
# derived attributes. The reason for this is that we push this
# data into kafka for snuba processing and our postprocessing
# picks up the data right from the snuba topic. For most usage
# however the data is dynamically overriden by Event.title and
# Event.location (See Event.as_dict)
materialized_metadata = self.materialize_metadata()
event_metadata = materialized_metadata['metadata']
data.update(materialized_metadata)
data['culprit'] = culprit
# index components into ``Event.message``
# See GH-3248
event.message = self.get_search_message(event_metadata, culprit)
received_timestamp = event.data.get('received') or float(event.datetime.strftime('%s'))
# The group gets the same metadata as the event when it's flushed but
# additionally the `last_received` key is set. This key is used by
# _save_aggregate.
group_metadata = dict(materialized_metadata)
group_metadata['last_received'] = received_timestamp
kwargs = {
'platform': platform,
'message': event.message,
'culprit': culprit,
'logger': logger_name,
'level': LOG_LEVELS_MAP.get(level),
'last_seen': date,
'first_seen': date,
'active_at': date,
'data': group_metadata,
}
if release:
kwargs['first_release'] = release
try:
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **kwargs
)
except HashDiscarded:
event_discarded.send_robust(
project=project,
sender=EventManager,
)
metrics.incr(
'events.discarded',
skip_internal=True,
tags={
'organization_id': project.organization_id,
'platform': platform,
},
)
raise
else:
event_saved.send_robust(
project=project,
event_size=event.size,
sender=EventManager,
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
# When an event was sampled, the canonical source of truth
# is the EventMapping table since we aren't going to be writing out an actual
# Event row. Otherwise, if the Event isn't being sampled, we can safely
# rely on the Event table itself as the source of truth and ignore
# EventMapping since it's redundant information.
if is_sample:
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project, group=group, event_id=event_id)
except IntegrityError:
logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': EventMapping.__name__,
}
)
return event
environment = Environment.get_or_create(
project=project,
name=environment,
)
group_environment, is_new_group_environment = GroupEnvironment.get_or_create(
group_id=group.id,
environment_id=environment.id,
defaults={
'first_release': release if release else None,
},
)
if release:
ReleaseEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
ReleaseProjectEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date,
)
counters = [
(tsdb.models.group, group.id),
(tsdb.models.project, project.id),
]
if release:
counters.append((tsdb.models.release, release.id))
tsdb.incr_multi(counters, timestamp=event.datetime, environment_id=environment.id)
frequencies = [
# (tsdb.models.frequent_projects_by_organization, {
# project.organization_id: {
# project.id: 1,
# },
# }),
# (tsdb.models.frequent_issues_by_project, {
# project.id: {
# group.id: 1,
# },
# })
(tsdb.models.frequent_environments_by_group, {
group.id: {
environment.id: 1,
},
})
]
if release:
frequencies.append(
(tsdb.models.frequent_releases_by_group, {
group.id: {
grouprelease.id: 1,
},
})
)
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(
group=group,
environment=environment,
)
# Update any event attachment that arrived before the event group was defined.
EventAttachment.objects.filter(
project_id=project.id,
event_id=event_id,
).update(
group_id=group.id,
)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': Event.__name__,
}
)
return event
tagstore.delay_index_event_tags(
organization_id=project.organization_id,
project_id=project.id,
group_id=group.id,
environment_id=environment.id,
event_id=event.id,
tags=event.tags,
date_added=event.datetime,
)
if event_user:
tsdb.record_multi(
(
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value, )),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value, )),
),
timestamp=event.datetime,
environment_id=environment.id,
)
if release:
if is_new:
buffer.incr(
ReleaseProject, {'new_groups': 1}, {
'release_id': release.id,
'project_id': project.id,
}
)
if is_new_group_environment:
buffer.incr(
ReleaseProjectEnvironment, {'new_issues_count': 1}, {
'project_id': project.id,
'release_id': release.id,
'environment_id': environment.id,
}
)
safe_execute(
Group.objects.add_tags,
group,
environment,
event.get_tags(),
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send_robust(project=project, group=group, sender=Project)
eventstream.insert(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
is_new_group_environment=is_new_group_environment,
primary_hash=hashes[0],
# We are choosing to skip consuming the event back
# in the eventstream if it's flagged as raw.
# This means that we want to publish the event
# through the event stream, but we don't care
# about post processing and handling the commit.
skip_consume=raw,
)
metrics.timing(
'events.latency',
received_timestamp - recorded_timestamp,
tags={
'project_id': project.id,
},
)
metrics.timing(
'events.size.data.post_save',
event.size,
tags={'project_id': project.id}
)
return event
def create_message_event(template, parameters, environment, release):
i = next(sequence)
event_id = uuid.UUID(fields=(
i,
0x0,
0x1000,
0x80,
0x80,
0x808080808080,
), ).hex
tags = [['color', next(tag_values)]]
if environment:
tags.append(['environment', environment])
if release:
tags.append(['sentry:release', release])
event = Event.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
message='%s' % (id, ),
datetime=now + shift(i),
data={
'environment': environment,
'type': 'default',
'metadata': {
'title': template % parameters,
},
'logentry': {
'message': template,
'params': parameters,
'formatted': template % parameters,
},
'user': next(user_values),
'tags': tags,
},
)
with self.tasks():
Group.objects.add_tags(
source,
Environment.objects.get(
organization_id=project.organization_id,
name=environment),
tags=event.tags,
)
EventMapping.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
date_added=event.datetime,
)
UserReport.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
name='Log Hat',
email='*****@*****.**',
comments='Quack',
)
if release:
Release.get_or_create(
project=project,
version=event.get_tag('sentry:release'),
date_added=event.datetime,
)
features.record([event])
return event
def create_message_event(template, parameters, environment, release):
i = next(sequence)
event_id = uuid.UUID(fields=(i, 0x0, 0x1000, 0x80, 0x80,
0x808080808080)).hex
tags = [["color", next(tag_values)]]
if environment:
tags.append(["environment", environment])
if release:
tags.append(["sentry:release", release])
event = Event.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
message="%s" % (id, ),
datetime=now + shift(i),
data={
"environment": environment,
"type": "default",
"metadata": {
"title": template % parameters
},
"logentry": {
"message": template,
"params": parameters,
"formatted": template % parameters,
},
"user": next(user_values),
"tags": tags,
},
)
with self.tasks():
Group.objects.add_tags(
source,
Environment.objects.get(
organization_id=project.organization_id,
name=environment),
tags=event.tags,
)
UserReport.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
name="Log Hat",
email="*****@*****.**",
comments="Quack",
)
if release:
Release.get_or_create(
project=project,
version=event.get_tag("sentry:release"),
date_added=event.datetime,
)
features.record([event])
return event
def save(self, project, raw=False):
from sentry.tasks.post_process import index_event_tags
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
level = data.pop('level')
culprit = data.pop('transaction', None)
if not culprit:
culprit = data.pop('culprit', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
dist = data.pop('dist', None)
environment = data.pop('environment', None)
# unused
time_spent = data.pop('time_spent', None)
message = data.pop('message', '')
if not culprit:
# if we generate an implicit culprit, lets not call it a
# transaction
transaction_name = None
culprit = generate_culprit(data, platform=platform)
else:
transaction_name = culprit
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'platform': platform,
}
event = Event(
project_id=project.id,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs
)
event._project_cache = project
# convert this to a dict to ensure we're only storing one value per key
# as most parts of Sentry dont currently play well with multiple values
tags = dict(data.get('tags') or [])
tags['level'] = LOG_LEVELS[level]
if logger_name:
tags['logger'] = logger_name
if server_name:
tags['server_name'] = server_name
if site:
tags['site'] = site
if environment:
tags['environment'] = environment
if transaction_name:
tags['transaction'] = transaction_name
if release:
# dont allow a conflicting 'release' tag
if 'release' in tags:
del tags['release']
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
tags['sentry:release'] = release.version
if dist and release:
dist = release.add_dist(dist, date)
tags['sentry:dist'] = dist.name
else:
dist = None
event_user = self._get_event_user(project, data)
if event_user:
# dont allow a conflicting 'user' tag
if 'user' in tags:
del tags['user']
tags['sentry:user'] = event_user.tag_value
# At this point we want to normalize the in_app values in case the
# clients did not set this appropriately so far.
normalize_in_app(data)
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event, _with_transaction=False)
if added_tags:
# plugins should not override user provided tags
for key, value in added_tags:
tags.setdefault(key, value)
# tags are stored as a tuple
tags = tags.items()
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
data['fingerprint'] = fingerprint or ['{{ default }}']
for path, iface in six.iteritems(event.interfaces):
data['tags'].extend(iface.iter_tags())
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.get_path(), None)
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = [md5_from_hash(h) for h in get_hashes_from_fingerprint(event, fingerprint)]
elif checksum:
hashes = [checksum]
data['checksum'] = checksum
else:
hashes = [md5_from_hash(h) for h in get_hashes_for_event(event)]
# TODO(dcramer): temp workaround for complexity
data['message'] = message
event_type = eventtypes.get(data.get('type', 'default'))(data)
event_metadata = event_type.get_metadata()
# TODO(dcramer): temp workaround for complexity
del data['message']
data['type'] = event_type.key
data['metadata'] = event_metadata
# index components into ``Event.message``
# See GH-3248
if event_type.key != 'default':
if 'sentry.interfaces.Message' in data and \
data['sentry.interfaces.Message']['message'] != message:
message = u'{} {}'.format(
message,
data['sentry.interfaces.Message']['message'],
)
if not message:
message = ''
elif not isinstance(message, six.string_types):
message = force_text(message)
for value in six.itervalues(event_metadata):
value_u = force_text(value, errors='replace')
if value_u not in message:
message = u'{} {}'.format(message, value_u)
if culprit and culprit not in message:
culprit_u = force_text(culprit, errors='replace')
message = u'{} {}'.format(message, culprit_u)
message = trim(message.strip(), settings.SENTRY_MAX_MESSAGE_LENGTH)
event.message = message
kwargs['message'] = message
group_kwargs = kwargs.copy()
group_kwargs.update(
{
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'active_at': date,
'data': {
'last_received':
event.data.get('received') or float(event.datetime.strftime('%s')),
'type':
event_type.key,
# we cache the events metadata on the group to ensure its
# accessible in the stream
'metadata':
event_metadata,
},
}
)
if release:
group_kwargs['first_release'] = release
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **group_kwargs
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project, group=group, event_id=event_id)
except IntegrityError:
self.logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': EventMapping.__name__,
}
)
return event
environment = Environment.get_or_create(
project=project,
name=environment,
)
if release:
ReleaseEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date,
)
counters = [
(tsdb.models.group, group.id),
(tsdb.models.project, project.id),
]
if release:
counters.append((tsdb.models.release, release.id))
tsdb.incr_multi(counters, timestamp=event.datetime)
frequencies = [
# (tsdb.models.frequent_projects_by_organization, {
# project.organization_id: {
# project.id: 1,
# },
# }),
# (tsdb.models.frequent_issues_by_project, {
# project.id: {
# group.id: 1,
# },
# })
(tsdb.models.frequent_environments_by_group, {
group.id: {
environment.id: 1,
},
})
]
if release:
frequencies.append(
(tsdb.models.frequent_releases_by_group, {
group.id: {
grouprelease.id: 1,
},
})
)
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': Event.__name__,
}
)
return event
index_event_tags.delay(
organization_id=project.organization_id,
project_id=project.id,
group_id=group.id,
event_id=event.id,
tags=tags,
)
if event_user:
tsdb.record_multi(
(
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value, )),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value, )),
),
timestamp=event.datetime
)
if is_new and release:
buffer.incr(
ReleaseProject, {'new_groups': 1}, {
'release_id': release.id,
'project_id': project.id,
}
)
safe_execute(Group.objects.add_tags, group, tags, _with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send(project=project, group=group, sender=Project)
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info('post_process.skip.raw_event', extra={'event_id': event.id})
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def save(self, project, raw=False):
from sentry.tasks.post_process import index_event_tags
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
level = data.pop('level')
culprit = data.pop('culprit', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
environment = data.pop('environment', None)
# unused
time_spent = data.pop('time_spent', None)
message = data.pop('message', '')
if not culprit:
# if we generate an implicit culprit, lets not call it a
# transaction
transaction_name = None
culprit = generate_culprit(data, platform=platform)
else:
transaction_name = culprit
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'platform': platform,
}
event = Event(
project_id=project.id,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs
)
# convert this to a dict to ensure we're only storing one value per key
# as most parts of Sentry dont currently play well with multiple values
tags = dict(data.get('tags') or [])
tags['level'] = LOG_LEVELS[level]
if logger_name:
tags['logger'] = logger_name
if server_name:
tags['server_name'] = server_name
if site:
tags['site'] = site
if environment:
tags['environment'] = environment
if transaction_name:
tags['transaction'] = transaction_name
if release:
# dont allow a conflicting 'release' tag
if 'release' in tags:
del tags['release']
tags['sentry:release'] = release
event_user = self._get_event_user(project, data)
if event_user:
# dont allow a conflicting 'user' tag
if 'user' in tags:
del tags['user']
tags['sentry:user'] = event_user.tag_value
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event,
_with_transaction=False)
if added_tags:
# plugins should not override user provided tags
for key, value in added_tags:
tags.setdefault(key, value)
# tags are stored as a tuple
tags = tags.items()
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
data['fingerprint'] = fingerprint or ['{{ default }}']
for path, iface in six.iteritems(event.interfaces):
data['tags'].extend(iface.iter_tags())
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.get_path(), None)
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = [
md5_from_hash(h)
for h in get_hashes_from_fingerprint(event, fingerprint)
]
elif checksum:
hashes = [checksum]
data['checksum'] = checksum
else:
hashes = [
md5_from_hash(h)
for h in get_hashes_for_event(event)
]
# TODO(dcramer): temp workaround for complexity
data['message'] = message
event_type = eventtypes.get(data.get('type', 'default'))(data)
event_metadata = event_type.get_metadata()
# TODO(dcramer): temp workaround for complexity
del data['message']
data['type'] = event_type.key
data['metadata'] = event_metadata
# index components into ``Event.message``
# See GH-3248
if event_type.key != 'default':
if 'sentry.interfaces.Message' in data and \
data['sentry.interfaces.Message']['message'] != message:
message = u'{} {}'.format(
message,
data['sentry.interfaces.Message']['message'],
)
if not message:
message = ''
elif not isinstance(message, six.string_types):
message = force_text(message)
for value in six.itervalues(event_metadata):
value_u = force_text(value, errors='replace')
if value_u not in message:
message = u'{} {}'.format(message, value_u)
if culprit and culprit not in message:
culprit_u = force_text(culprit, errors='replace')
message = u'{} {}'.format(message, culprit_u)
message = trim(message.strip(), settings.SENTRY_MAX_MESSAGE_LENGTH)
event.message = message
kwargs['message'] = message
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'active_at': date,
'data': {
'last_received': event.data.get('received') or float(event.datetime.strftime('%s')),
'type': event_type.key,
# we cache the events metadata on the group to ensure its
# accessible in the stream
'metadata': event_metadata,
},
})
if release:
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
group_kwargs['first_release'] = release
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event,
hashes=hashes,
release=release,
**group_kwargs
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(
project=project, group=group, event_id=event_id)
except IntegrityError:
self.logger.info('duplicate.found', extra={'event_id': event.id}, exc_info=True)
return event
environment = Environment.get_or_create(
project=project,
name=environment,
)
if release:
ReleaseEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date,
)
counters = [
(tsdb.models.group, group.id),
(tsdb.models.project, project.id),
]
if release:
counters.append((tsdb.models.release, release.id))
tsdb.incr_multi(counters, timestamp=event.datetime)
frequencies = [
# (tsdb.models.frequent_projects_by_organization, {
# project.organization_id: {
# project.id: 1,
# },
# }),
# (tsdb.models.frequent_issues_by_project, {
# project.id: {
# group.id: 1,
# },
# })
(tsdb.models.frequent_environments_by_group, {
group.id: {
environment.id: 1,
},
})
]
if release:
frequencies.append(
(tsdb.models.frequent_releases_by_group, {
group.id: {
grouprelease.id: 1,
},
})
)
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
UserReport.objects.filter(
project=project, event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info('duplicate.found', extra={'event_id': event.id}, exc_info=True)
return event
index_event_tags.delay(
project_id=project.id,
group_id=group.id,
event_id=event.id,
tags=tags,
)
if event_user:
tsdb.record_multi((
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value,)),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value,)),
), timestamp=event.datetime)
if is_new and release:
buffer.incr(Release, {'new_groups': 1}, {
'id': release.id,
})
safe_execute(Group.objects.add_tags, group, tags,
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send(project=project, group=group, sender=Project)
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info('post_process.skip.raw_event', extra={'event_id': event.id})
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def save(self,
project_id,
raw=False,
assume_normalized=False,
cache_key=None):
"""
We re-insert events with duplicate IDs into Snuba, which is responsible
for deduplicating events. Since deduplication in Snuba is on the primary
key (based on event ID, project ID and day), events with same IDs are only
deduplicated if their timestamps fall on the same day. The latest event
always wins and overwrites the value of events received earlier in that day.
Since we increment counters and frequencies here before events get inserted
to eventstream these numbers may be larger than the total number of
events if we receive duplicate event IDs that fall on the same day
(that do not hit cache first).
"""
# Normalize if needed
if not self._normalized:
if not assume_normalized:
self.normalize()
self._normalized = True
data = self._data
project = Project.objects.get_from_cache(id=project_id)
project._organization_cache = Organization.objects.get_from_cache(
id=project.organization_id)
# Pull out the culprit
culprit = self.get_culprit()
# Pull the toplevel data we're interested in
level = data.get("level")
# TODO(mitsuhiko): this code path should be gone by July 2018.
# This is going to be fine because no code actually still depends
# on integers here. When we need an integer it will be converted
# into one later. Old workers used to send integers here.
if level is not None and isinstance(level, six.integer_types):
level = LOG_LEVELS[level]
transaction_name = data.get("transaction")
logger_name = data.get("logger")
release = data.get("release")
dist = data.get("dist")
environment = data.get("environment")
recorded_timestamp = data.get("timestamp")
# We need to swap out the data with the one internal to the newly
# created event object
event = self._get_event_instance(project_id=project_id)
self._data = data = event.data.data
event._project_cache = project
date = event.datetime
platform = event.platform
event_id = event.event_id
if transaction_name:
transaction_name = force_text(transaction_name)
# Right now the event type is the signal to skip the group. This
# is going to change a lot.
if event.get_event_type() == "transaction":
issueless_event = True
else:
issueless_event = False
# Some of the data that are toplevel attributes are duplicated
# into tags (logger, level, environment, transaction). These are
# different from legacy attributes which are normalized into tags
# ahead of time (site, server_name).
setdefault_path(data, "tags", value=[])
set_tag(data, "level", level)
if logger_name:
set_tag(data, "logger", logger_name)
if environment:
set_tag(data, "environment", environment)
if transaction_name:
set_tag(data, "transaction", transaction_name)
if release:
# dont allow a conflicting 'release' tag
pop_tag(data, "release")
release = Release.get_or_create(project=project,
version=release,
date_added=date)
set_tag(data, "sentry:release", release.version)
if dist and release:
dist = release.add_dist(dist, date)
# dont allow a conflicting 'dist' tag
pop_tag(data, "dist")
set_tag(data, "sentry:dist", dist.name)
else:
dist = None
event_user = self._get_event_user(project, data)
if event_user:
# dont allow a conflicting 'user' tag
pop_tag(data, "user")
set_tag(data, "sentry:user", event_user.tag_value)
# At this point we want to normalize the in_app values in case the
# clients did not set this appropriately so far.
grouping_config = load_grouping_config(
get_grouping_config_dict_for_event_data(data, project))
normalize_stacktraces_for_grouping(data, grouping_config)
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags,
event,
_with_transaction=False)
if added_tags:
# plugins should not override user provided tags
for key, value in added_tags:
if get_tag(data, key) is None:
set_tag(data, key, value)
for path, iface in six.iteritems(event.interfaces):
for k, v in iface.iter_tags():
set_tag(data, k, v)
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.path, None)
# The active grouping config was put into the event in the
# normalize step before. We now also make sure that the
# fingerprint was set to `'{{ default }}' just in case someone
# removed it from the payload. The call to get_hashes will then
# look at `grouping_config` to pick the right parameters.
data["fingerprint"] = data.get("fingerprint") or ["{{ default }}"]
apply_server_fingerprinting(
data, get_fingerprinting_config_for_project(project))
# Here we try to use the grouping config that was requested in the
# event. If that config has since been deleted (because it was an
# experimental grouping config) we fall back to the default.
try:
hashes = event.get_hashes()
except GroupingConfigNotFound:
data["grouping_config"] = get_grouping_config_dict_for_project(
project)
hashes = event.get_hashes()
data["hashes"] = hashes
# we want to freeze not just the metadata and type in but also the
# derived attributes. The reason for this is that we push this
# data into kafka for snuba processing and our postprocessing
# picks up the data right from the snuba topic. For most usage
# however the data is dynamically overridden by Event.title and
# Event.location (See Event.as_dict)
materialized_metadata = self.materialize_metadata()
data.update(materialized_metadata)
data["culprit"] = culprit
received_timestamp = event.data.get("received") or float(
event.datetime.strftime("%s"))
if not issueless_event:
# The group gets the same metadata as the event when it's flushed but
# additionally the `last_received` key is set. This key is used by
# _save_aggregate.
group_metadata = dict(materialized_metadata)
group_metadata["last_received"] = received_timestamp
kwargs = {
"platform": platform,
"message": event.search_message,
"culprit": culprit,
"logger": logger_name,
"level": LOG_LEVELS_MAP.get(level),
"last_seen": date,
"first_seen": date,
"active_at": date,
"data": group_metadata,
}
if release:
kwargs["first_release"] = release
try:
group, is_new, is_regression = self._save_aggregate(
event=event, hashes=hashes, release=release, **kwargs)
except HashDiscarded:
event_discarded.send_robust(project=project,
sender=EventManager)
metrics.incr(
"events.discarded",
skip_internal=True,
tags={
"organization_id": project.organization_id,
"platform": platform
},
)
raise
else:
event_saved.send_robust(project=project,
event_size=event.size,
sender=EventManager)
event.group = group
else:
group = None
is_new = False
is_regression = False
event_saved.send_robust(project=project,
event_size=event.size,
sender=EventManager)
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
environment = Environment.get_or_create(project=project,
name=environment)
if group:
group_environment, is_new_group_environment = GroupEnvironment.get_or_create(
group_id=group.id,
environment_id=environment.id,
defaults={"first_release": release if release else None},
)
else:
is_new_group_environment = False
if release:
ReleaseEnvironment.get_or_create(project=project,
release=release,
environment=environment,
datetime=date)
ReleaseProjectEnvironment.get_or_create(project=project,
release=release,
environment=environment,
datetime=date)
if group:
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date)
counters = [(tsdb.models.project, project.id)]
if group:
counters.append((tsdb.models.group, group.id))
if release:
counters.append((tsdb.models.release, release.id))
tsdb.incr_multi(counters,
timestamp=event.datetime,
environment_id=environment.id)
frequencies = []
if group:
frequencies.append((tsdb.models.frequent_environments_by_group, {
group.id: {
environment.id: 1
}
}))
if release:
frequencies.append((tsdb.models.frequent_releases_by_group, {
group.id: {
grouprelease.id: 1
}
}))
if frequencies:
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
if group:
UserReport.objects.filter(project=project,
event_id=event_id).update(
group=group, environment=environment)
# Enusre the _metrics key exists. This is usually created during
# and prefilled with ingestion sizes.
event_metrics = event.data.get("_metrics") or {}
event.data["_metrics"] = event_metrics
# Capture the actual size that goes into node store.
event_metrics["bytes.stored.event"] = len(
json.dumps(dict(event.data.items())))
# Load attachments first, but persist them at the very last after
# posting to eventstream to make sure all counters and eventstream are
# incremented for sure.
attachments = self.get_attachments(cache_key, event)
for attachment in attachments:
key = "bytes.stored.%s" % (attachment.type, )
event_metrics[key] = (event_metrics.get(key) or 0) + len(
attachment.data)
# Write the event to Nodestore
event.data.save()
if event_user:
counters = [(tsdb.models.users_affected_by_project, project.id,
(event_user.tag_value, ))]
if group:
counters.append((tsdb.models.users_affected_by_group, group.id,
(event_user.tag_value, )))
tsdb.record_multi(counters,
timestamp=event.datetime,
environment_id=environment.id)
if release:
if is_new:
buffer.incr(
ReleaseProject,
{"new_groups": 1},
{
"release_id": release.id,
"project_id": project.id
},
)
if is_new_group_environment:
buffer.incr(
ReleaseProjectEnvironment,
{"new_issues_count": 1},
{
"project_id": project.id,
"release_id": release.id,
"environment_id": environment.id,
},
)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send_robust(project=project,
event=event,
sender=Project)
eventstream.insert(
group=group,
event=event,
is_new=is_new,
is_regression=is_regression,
is_new_group_environment=is_new_group_environment,
primary_hash=hashes[0],
# We are choosing to skip consuming the event back
# in the eventstream if it's flagged as raw.
# This means that we want to publish the event
# through the event stream, but we don't care
# about post processing and handling the commit.
skip_consume=raw,
)
# Do this last to ensure signals get emitted even if connection to the
# file store breaks temporarily.
self.save_attachments(attachments, event)
metric_tags = {"from_relay": "_relay_processed" in self._data}
metrics.timing("events.latency",
received_timestamp - recorded_timestamp,
tags=metric_tags)
metrics.timing("events.size.data.post_save",
event.size,
tags=metric_tags)
metrics.incr(
"events.post_save.normalize.errors",
amount=len(self._data.get("errors") or ()),
tags=metric_tags,
)
return event
def save(self, project, raw=False):
from sentry.tasks.post_process import index_event_tags
data = self.data
project = Project.objects.get_from_cache(id=project)
# Check to make sure we're not about to do a bunch of work that's
# already been done if we've processed an event with this ID. (This
# isn't a perfect solution -- this doesn't handle ``EventMapping`` and
# there's a race condition between here and when the event is actually
# saved, but it's an improvement. See GH-7677.)
try:
event = Event.objects.get(
project_id=project.id,
event_id=data['event_id'],
)
except Event.DoesNotExist:
pass
else:
self.logger.info('duplicate.found',
exc_info=True,
extra={
'event_uuid': data['event_id'],
'project_id': project.id,
'model': Event.__name__,
})
return event
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
level = data.pop('level')
culprit = data.pop('transaction', None)
if not culprit:
culprit = data.pop('culprit', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
dist = data.pop('dist', None)
environment = data.pop('environment', None)
# unused
time_spent = data.pop('time_spent', None)
message = data.pop('message', '')
if not culprit:
# if we generate an implicit culprit, lets not call it a
# transaction
transaction_name = None
culprit = generate_culprit(data, platform=platform)
else:
transaction_name = culprit
culprit = force_text(culprit)
recorded_timestamp = data.pop('timestamp')
date = datetime.fromtimestamp(recorded_timestamp)
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'platform': platform,
}
event = Event(project_id=project.id,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs)
event._project_cache = project
# convert this to a dict to ensure we're only storing one value per key
# as most parts of Sentry dont currently play well with multiple values
tags = dict(data.get('tags') or [])
tags['level'] = LOG_LEVELS[level]
if logger_name:
tags['logger'] = logger_name
if server_name:
tags['server_name'] = server_name
if site:
tags['site'] = site
if environment:
tags['environment'] = environment
if transaction_name:
tags['transaction'] = transaction_name
if release:
# dont allow a conflicting 'release' tag
if 'release' in tags:
del tags['release']
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
tags['sentry:release'] = release.version
if dist and release:
dist = release.add_dist(dist, date)
tags['sentry:dist'] = dist.name
else:
dist = None
event_user = self._get_event_user(project, data)
if event_user:
# dont allow a conflicting 'user' tag
if 'user' in tags:
del tags['user']
tags['sentry:user'] = event_user.tag_value
# At this point we want to normalize the in_app values in case the
# clients did not set this appropriately so far.
normalize_in_app(data)
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags,
event,
_with_transaction=False)
if added_tags:
# plugins should not override user provided tags
for key, value in added_tags:
tags.setdefault(key, value)
for path, iface in six.iteritems(event.interfaces):
for k, v in iface.iter_tags():
tags[k] = v
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.get_path(), None)
# tags are stored as a tuple
tags = tags.items()
data['tags'] = tags
data['fingerprint'] = fingerprint or ['{{ default }}']
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = [
md5_from_hash(h)
for h in get_hashes_from_fingerprint(event, fingerprint)
]
elif checksum:
if HASH_RE.match(checksum):
hashes = [checksum]
else:
hashes = [md5_from_hash([checksum]), checksum]
data['checksum'] = checksum
else:
hashes = [md5_from_hash(h) for h in get_hashes_for_event(event)]
# TODO(dcramer): temp workaround for complexity
data['message'] = message
event_type = eventtypes.get(data.get('type', 'default'))(data)
event_metadata = event_type.get_metadata()
# TODO(dcramer): temp workaround for complexity
del data['message']
data['type'] = event_type.key
data['metadata'] = event_metadata
# index components into ``Event.message``
# See GH-3248
if event_type.key != 'default':
if 'sentry.interfaces.Message' in data and \
data['sentry.interfaces.Message']['message'] != message:
message = u'{} {}'.format(
message,
data['sentry.interfaces.Message']['message'],
)
if not message:
message = ''
elif not isinstance(message, six.string_types):
message = force_text(message)
for value in six.itervalues(event_metadata):
value_u = force_text(value, errors='replace')
if value_u not in message:
message = u'{} {}'.format(message, value_u)
if culprit and culprit not in message:
culprit_u = force_text(culprit, errors='replace')
message = u'{} {}'.format(message, culprit_u)
message = trim(message.strip(), settings.SENTRY_MAX_MESSAGE_LENGTH)
event.message = message
kwargs['message'] = message
received_timestamp = event.data.get('received') or float(
event.datetime.strftime('%s'))
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'active_at': date,
'data': {
'last_received': received_timestamp,
'type': event_type.key,
# we cache the events metadata on the group to ensure its
# accessible in the stream
'metadata': event_metadata,
},
})
if release:
group_kwargs['first_release'] = release
try:
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **group_kwargs)
except HashDiscarded:
event_discarded.send_robust(
project=project,
sender=EventManager,
)
metrics.incr(
'events.discarded',
skip_internal=True,
tags={
'organization_id': project.organization_id,
'platform': platform,
},
)
raise
else:
event_saved.send_robust(
project=project,
sender=EventManager,
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
# When an event was sampled, the canonical source of truth
# is the EventMapping table since we aren't going to be writing out an actual
# Event row. Otherwise, if the Event isn't being sampled, we can safely
# rely on the Event table itself as the source of truth and ignore
# EventMapping since it's redundant information.
if is_sample:
try:
with transaction.atomic(
using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project,
group=group,
event_id=event_id)
except IntegrityError:
self.logger.info('duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': EventMapping.__name__,
})
return event
environment = Environment.get_or_create(
project=project,
name=environment,
)
group_environment, is_new_group_environment = GroupEnvironment.get_or_create(
group_id=group.id,
environment_id=environment.id,
defaults={
'first_release_id': release.id if release else None,
},
)
if release:
ReleaseEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
ReleaseProjectEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date,
)
counters = [
(tsdb.models.group, group.id),
(tsdb.models.project, project.id),
]
if release:
counters.append((tsdb.models.release, release.id))
tsdb.incr_multi(counters,
timestamp=event.datetime,
environment_id=environment.id)
frequencies = [
# (tsdb.models.frequent_projects_by_organization, {
# project.organization_id: {
# project.id: 1,
# },
# }),
# (tsdb.models.frequent_issues_by_project, {
# project.id: {
# group.id: 1,
# },
# })
(tsdb.models.frequent_environments_by_group, {
group.id: {
environment.id: 1,
},
})
]
if release:
frequencies.append((tsdb.models.frequent_releases_by_group, {
group.id: {
grouprelease.id: 1,
},
}))
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(
group=group,
environment=environment,
)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info('duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': Event.__name__,
})
return event
index_event_tags.delay(
organization_id=project.organization_id,
project_id=project.id,
group_id=group.id,
environment_id=environment.id,
event_id=event.id,
tags=tags,
date_added=event.datetime,
)
if event_user:
tsdb.record_multi(
(
(tsdb.models.users_affected_by_group, group.id,
(event_user.tag_value, )),
(tsdb.models.users_affected_by_project, project.id,
(event_user.tag_value, )),
),
timestamp=event.datetime,
environment_id=environment.id,
)
if release:
if is_new:
buffer.incr(ReleaseProject, {'new_groups': 1}, {
'release_id': release.id,
'project_id': project.id,
})
if is_new_group_environment:
buffer.incr(ReleaseProjectEnvironment, {'new_issues_count': 1},
{
'project_id': project.id,
'release_id': release.id,
'environment_id': environment.id,
})
safe_execute(Group.objects.add_tags,
group,
environment,
tags,
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send(project=project,
group=group,
sender=Project)
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
is_new_group_environment=is_new_group_environment,
primary_hash=hashes[0],
)
else:
self.logger.info('post_process.skip.raw_event',
extra={'event_id': event.id})
metrics.timing(
'events.latency',
received_timestamp - recorded_timestamp,
tags={
'project_id': project.id,
},
)
return event
def save(self, project, raw=False):
# TODO: culprit should default to "most recent" frame in stacktraces when
# it's not provided.
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
message = data.pop('message')
level = data.pop('level')
culprit = data.pop('culprit', None) or ''
time_spent = data.pop('time_spent', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'message': message,
'platform': platform,
}
event = Event(project=project,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs)
tags = data.get('tags') or []
tags.append(('level', LOG_LEVELS[level]))
if logger_name:
tags.append(('logger', logger_name))
if server_name:
tags.append(('server_name', server_name))
if site:
tags.append(('site', site))
if release:
# TODO(dcramer): we should ensure we create Release objects
tags.append(('sentry:release', release))
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags,
event,
_with_transaction=False)
if added_tags:
tags.extend(added_tags)
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
# Calculate the checksum from the first highest scoring interface
if checksum:
hashes = [checksum]
elif fingerprint:
hashes = map(md5_from_hash,
get_hashes_from_fingerprint(event, fingerprint))
else:
hashes = map(md5_from_hash, get_hashes_for_event(event))
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'time_spent_total': time_spent or 0,
'time_spent_count': time_spent and 1 or 0,
})
if release:
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
group_kwargs['first_release'] = release
Activity.objects.create(
type=Activity.RELEASE,
project=project,
ident=release,
data={'version': release},
datetime=date,
)
group, is_new, is_regression, is_sample = safe_execute(
self._save_aggregate, event=event, hashes=hashes, **group_kwargs)
using = group._state.db
event.group = group
try:
with transaction.atomic():
EventMapping.objects.create(project=project,
group=group,
event_id=event_id)
except IntegrityError:
self.logger.info('Duplicate EventMapping found for event_id=%s',
event_id)
return event
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic():
event.save()
except IntegrityError:
self.logger.info('Duplicate Event found for event_id=%s',
event_id)
return event
if is_new and release:
buffer.incr(Release, {'new_groups': 1}, {
'id': release.id,
})
safe_execute(Group.objects.add_tags,
group,
tags,
_with_transaction=False)
if not raw:
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info(
'Raw event passed; skipping post process for event_id=%s',
event_id)
index_event.delay(event)
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def save(self, project, raw=False):
from sentry.tasks.post_process import index_event_tags
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop("event_id")
message = data.pop("message")
level = data.pop("level")
culprit = data.pop("culprit", None)
logger_name = data.pop("logger", None)
server_name = data.pop("server_name", None)
site = data.pop("site", None)
checksum = data.pop("checksum", None)
fingerprint = data.pop("fingerprint", None)
platform = data.pop("platform", None)
release = data.pop("release", None)
environment = data.pop("environment", None)
# unused
time_spent = data.pop("time_spent", None)
if not culprit:
culprit = generate_culprit(data, platform=platform)
date = datetime.fromtimestamp(data.pop("timestamp"))
date = date.replace(tzinfo=timezone.utc)
kwargs = {"message": message, "platform": platform}
event = Event(
project_id=project.id, event_id=event_id, data=data, time_spent=time_spent, datetime=date, **kwargs
)
tags = data.get("tags") or []
tags.append(("level", LOG_LEVELS[level]))
if logger_name:
tags.append(("logger", logger_name))
if server_name:
tags.append(("server_name", server_name))
if site:
tags.append(("site", site))
if release:
# TODO(dcramer): we should ensure we create Release objects
tags.append(("sentry:release", release))
if environment:
tags.append(("environment", environment))
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event, _with_transaction=False)
if added_tags:
tags.extend(added_tags)
event_user = self._get_event_user(project, data)
if event_user:
tags.append(("sentry:user", event_user.tag_value))
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data["tags"] = tags
data["fingerprint"] = fingerprint or ["{{ default }}"]
# Get rid of ephemeral interface data
for interface_class, _ in iter_interfaces():
interface = interface_class()
if interface.ephemeral:
data.pop(interface.get_path(), None)
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = map(md5_from_hash, get_hashes_from_fingerprint(event, fingerprint))
elif checksum:
hashes = [checksum]
else:
hashes = map(md5_from_hash, get_hashes_for_event(event))
# TODO(dcramer): temp workaround for complexity
data["message"] = message
event_type = eventtypes.get(data.get("type", "default"))(data)
group_kwargs = kwargs.copy()
group_kwargs.update(
{
"culprit": culprit,
"logger": logger_name,
"level": level,
"last_seen": date,
"first_seen": date,
"data": {
"last_received": event.data.get("received") or float(event.datetime.strftime("%s")),
"type": event_type.key,
# we cache the events metadata on the group to ensure its
# accessible in the stream
"metadata": event_type.get_metadata(),
},
}
)
# TODO(dcramer): temp workaround for complexity
del data["message"]
if release:
release = Release.get_or_create(project=project, version=release, date_added=date)
group_kwargs["first_release"] = release
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **group_kwargs
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project, group=group, event_id=event_id)
except IntegrityError:
self.logger.info("Duplicate EventMapping found for event_id=%s", event_id, exc_info=True)
return event
UserReport.objects.filter(project=project, event_id=event_id).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info("Duplicate Event found for event_id=%s", event_id, exc_info=True)
return event
index_event_tags.delay(project_id=project.id, event_id=event.id, tags=tags)
if event_user:
tsdb.record_multi(
(
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value,)),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value,)),
),
timestamp=event.datetime,
)
if is_new and release:
buffer.incr(Release, {"new_groups": 1}, {"id": release.id})
safe_execute(Group.objects.add_tags, group, tags, _with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send(project=project, group=group, sender=Project)
post_process_group.delay(
group=group, event=event, is_new=is_new, is_sample=is_sample, is_regression=is_regression
)
else:
self.logger.info("Raw event passed; skipping post process for event_id=%s", event_id)
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def save(self, project, raw=False):
from sentry.tasks.post_process import index_event_tags
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
level = data.pop('level')
culprit = data.pop('culprit', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
environment = data.pop('environment', None)
# unused
time_spent = data.pop('time_spent', None)
message = data.pop('message', '')
if not culprit:
culprit = generate_culprit(data, platform=platform)
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'platform': platform,
}
event = Event(project_id=project.id,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs)
tags = data.get('tags') or []
tags.append(('level', LOG_LEVELS[level]))
if logger_name:
tags.append(('logger', logger_name))
if server_name:
tags.append(('server_name', server_name))
if site:
tags.append(('site', site))
if release:
# TODO(dcramer): we should ensure we create Release objects
tags.append(('sentry:release', release))
if environment:
tags.append(('environment', environment))
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags,
event,
_with_transaction=False)
if added_tags:
tags.extend(added_tags)
event_user = self._get_event_user(project, data)
if event_user:
tags.append(('sentry:user', event_user.tag_value))
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
data['fingerprint'] = fingerprint or ['{{ default }}']
# Get rid of ephemeral interface data
for interface_class, _ in iter_interfaces():
interface = interface_class()
if interface.ephemeral:
data.pop(interface.get_path(), None)
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = map(md5_from_hash,
get_hashes_from_fingerprint(event, fingerprint))
elif checksum:
hashes = [checksum]
else:
hashes = map(md5_from_hash, get_hashes_for_event(event))
# TODO(dcramer): temp workaround for complexity
data['message'] = message
event_type = eventtypes.get(data.get('type', 'default'))(data)
event_metadata = event_type.get_metadata()
# TODO(dcramer): temp workaround for complexity
del data['message']
data['type'] = event_type.key
data['metadata'] = event_metadata
# index components into ``Event.message``
# See GH-3248
if event_type.key != 'default':
if 'sentry.interfaces.Message' in data and \
data['sentry.interfaces.Message']['message'] != message:
message = u'{} {}'.format(
message,
data['sentry.interfaces.Message']['message'],
)
if not message:
message = ''
elif not isinstance(message, basestring):
message = force_text(message)
for value in event_metadata.itervalues():
value_u = force_text(value, errors='replace')
if value_u not in message:
message = u'{} {}'.format(message, value_u)
message = trim(message.strip(), settings.SENTRY_MAX_MESSAGE_LENGTH)
event.message = message
kwargs['message'] = message
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'data': {
'last_received':
event.data.get('received')
or float(event.datetime.strftime('%s')),
'type':
event_type.key,
# we cache the events metadata on the group to ensure its
# accessible in the stream
'metadata':
event_metadata,
},
})
if release:
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
group_kwargs['first_release'] = release
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **group_kwargs)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project,
group=group,
event_id=event_id)
except IntegrityError:
self.logger.info('Duplicate EventMapping found for event_id=%s',
event_id,
exc_info=True)
return event
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info('Duplicate Event found for event_id=%s',
event_id,
exc_info=True)
return event
index_event_tags.delay(
project_id=project.id,
group_id=group.id,
event_id=event.id,
tags=tags,
)
if event_user:
tsdb.record_multi((
(tsdb.models.users_affected_by_group, group.id,
(event_user.tag_value, )),
(tsdb.models.users_affected_by_project, project.id,
(event_user.tag_value, )),
),
timestamp=event.datetime)
if is_new and release:
buffer.incr(Release, {'new_groups': 1}, {
'id': release.id,
})
safe_execute(Group.objects.add_tags,
group,
tags,
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send(project=project,
group=group,
sender=Project)
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info(
'Raw event passed; skipping post process for event_id=%s',
event_id)
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def save(self, project, raw=False):
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
message = data.pop('message')
level = data.pop('level')
culprit = data.pop('culprit', None)
time_spent = data.pop('time_spent', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
environment = data.pop('environment', None)
if not culprit:
culprit = generate_culprit(data)
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'message': message,
'platform': platform,
}
event = Event(
project=project,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs
)
tags = data.get('tags') or []
tags.append(('level', LOG_LEVELS[level]))
if logger_name:
tags.append(('logger', logger_name))
if server_name:
tags.append(('server_name', server_name))
if site:
tags.append(('site', site))
if release:
# TODO(dcramer): we should ensure we create Release objects
tags.append(('sentry:release', release))
if environment:
tags.append(('environment', environment))
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event,
_with_transaction=False)
if added_tags:
tags.extend(added_tags)
event_user = self._get_event_user(project, data)
if event_user:
tags.append(('sentry:user', event_user.tag_value))
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
data['fingerprint'] = fingerprint or '{{ default }}'
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = map(md5_from_hash, get_hashes_from_fingerprint(event, fingerprint))
elif checksum:
hashes = [checksum]
else:
hashes = map(md5_from_hash, get_hashes_for_event(event))
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'time_spent_total': time_spent or 0,
'time_spent_count': time_spent and 1 or 0,
})
if release:
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
group_kwargs['first_release'] = release
Activity.objects.create(
type=Activity.RELEASE,
project=project,
ident=release,
data={'version': release},
datetime=date,
)
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event,
hashes=hashes,
release=release,
**group_kwargs
)
event.group = group
event.group_id = group.id
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
try:
with transaction.atomic():
EventMapping.objects.create(
project=project, group=group, event_id=event_id)
except IntegrityError:
self.logger.info('Duplicate EventMapping found for event_id=%s', event_id)
return event
UserReport.objects.filter(
project=project, event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic():
event.save()
except IntegrityError:
self.logger.info('Duplicate Event found for event_id=%s', event_id)
return event
if event_user:
tsdb.record_multi((
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value,)),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value,)),
), timestamp=event.datetime)
if is_new and release:
buffer.incr(Release, {'new_groups': 1}, {
'id': release.id,
})
safe_execute(Group.objects.add_tags, group, tags,
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info('Raw event passed; skipping post process for event_id=%s', event_id)
index_event.delay(event)
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def setUp(self):
self.project = self.create_project()
self.release = Release.get_or_create(self.project, '1.0')
self.environment1 = Environment.get_or_create(self.project, 'prod')
self.environment2 = Environment.get_or_create(self.project, 'staging')
self.timestamp = 1403007314
def save(self, project, raw=False):
from sentry.tasks.post_process import index_event_tags
data = self.data
project = Project.objects.get_from_cache(id=project)
# Check to make sure we're not about to do a bunch of work that's
# already been done if we've processed an event with this ID. (This
# isn't a perfect solution -- this doesn't handle ``EventMapping`` and
# there's a race condition between here and when the event is actually
# saved, but it's an improvement. See GH-7677.)
try:
event = Event.objects.get(
project_id=project.id,
event_id=data['event_id'],
)
except Event.DoesNotExist:
pass
else:
self.logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': data['event_id'],
'project_id': project.id,
'model': Event.__name__,
}
)
return event
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
level = data.pop('level')
transaction_name = data.pop('transaction', None)
culprit = data.pop('culprit', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
dist = data.pop('dist', None)
environment = data.pop('environment', None)
# unused
time_spent = data.pop('time_spent', None)
message = data.pop('message', '')
if not culprit:
if transaction_name:
culprit = transaction_name
else:
culprit = generate_culprit(data, platform=platform)
culprit = force_text(culprit)
if transaction_name:
transaction_name = force_text(transaction_name)
recorded_timestamp = data.pop('timestamp')
date = datetime.fromtimestamp(recorded_timestamp)
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'platform': platform,
}
event = Event(
project_id=project.id,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs
)
event._project_cache = project
data = event.data.data
# convert this to a dict to ensure we're only storing one value per key
# as most parts of Sentry dont currently play well with multiple values
tags = dict(data.get('tags') or [])
tags['level'] = LOG_LEVELS[level]
if logger_name:
tags['logger'] = logger_name
if server_name:
tags['server_name'] = server_name
if site:
tags['site'] = site
if environment:
tags['environment'] = environment
if transaction_name:
tags['transaction'] = transaction_name
if release:
# dont allow a conflicting 'release' tag
if 'release' in tags:
del tags['release']
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
tags['sentry:release'] = release.version
if dist and release:
dist = release.add_dist(dist, date)
tags['sentry:dist'] = dist.name
else:
dist = None
event_user = self._get_event_user(project, data)
if event_user:
# dont allow a conflicting 'user' tag
if 'user' in tags:
del tags['user']
tags['sentry:user'] = event_user.tag_value
# At this point we want to normalize the in_app values in case the
# clients did not set this appropriately so far.
normalize_in_app(data)
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event, _with_transaction=False)
if added_tags:
# plugins should not override user provided tags
for key, value in added_tags:
tags.setdefault(key, value)
for path, iface in six.iteritems(event.interfaces):
for k, v in iface.iter_tags():
tags[k] = v
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.get_path(), None)
# tags are stored as a tuple
tags = tags.items()
data['tags'] = tags
data['fingerprint'] = fingerprint or ['{{ default }}']
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = [md5_from_hash(h) for h in get_hashes_from_fingerprint(event, fingerprint)]
elif checksum:
if HASH_RE.match(checksum):
hashes = [checksum]
else:
hashes = [md5_from_hash([checksum]), checksum]
data['checksum'] = checksum
else:
hashes = [md5_from_hash(h) for h in get_hashes_for_event(event)]
# TODO(dcramer): temp workaround for complexity
data['message'] = message
event_type = eventtypes.get(data.get('type', 'default'))(data)
event_metadata = event_type.get_metadata()
# TODO(dcramer): temp workaround for complexity
del data['message']
data['type'] = event_type.key
data['metadata'] = event_metadata
# index components into ``Event.message``
# See GH-3248
if event_type.key != 'default':
if 'sentry.interfaces.Message' in data and \
data['sentry.interfaces.Message']['message'] != message:
message = u'{} {}'.format(
message,
data['sentry.interfaces.Message']['message'],
)
if not message:
message = ''
elif not isinstance(message, six.string_types):
message = force_text(message)
for value in six.itervalues(event_metadata):
value_u = force_text(value, errors='replace')
if value_u not in message:
message = u'{} {}'.format(message, value_u)
if culprit and culprit not in message:
culprit_u = force_text(culprit, errors='replace')
message = u'{} {}'.format(message, culprit_u)
message = trim(message.strip(), settings.SENTRY_MAX_MESSAGE_LENGTH)
event.message = message
kwargs['message'] = message
received_timestamp = event.data.get('received') or float(event.datetime.strftime('%s'))
group_kwargs = kwargs.copy()
group_kwargs.update(
{
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'active_at': date,
'data': {
'last_received': received_timestamp,
'type':
event_type.key,
# we cache the events metadata on the group to ensure its
# accessible in the stream
'metadata':
event_metadata,
},
}
)
if release:
group_kwargs['first_release'] = release
try:
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **group_kwargs
)
except HashDiscarded:
event_discarded.send_robust(
project=project,
sender=EventManager,
)
metrics.incr(
'events.discarded',
skip_internal=True,
tags={
'organization_id': project.organization_id,
'platform': platform,
},
)
raise
else:
event_saved.send_robust(
project=project,
event_size=event.size,
sender=EventManager,
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
# When an event was sampled, the canonical source of truth
# is the EventMapping table since we aren't going to be writing out an actual
# Event row. Otherwise, if the Event isn't being sampled, we can safely
# rely on the Event table itself as the source of truth and ignore
# EventMapping since it's redundant information.
if is_sample:
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project, group=group, event_id=event_id)
except IntegrityError:
self.logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': EventMapping.__name__,
}
)
return event
environment = Environment.get_or_create(
project=project,
name=environment,
)
group_environment, is_new_group_environment = GroupEnvironment.get_or_create(
group_id=group.id,
environment_id=environment.id,
defaults={
'first_release_id': release.id if release else None,
},
)
if release:
ReleaseEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
ReleaseProjectEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date,
)
counters = [
(tsdb.models.group, group.id),
(tsdb.models.project, project.id),
]
if release:
counters.append((tsdb.models.release, release.id))
tsdb.incr_multi(counters, timestamp=event.datetime, environment_id=environment.id)
frequencies = [
# (tsdb.models.frequent_projects_by_organization, {
# project.organization_id: {
# project.id: 1,
# },
# }),
# (tsdb.models.frequent_issues_by_project, {
# project.id: {
# group.id: 1,
# },
# })
(tsdb.models.frequent_environments_by_group, {
group.id: {
environment.id: 1,
},
})
]
if release:
frequencies.append(
(tsdb.models.frequent_releases_by_group, {
group.id: {
grouprelease.id: 1,
},
})
)
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(
group=group,
environment=environment,
)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': Event.__name__,
}
)
return event
index_event_tags.delay(
organization_id=project.organization_id,
project_id=project.id,
group_id=group.id,
environment_id=environment.id,
event_id=event.id,
tags=tags,
date_added=event.datetime,
)
if event_user:
tsdb.record_multi(
(
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value, )),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value, )),
),
timestamp=event.datetime,
environment_id=environment.id,
)
if release:
if is_new:
buffer.incr(
ReleaseProject, {'new_groups': 1}, {
'release_id': release.id,
'project_id': project.id,
}
)
if is_new_group_environment:
buffer.incr(
ReleaseProjectEnvironment, {'new_issues_count': 1}, {
'project_id': project.id,
'release_id': release.id,
'environment_id': environment.id,
}
)
safe_execute(Group.objects.add_tags, group, environment, tags, _with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send_robust(project=project, group=group, sender=Project)
eventstream.insert(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
is_new_group_environment=is_new_group_environment,
primary_hash=hashes[0],
# We are choosing to skip consuming the event back
# in the eventstream if it's flagged as raw.
# This means that we want to publish the event
# through the event stream, but we don't care
# about post processing and handling the commit.
skip_consume=raw,
)
metrics.timing(
'events.latency',
received_timestamp - recorded_timestamp,
tags={
'project_id': project.id,
},
)
return event
def setUp(self):
self.project = self.create_project()
self.release = Release.get_or_create(self.project, "1.0")
self.environment1 = Environment.get_or_create(self.project, "prod")
self.environment2 = Environment.get_or_create(self.project, "staging")
self.timestamp = float(int(time() - 300))
def debounce_update_release_health_data(organization, project_ids):
"""This causes a flush of snuba health data to the postgres tables once
per minute for the given projects.
"""
# Figure out which projects need to get updates from the snuba.
should_update = {}
cache_keys = ["debounce-health:%d" % id for id in project_ids]
cache_data = cache.get_many(cache_keys)
for project_id, cache_key in izip(project_ids, cache_keys):
if cache_data.get(cache_key) is None:
should_update[project_id] = cache_key
if not should_update:
return
projects = {
p.id: p
for p in Project.objects.get_many_from_cache(should_update.keys())
}
# This gives us updates for all release-projects which have seen new
# health data over the last days. It will miss releases where the last
# date is longer than what `get_changed_project_release_model_adoptions`
# considers recent.
project_releases = release_health.get_changed_project_release_model_adoptions(
should_update.keys())
# Check which we already have rows for.
existing = set(
ReleaseProject.objects.filter(
project_id__in=[x[0] for x in project_releases],
release__version__in=[x[1] for x in project_releases],
).values_list("project_id", "release__version"))
to_upsert = []
for key in project_releases:
if key not in existing:
to_upsert.append(key)
if to_upsert:
dates = release_health.get_oldest_health_data_for_releases(to_upsert)
for project_id, version in to_upsert:
project = projects.get(project_id)
if project is None:
# should not happen
continue
# We might have never observed the release. This for instance can
# happen if the release only had health data so far. For these cases
# we want to create the release the first time we observed it on the
# health side.
release = Release.get_or_create(project=project,
version=version,
date_added=dates.get(
(project_id, version)))
# Make sure that the release knows about this project. Like we had before
# the project might not have been associated with this release yet.
release.add_project(project)
# Debounce updates for a minute
cache.set_many(
dict(izip(should_update.values(), [True] * len(should_update))), 60)
def save(self, project, raw=False):
from sentry.tasks.post_process import index_event_tags
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
level = data.pop('level')
culprit = data.pop('culprit', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
fingerprint = data.pop('fingerprint', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
environment = data.pop('environment', None)
# unused
time_spent = data.pop('time_spent', None)
message = data.pop('message', '')
if not culprit:
culprit = generate_culprit(data, platform=platform)
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'platform': platform,
}
event = Event(
project_id=project.id,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs
)
tags = data.get('tags') or []
tags.append(('level', LOG_LEVELS[level]))
if logger_name:
tags.append(('logger', logger_name))
if server_name:
tags.append(('server_name', server_name))
if site:
tags.append(('site', site))
if release:
# TODO(dcramer): we should ensure we create Release objects
tags.append(('sentry:release', release))
if environment:
tags.append(('environment', environment))
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event,
_with_transaction=False)
if added_tags:
tags.extend(added_tags)
event_user = self._get_event_user(project, data)
if event_user:
tags.append(('sentry:user', event_user.tag_value))
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
data['fingerprint'] = fingerprint or ['{{ default }}']
for path, iface in event.interfaces.iteritems():
data['tags'].extend(iface.iter_tags())
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.get_path(), None)
# prioritize fingerprint over checksum as its likely the client defaulted
# a checksum whereas the fingerprint was explicit
if fingerprint:
hashes = map(md5_from_hash, get_hashes_from_fingerprint(event, fingerprint))
elif checksum:
hashes = [checksum]
else:
hashes = map(md5_from_hash, get_hashes_for_event(event))
# TODO(dcramer): temp workaround for complexity
data['message'] = message
event_type = eventtypes.get(data.get('type', 'default'))(data)
event_metadata = event_type.get_metadata()
# TODO(dcramer): temp workaround for complexity
del data['message']
data['type'] = event_type.key
data['metadata'] = event_metadata
# index components into ``Event.message``
# See GH-3248
if event_type.key != 'default':
if 'sentry.interfaces.Message' in data and \
data['sentry.interfaces.Message']['message'] != message:
message = u'{} {}'.format(
message,
data['sentry.interfaces.Message']['message'],
)
if not message:
message = ''
elif not isinstance(message, basestring):
message = force_text(message)
for value in event_metadata.itervalues():
value_u = force_text(value, errors='replace')
if value_u not in message:
message = u'{} {}'.format(message, value_u)
message = trim(message.strip(), settings.SENTRY_MAX_MESSAGE_LENGTH)
event.message = message
kwargs['message'] = message
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'data': {
'last_received': event.data.get('received') or float(event.datetime.strftime('%s')),
'type': event_type.key,
# we cache the events metadata on the group to ensure its
# accessible in the stream
'metadata': event_metadata,
},
})
if release:
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
group_kwargs['first_release'] = release
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event,
hashes=hashes,
release=release,
**group_kwargs
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(
project=project, group=group, event_id=event_id)
except IntegrityError:
self.logger.info('Duplicate EventMapping found for event_id=%s', event_id,
exc_info=True)
return event
if release:
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date,
)
tsdb.incr_multi([
(tsdb.models.group, group.id),
(tsdb.models.project, project.id),
], timestamp=event.datetime)
frequencies = [
(tsdb.models.frequent_projects_by_organization, {
project.organization_id: {
project.id: 1,
},
}),
(tsdb.models.frequent_issues_by_project, {
project.id: {
group.id: 1,
},
})
]
if release:
frequencies.append(
(tsdb.models.frequent_releases_by_groups, {
group.id: {
grouprelease.id: 1,
},
})
)
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
UserReport.objects.filter(
project=project, event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
self.logger.info('Duplicate Event found for event_id=%s', event_id,
exc_info=True)
return event
index_event_tags.delay(
project_id=project.id,
group_id=group.id,
event_id=event.id,
tags=tags,
)
if event_user:
tsdb.record_multi((
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value,)),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value,)),
), timestamp=event.datetime)
if is_new and release:
buffer.incr(Release, {'new_groups': 1}, {
'id': release.id,
})
safe_execute(Group.objects.add_tags, group, tags,
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send(project=project, group=group, sender=Project)
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info('Raw event passed; skipping post process for event_id=%s', event_id)
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def setUp(self):
self.project = self.create_project()
self.release = Release.get_or_create(self.project, '1.0')
self.environment1 = Environment.get_or_create(self.project, 'prod')
self.environment2 = Environment.get_or_create(self.project, 'staging')
self.timestamp = 1403007314
def save(self, project, raw=False):
# TODO: culprit should default to "most recent" frame in stacktraces when
# it's not provided.
project = Project.objects.get_from_cache(id=project)
data = self.data.copy()
# First we pull out our top-level (non-data attr) kwargs
event_id = data.pop('event_id')
message = data.pop('message')
level = data.pop('level')
culprit = data.pop('culprit', None) or ''
time_spent = data.pop('time_spent', None)
logger_name = data.pop('logger', None)
server_name = data.pop('server_name', None)
site = data.pop('site', None)
checksum = data.pop('checksum', None)
platform = data.pop('platform', None)
release = data.pop('release', None)
date = datetime.fromtimestamp(data.pop('timestamp'))
date = date.replace(tzinfo=timezone.utc)
kwargs = {
'message': message,
'platform': platform,
}
event = Event(
project=project,
event_id=event_id,
data=data,
time_spent=time_spent,
datetime=date,
**kwargs
)
tags = data.get('tags') or []
tags.append(('level', LOG_LEVELS[level]))
if logger_name:
tags.append(('logger', logger_name))
if server_name:
tags.append(('server_name', server_name))
if site:
tags.append(('site', site))
if release:
# TODO(dcramer): we should ensure we create Release objects
tags.append(('sentry:release', release))
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event,
_with_transaction=False)
if added_tags:
tags.extend(added_tags)
# XXX(dcramer): we're relying on mutation of the data object to ensure
# this propagates into Event
data['tags'] = tags
# Calculate the checksum from the first highest scoring interface
if checksum:
hashes = [checksum]
else:
hashes = get_hashes_for_event(event)
group_kwargs = kwargs.copy()
group_kwargs.update({
'culprit': culprit,
'logger': logger_name,
'level': level,
'last_seen': date,
'first_seen': date,
'time_spent_total': time_spent or 0,
'time_spent_count': time_spent and 1 or 0,
})
if release:
group_kwargs['first_release'] = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
Activity.objects.create(
type=Activity.RELEASE,
project=project,
ident=release,
data={'version': release},
datetime=date,
)
group, is_new, is_regression, is_sample = safe_execute(
self._save_aggregate,
event=event,
hashes=hashes,
**group_kwargs
)
using = group._state.db
event.group = group
try:
with transaction.atomic():
EventMapping.objects.create(
project=project, group=group, event_id=event_id)
except IntegrityError:
self.logger.info('Duplicate EventMapping found for event_id=%s', event_id)
return event
UserReport.objects.filter(
project=project, event_id=event_id,
).update(group=group)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic():
event.save()
except IntegrityError:
self.logger.info('Duplicate Event found for event_id=%s', event_id)
return event
safe_execute(Group.objects.add_tags, group, tags,
_with_transaction=False)
if not raw:
post_process_group.delay(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
)
else:
self.logger.info('Raw event passed; skipping post process for event_id=%s', event_id)
index_event.delay(event)
# TODO: move this to the queue
if is_regression and not raw:
regression_signal.send_robust(sender=Group, instance=group)
return event
def create_message_event(template, parameters, environment, release):
i = next(sequence)
event_id = uuid.UUID(
fields=(i, 0x0, 0x1000, 0x80, 0x80, 0x808080808080, ),
).hex
tags = [['color', next(tag_values)]]
if environment:
tags.append(['environment', environment])
if release:
tags.append(['sentry:release', release])
event = Event.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
message='%s' % (id, ),
datetime=now + shift(i),
data={
'environment': environment,
'type': 'default',
'metadata': {
'title': template % parameters,
},
'logentry': {
'message': template,
'params': parameters,
'formatted': template % parameters,
},
'user': next(user_values),
'tags': tags,
},
)
with self.tasks():
Group.objects.add_tags(
source,
Environment.objects.get(
organization_id=project.organization_id,
name=environment
),
tags=event.get_tags(),
)
EventMapping.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
date_added=event.datetime,
)
UserReport.objects.create(
project_id=project.id,
group_id=source.id,
event_id=event_id,
name='Log Hat',
email='*****@*****.**',
comments='Quack',
)
if release:
Release.get_or_create(
project=project,
version=event.get_tag('sentry:release'),
date_added=event.datetime,
)
features.record([event])
return event
def save(self, project_id, raw=False, assume_normalized=False):
# Normalize if needed
if not self._normalized:
if not assume_normalized:
self.normalize()
self._normalized = True
data = self._data
project = Project.objects.get_from_cache(id=project_id)
project._organization_cache = Organization.objects.get_from_cache(
id=project.organization_id)
# Check to make sure we're not about to do a bunch of work that's
# already been done if we've processed an event with this ID. (This
# isn't a perfect solution -- this doesn't handle ``EventMapping`` and
# there's a race condition between here and when the event is actually
# saved, but it's an improvement. See GH-7677.)
try:
event = Event.objects.get(
project_id=project.id,
event_id=data['event_id'],
)
except Event.DoesNotExist:
pass
else:
# Make sure we cache on the project before returning
event._project_cache = project
logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': data['event_id'],
'project_id': project.id,
'model': Event.__name__,
}
)
return event
# Pull out the culprit
culprit = self.get_culprit()
# Pull the toplevel data we're interested in
level = data.get('level')
# TODO(mitsuhiko): this code path should be gone by July 2018.
# This is going to be fine because no code actually still depends
# on integers here. When we need an integer it will be converted
# into one later. Old workers used to send integers here.
if level is not None and isinstance(level, six.integer_types):
level = LOG_LEVELS[level]
transaction_name = data.get('transaction')
logger_name = data.get('logger')
release = data.get('release')
dist = data.get('dist')
environment = data.get('environment')
recorded_timestamp = data.get('timestamp')
# We need to swap out the data with the one internal to the newly
# created event object
event = self._get_event_instance(project_id=project_id)
self._data = data = event.data.data
event._project_cache = project
date = event.datetime
platform = event.platform
event_id = event.event_id
if transaction_name:
transaction_name = force_text(transaction_name)
# Some of the data that are toplevel attributes are duplicated
# into tags (logger, level, environment, transaction). These are
# different from legacy attributes which are normalized into tags
# ahead of time (site, server_name).
setdefault_path(data, 'tags', value=[])
set_tag(data, 'level', level)
if logger_name:
set_tag(data, 'logger', logger_name)
if environment:
set_tag(data, 'environment', environment)
if transaction_name:
set_tag(data, 'transaction', transaction_name)
if release:
# dont allow a conflicting 'release' tag
pop_tag(data, 'release')
release = Release.get_or_create(
project=project,
version=release,
date_added=date,
)
set_tag(data, 'sentry:release', release.version)
if dist and release:
dist = release.add_dist(dist, date)
# dont allow a conflicting 'dist' tag
pop_tag(data, 'dist')
set_tag(data, 'sentry:dist', dist.name)
else:
dist = None
event_user = self._get_event_user(project, data)
if event_user:
# dont allow a conflicting 'user' tag
pop_tag(data, 'user')
set_tag(data, 'sentry:user', event_user.tag_value)
# At this point we want to normalize the in_app values in case the
# clients did not set this appropriately so far.
grouping_config = load_grouping_config(
get_grouping_config_dict_for_event_data(data, project))
normalize_stacktraces_for_grouping(data, grouping_config)
for plugin in plugins.for_project(project, version=None):
added_tags = safe_execute(plugin.get_tags, event, _with_transaction=False)
if added_tags:
# plugins should not override user provided tags
for key, value in added_tags:
if get_tag(data, key) is None:
set_tag(data, key, value)
for path, iface in six.iteritems(event.interfaces):
for k, v in iface.iter_tags():
set_tag(data, k, v)
# Get rid of ephemeral interface data
if iface.ephemeral:
data.pop(iface.path, None)
# The active grouping config was put into the event in the
# normalize step before. We now also make sure that the
# fingerprint was set to `'{{ default }}' just in case someone
# removed it from the payload. The call to get_hashes will then
# look at `grouping_config` to pick the right paramters.
data['fingerprint'] = data.get('fingerprint') or ['{{ default }}']
apply_server_fingerprinting(data, get_fingerprinting_config_for_project(project))
hashes = event.get_hashes()
data['hashes'] = hashes
# we want to freeze not just the metadata and type in but also the
# derived attributes. The reason for this is that we push this
# data into kafka for snuba processing and our postprocessing
# picks up the data right from the snuba topic. For most usage
# however the data is dynamically overriden by Event.title and
# Event.location (See Event.as_dict)
materialized_metadata = self.materialize_metadata()
event_metadata = materialized_metadata['metadata']
data.update(materialized_metadata)
data['culprit'] = culprit
# index components into ``Event.message``
# See GH-3248
event.message = self.get_search_message(event_metadata, culprit)
received_timestamp = event.data.get('received') or float(event.datetime.strftime('%s'))
# The group gets the same metadata as the event when it's flushed but
# additionally the `last_received` key is set. This key is used by
# _save_aggregate.
group_metadata = dict(materialized_metadata)
group_metadata['last_received'] = received_timestamp
kwargs = {
'platform': platform,
'message': event.message,
'culprit': culprit,
'logger': logger_name,
'level': LOG_LEVELS_MAP.get(level),
'last_seen': date,
'first_seen': date,
'active_at': date,
'data': group_metadata,
}
if release:
kwargs['first_release'] = release
try:
group, is_new, is_regression, is_sample = self._save_aggregate(
event=event, hashes=hashes, release=release, **kwargs
)
except HashDiscarded:
event_discarded.send_robust(
project=project,
sender=EventManager,
)
metrics.incr(
'events.discarded',
skip_internal=True,
tags={
'organization_id': project.organization_id,
'platform': platform,
},
)
raise
else:
event_saved.send_robust(
project=project,
event_size=event.size,
sender=EventManager,
)
event.group = group
# store a reference to the group id to guarantee validation of isolation
event.data.bind_ref(event)
# When an event was sampled, the canonical source of truth
# is the EventMapping table since we aren't going to be writing out an actual
# Event row. Otherwise, if the Event isn't being sampled, we can safely
# rely on the Event table itself as the source of truth and ignore
# EventMapping since it's redundant information.
if is_sample:
try:
with transaction.atomic(using=router.db_for_write(EventMapping)):
EventMapping.objects.create(project=project, group=group, event_id=event_id)
except IntegrityError:
logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': EventMapping.__name__,
}
)
return event
environment = Environment.get_or_create(
project=project,
name=environment,
)
group_environment, is_new_group_environment = GroupEnvironment.get_or_create(
group_id=group.id,
environment_id=environment.id,
defaults={
'first_release': release if release else None,
},
)
if release:
ReleaseEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
ReleaseProjectEnvironment.get_or_create(
project=project,
release=release,
environment=environment,
datetime=date,
)
grouprelease = GroupRelease.get_or_create(
group=group,
release=release,
environment=environment,
datetime=date,
)
counters = [
(tsdb.models.group, group.id),
(tsdb.models.project, project.id),
]
if release:
counters.append((tsdb.models.release, release.id))
tsdb.incr_multi(counters, timestamp=event.datetime, environment_id=environment.id)
frequencies = [
# (tsdb.models.frequent_projects_by_organization, {
# project.organization_id: {
# project.id: 1,
# },
# }),
# (tsdb.models.frequent_issues_by_project, {
# project.id: {
# group.id: 1,
# },
# })
(tsdb.models.frequent_environments_by_group, {
group.id: {
environment.id: 1,
},
})
]
if release:
frequencies.append(
(tsdb.models.frequent_releases_by_group, {
group.id: {
grouprelease.id: 1,
},
})
)
tsdb.record_frequency_multi(frequencies, timestamp=event.datetime)
UserReport.objects.filter(
project=project,
event_id=event_id,
).update(
group=group,
environment=environment,
)
# save the event unless its been sampled
if not is_sample:
try:
with transaction.atomic(using=router.db_for_write(Event)):
event.save()
except IntegrityError:
logger.info(
'duplicate.found',
exc_info=True,
extra={
'event_uuid': event_id,
'project_id': project.id,
'group_id': group.id,
'model': Event.__name__,
}
)
return event
tagstore.delay_index_event_tags(
organization_id=project.organization_id,
project_id=project.id,
group_id=group.id,
environment_id=environment.id,
event_id=event.id,
tags=event.tags,
date_added=event.datetime,
)
if event_user:
tsdb.record_multi(
(
(tsdb.models.users_affected_by_group, group.id, (event_user.tag_value, )),
(tsdb.models.users_affected_by_project, project.id, (event_user.tag_value, )),
),
timestamp=event.datetime,
environment_id=environment.id,
)
if release:
if is_new:
buffer.incr(
ReleaseProject, {'new_groups': 1}, {
'release_id': release.id,
'project_id': project.id,
}
)
if is_new_group_environment:
buffer.incr(
ReleaseProjectEnvironment, {'new_issues_count': 1}, {
'project_id': project.id,
'release_id': release.id,
'environment_id': environment.id,
}
)
safe_execute(
Group.objects.add_tags,
group,
environment,
event.get_tags(),
_with_transaction=False)
if not raw:
if not project.first_event:
project.update(first_event=date)
first_event_received.send_robust(project=project, group=group, sender=Project)
eventstream.insert(
group=group,
event=event,
is_new=is_new,
is_sample=is_sample,
is_regression=is_regression,
is_new_group_environment=is_new_group_environment,
primary_hash=hashes[0],
# We are choosing to skip consuming the event back
# in the eventstream if it's flagged as raw.
# This means that we want to publish the event
# through the event stream, but we don't care
# about post processing and handling the commit.
skip_consume=raw,
)
metrics.timing(
'events.latency',
received_timestamp - recorded_timestamp,
tags={
'project_id': project.id,
},
)
metrics.timing(
'events.size.data.post_save',
event.size,
tags={'project_id': project.id}
)
return event
