[Notes] [Git][BuildStream/buildstream][valentindavid/cache_server_fill_up] 11 commits: setup.py: require ruamel.yaml >= 0.15.41 < 0.15.52

[Valentin David <gitlab mg gitlab com>]

Title: GitLab

Valentin David pushed to branch valentindavid/cache_server_fill_up at BuildStream / buildstream

Commits:

• 499c70fd
  by Valentin David at 2018-11-28T09:11:21Z

  setup.py: require ruamel.yaml >= 0.15.41 < 0.15.52
  

• 3513580c
  by Valentin David at 2018-11-28T14:02:32Z

  Merge branch 'valentindavid/ruamel_0_15_41' into 'master'
  
  setup.py: require ruamel.yaml >= 0.15.41 < 0.15.52
  
  See merge request BuildStream/buildstream!975

• 26cdee08
  by Valentin David at 2018-11-28T14:29:52Z

  Use f_bavail to query available space. Not f_bfree.
  
  f_bfree space might not be usable. In practice we see failures in big
  disks because f_bfree is over 2GB and f_bavail is 0. We get ENOSPC if
  writing on disk then.
  

• 58ca298f
  by Valentin David at 2018-11-28T14:29:52Z

  Make cache clients not fail when a blob is not available.
  
  We plan to make cache incomplete. That is some blobs are missing.  For
  most of cases we will delete references when requested if they are
  incomplete. But there will be corner cases where objects are removed
  after the reference is requested.
  

• 227fa26d
  by Valentin David at 2018-11-28T14:29:52Z

  Fix type error in RPC messages
  

• 5ef19a0b
  by Valentin David at 2018-11-28T14:29:52Z

  Avoid copying temporary file when adding object to CAS in server.
  
  The file is already a temporary file and does not need copy.  ENOSPC
  is thrown during that copy in issue #609.
  
  Fixes #678.
  

• 8d2946ff
  by Valentin David at 2018-11-28T14:29:52Z

  "Fallocate" object disk space to avoid getting NOSPC error when writing
  
  This locks the temporary object file so that cleanup does not need to
  be done for every write.
  

• b587953a
  by Valentin David at 2018-11-28T14:29:52Z

  Update mtimes of objects for requested references.
  
  This also remove references when some objects are missing. This is in
  preparation for the move from reference to object garbage collection.
  

• a64f667d
  by Valentin David at 2018-11-28T14:29:52Z

  Move cas server from ref-based to object-based garbage collection.
  

• 353b90dd
  by Valentin David at 2018-11-28T14:29:52Z

  Cleanup cache in cas server more agressively
  
  When there is less than 2GB left, it cleans up have 10GB available.
  These values are configurable.
  

• ba9afa98
  by Valentin David at 2018-11-28T14:29:52Z

  Lock cache cleanup in cas server
  
  Cleaning up in parallel might slow down the cleaning process
  

5 changed files:

• buildstream/_artifactcache/cascache.py
• buildstream/_artifactcache/casserver.py
• setup.py
• tests/frontend/push.py
• tests/testutils/artifactshare.py

Changes:

buildstream/_artifactcache/cascache.py

---

...	...	@@ -25,6 +25,7 @@ import os
25	25	import stat
26	26	import tempfile
27	27	import uuid
	28	+import contextlib
28	29	from urllib.parse import urlparse
29	30
30	31	import grpc
...	...	@@ -88,6 +89,13 @@ class CASRemoteSpec(namedtuple('CASRemoteSpec', 'url push server_cert client_key
88	89	CASRemoteSpec.__new__.__defaults__ = (None, None, None)
89	90
90	91
	92	+class BlobNotFound(CASError):
	93	+
	94	+ def __init__(self, blob, msg):
	95	+ self.blob = blob
	96	+ super().__init__(msg)
	97	+
	98	+
91	99	# A CASCache manages a CAS repository as specified in the Remote Execution API.
92	100	#
93	101	# Args:
...	...	@@ -299,6 +307,8 @@ class CASCache():
299	307	raise CASError("Failed to pull ref {}: {}".format(ref, e)) from e
300	308	else:
301	309	return False
	310	+ except BlobNotFound as e:
	311	+ return False
302	312
303	313	# pull_tree():
304	314	#
...	...	@@ -471,13 +481,14 @@ class CASCache():
471	481	# digest (Digest): An optional Digest object to populate
472	482	# path (str): Path to file to add
473	483	# buffer (bytes): Byte buffer to add
	484	+ # link_directly (bool): Whether file given by path can be linked
474	485	#
475	486	# Returns:
476	487	# (Digest): The digest of the added object
477	488	#
478	489	# Either `path` or `buffer` must be passed, but not both.
479	490	#
480		- def add_object(self, *, digest=None, path=None, buffer=None):
	491	+ def add_object(self, *, digest=None, path=None, buffer=None, link_directly=False):
481	492	# Exactly one of the two parameters has to be specified
482	493	assert (path is None) != (buffer is None)
483	494
...	...	@@ -487,28 +498,34 @@ class CASCache():
487	498	try:
488	499	h = hashlib.sha256()
489	500	# Always write out new file to avoid corruption if input file is modified
490		- with tempfile.NamedTemporaryFile(dir=self.tmpdir) as out:
491		- # Set mode bits to 0644
492		- os.chmod(out.name, stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP | stat.S_IROTH)
493		-
494		- if path:
495		- with open(path, 'rb') as f:
496		- for chunk in iter(lambda: f.read(4096), b""):
497		- h.update(chunk)
498		- out.write(chunk)
	501	+ with contextlib.ExitStack() as stack:
	502	+ if path is not None and link_directly:
	503	+ tmp = stack.enter_context(open(path, 'rb'))
	504	+ for chunk in iter(lambda: tmp.read(4096), b""):
	505	+ h.update(chunk)
499	506	else:
500		- h.update(buffer)
501		- out.write(buffer)
	507	+ tmp = stack.enter_context(tempfile.NamedTemporaryFile(dir=self.tmpdir))
	508	+ # Set mode bits to 0644
	509	+ os.chmod(tmp.name, stat.S_IRUSR | stat.S_IWUSR | stat.S_IRGRP | stat.S_IROTH)
502	510
503		- out.flush()
	511	+ if path:
	512	+ with open(path, 'rb') as f:
	513	+ for chunk in iter(lambda: f.read(4096), b""):
	514	+ h.update(chunk)
	515	+ tmp.write(chunk)
	516	+ else:
	517	+ h.update(buffer)
	518	+ tmp.write(buffer)
	519	+
	520	+ tmp.flush()
504	521
505	522	digest.hash = h.hexdigest()
506		- digest.size_bytes = os.fstat(out.fileno()).st_size
	523	+ digest.size_bytes = os.fstat(tmp.fileno()).st_size
507	524
508	525	# Place file at final location
509	526	objpath = self.objpath(digest)
510	527	os.makedirs(os.path.dirname(objpath), exist_ok=True)
511		- os.link(out.name, objpath)
	528	+ os.link(tmp.name, objpath)
512	529
513	530	except FileExistsError as e:
514	531	# We can ignore the failed link() if the object is already in the repo.
...	...	@@ -606,6 +623,41 @@ class CASCache():
606	623	# first ref of this list will be the file modified earliest.
607	624	return [ref for _, ref in sorted(zip(mtimes, refs))]
608	625
	626	+ # list_objects():
	627	+ #
	628	+ # List cached objects in Least Recently Modified (LRM) order.
	629	+ #
	630	+ # Returns:
	631	+ # (list) - A list of objects and timestamps in LRM order
	632	+ #
	633	+ def list_objects(self):
	634	+ objs = []
	635	+ mtimes = []
	636	+
	637	+ for root, _, files in os.walk(os.path.join(self.casdir, 'objects')):
	638	+ for filename in files:
	639	+ obj_path = os.path.join(root, filename)
	640	+ try:
	641	+ mtimes.append(os.path.getmtime(obj_path))
	642	+ except FileNotFoundError:
	643	+ pass
	644	+ else:
	645	+ objs.append(obj_path)
	646	+
	647	+ # NOTE: Sorted will sort from earliest to latest, thus the
	648	+ # first element of this list will be the file modified earliest.
	649	+ return sorted(zip(mtimes, objs))
	650	+
	651	+ def clean_up_refs_until(self, time):
	652	+ ref_heads = os.path.join(self.casdir, 'refs', 'heads')
	653	+
	654	+ for root, _, files in os.walk(ref_heads):
	655	+ for filename in files:
	656	+ ref_path = os.path.join(root, filename)
	657	+ # Obtain the mtime (the time a file was last modified)
	658	+ if os.path.getmtime(ref_path) < time:
	659	+ os.unlink(ref_path)
	660	+
609	661	# remove():
610	662	#
611	663	# Removes the given symbolic ref from the repo.
...	...	@@ -665,6 +717,10 @@ class CASCache():
665	717
666	718	return pruned
667	719
	720	+ def update_tree_mtime(self, tree):
	721	+ reachable = set()
	722	+ self._reachable_refs_dir(reachable, tree, update_mtime=True)
	723	+
668	724	################################################
669	725	# Local Private Methods #
670	726	################################################
...	...	@@ -811,10 +867,13 @@ class CASCache():
811	867	a += 1
812	868	b += 1
813	869
814		- def _reachable_refs_dir(self, reachable, tree):
	870	+ def _reachable_refs_dir(self, reachable, tree, update_mtime=False):
815	871	if tree.hash in reachable:
816	872	return
817	873
	874	+ if update_mtime:
	875	+ os.utime(self.objpath(tree))
	876	+
818	877	reachable.add(tree.hash)
819	878
820	879	directory = remote_execution_pb2.Directory()
...	...	@@ -823,10 +882,12 @@ class CASCache():
823	882	directory.ParseFromString(f.read())
824	883
825	884	for filenode in directory.files:
	885	+ if update_mtime:
	886	+ os.utime(self.objpath(filenode.digest))
826	887	reachable.add(filenode.digest.hash)
827	888
828	889	for dirnode in directory.directories:
829		- self._reachable_refs_dir(reachable, dirnode.digest)
	890	+ self._reachable_refs_dir(reachable, dirnode.digest, update_mtime=update_mtime)
830	891
831	892	def _required_blobs(self, directory_digest):
832	893	# parse directory, and recursively add blobs
...	...	@@ -880,7 +941,7 @@ class CASCache():
880	941	with tempfile.NamedTemporaryFile(dir=self.tmpdir) as f:
881	942	self._fetch_blob(remote, digest, f)
882	943
883		- added_digest = self.add_object(path=f.name)
	944	+ added_digest = self.add_object(path=f.name, link_directly=True)
884	945	assert added_digest.hash == digest.hash
885	946
886	947	return objpath
...	...	@@ -891,7 +952,7 @@ class CASCache():
891	952	f.write(data)
892	953	f.flush()
893	954
894		- added_digest = self.add_object(path=f.name)
	955	+ added_digest = self.add_object(path=f.name, link_directly=True)
895	956	assert added_digest.hash == digest.hash
896	957
897	958	# Helper function for _fetch_directory().
...	...	@@ -1203,6 +1264,9 @@ class _CASBatchRead():
1203	1264	batch_response = self._remote.cas.BatchReadBlobs(self._request)
1204	1265
1205	1266	for response in batch_response.responses:
	1267	+ if response.status.code == code_pb2.NOT_FOUND:
	1268	+ raise BlobNotFound(response.digest.hash, "Failed to download blob {}: {}".format(
	1269	+ response.digest.hash, response.status.code))
1206	1270	if response.status.code != code_pb2.OK:
1207	1271	raise CASError("Failed to download blob {}: {}".format(
1208	1272	response.digest.hash, response.status.code))

buildstream/_artifactcache/casserver.py

---

...	...	@@ -24,6 +24,8 @@ import signal
24	24	import sys
25	25	import tempfile
26	26	import uuid
	27	+import errno
	28	+import threading
27	29
28	30	import click
29	31	import grpc
...	...	@@ -31,6 +33,7 @@ import grpc
31	33	from .._protos.build.bazel.remote.execution.v2 import remote_execution_pb2, remote_execution_pb2_grpc
32	34	from .._protos.google.bytestream import bytestream_pb2, bytestream_pb2_grpc
33	35	from .._protos.buildstream.v2 import buildstream_pb2, buildstream_pb2_grpc
	36	+from .._protos.google.rpc import code_pb2
34	37
35	38	from .._exceptions import CASError
36	39
...	...	@@ -55,18 +58,22 @@ class ArtifactTooLargeException(Exception):
55	58	# repo (str): Path to CAS repository
56	59	# enable_push (bool): Whether to allow blob uploads and artifact updates
57	60	#
58		-def create_server(repo, *, enable_push):
	61	+def create_server(repo, *, enable_push,
	62	+ max_head_size=int(10e9),
	63	+ min_head_size=int(2e9)):
59	64	cas = CASCache(os.path.abspath(repo))
60	65
61	66	# Use max_workers default from Python 3.5+
62	67	max_workers = (os.cpu_count() or 1) * 5
63	68	server = grpc.server(futures.ThreadPoolExecutor(max_workers))
64	69
	70	+ cache_cleaner = _CacheCleaner(cas, max_head_size, min_head_size)
	71	+
65	72	bytestream_pb2_grpc.add_ByteStreamServicer_to_server(
66		- _ByteStreamServicer(cas, enable_push=enable_push), server)
	73	+ _ByteStreamServicer(cas, cache_cleaner, enable_push=enable_push), server)
67	74
68	75	remote_execution_pb2_grpc.add_ContentAddressableStorageServicer_to_server(
69		- _ContentAddressableStorageServicer(cas, enable_push=enable_push), server)
	76	+ _ContentAddressableStorageServicer(cas, cache_cleaner, enable_push=enable_push), server)
70	77
71	78	remote_execution_pb2_grpc.add_CapabilitiesServicer_to_server(
72	79	_CapabilitiesServicer(), server)
...	...	@@ -84,9 +91,19 @@ def create_server(repo, *, enable_push):
84	91	@click.option('--client-certs', help="Public client certificates for TLS (PEM-encoded)")
85	92	@click.option('--enable-push', default=False, is_flag=True,
86	93	help="Allow clients to upload blobs and update artifact cache")
	94	+@click.option('--head-room-min', type=click.INT,
	95	+ help="Disk head room minimum in bytes",
	96	+ default=2e9)
	97	+@click.option('--head-room-max', type=click.INT,
	98	+ help="Disk head room maximum in bytes",
	99	+ default=10e9)
87	100	@click.argument('repo')
88		-def server_main(repo, port, server_key, server_cert, client_certs, enable_push):
89		- server = create_server(repo, enable_push=enable_push)
	101	+def server_main(repo, port, server_key, server_cert, client_certs, enable_push,
	102	+ head_room_min, head_room_max):
	103	+ server = create_server(repo,
	104	+ max_head_size=head_room_max,
	105	+ min_head_size=head_room_min,
	106	+ enable_push=enable_push)
90	107
91	108	use_tls = bool(server_key)
92	109
...	...	@@ -128,10 +145,11 @@ def server_main(repo, port, server_key, server_cert, client_certs, enable_push):
128	145
129	146
130	147	class _ByteStreamServicer(bytestream_pb2_grpc.ByteStreamServicer):
131		- def __init__(self, cas, *, enable_push):
	148	+ def __init__(self, cas, cache_cleaner, *, enable_push):
132	149	super().__init__()
133	150	self.cas = cas
134	151	self.enable_push = enable_push
	152	+ self.cache_cleaner = cache_cleaner
135	153
136	154	def Read(self, request, context):
137	155	resource_name = request.resource_name
...	...	@@ -189,17 +207,34 @@ class _ByteStreamServicer(bytestream_pb2_grpc.ByteStreamServicer):
189	207	context.set_code(grpc.StatusCode.NOT_FOUND)
190	208	return response
191	209
192		- try:
193		- _clean_up_cache(self.cas, client_digest.size_bytes)
194		- except ArtifactTooLargeException as e:
195		- context.set_code(grpc.StatusCode.RESOURCE_EXHAUSTED)
196		- context.set_details(str(e))
197		- return response
	210	+ while True:
	211	+ if client_digest.size_bytes == 0:
	212	+ break
	213	+ try:
	214	+ self.cache_cleaner.clean_up(client_digest.size_bytes)
	215	+ except ArtifactTooLargeException as e:
	216	+ context.set_code(grpc.StatusCode.RESOURCE_EXHAUSTED)
	217	+ context.set_details(str(e))
	218	+ return response
	219	+
	220	+ try:
	221	+ os.posix_fallocate(out.fileno(), 0, client_digest.size_bytes)
	222	+ break
	223	+ except OSError as e:
	224	+ # Multiple upload can happen in the same time
	225	+ if e.errno != errno.ENOSPC:
	226	+ raise
	227	+
198	228	elif request.resource_name:
199	229	# If it is set on subsequent calls, it **must** match the value of the first request.
200	230	if request.resource_name != resource_name:
201	231	context.set_code(grpc.StatusCode.FAILED_PRECONDITION)
202	232	return response
	233	+
	234	+ if (offset + len(request.data)) > client_digest.size_bytes:
	235	+ context.set_code(grpc.StatusCode.FAILED_PRECONDITION)
	236	+ return response
	237	+
203	238	out.write(request.data)
204	239	offset += len(request.data)
205	240	if request.finish_write:
...	...	@@ -207,7 +242,7 @@ class _ByteStreamServicer(bytestream_pb2_grpc.ByteStreamServicer):
207	242	context.set_code(grpc.StatusCode.FAILED_PRECONDITION)
208	243	return response
209	244	out.flush()
210		- digest = self.cas.add_object(path=out.name)
	245	+ digest = self.cas.add_object(path=out.name, link_directly=True)
211	246	if digest.hash != client_digest.hash:
212	247	context.set_code(grpc.StatusCode.FAILED_PRECONDITION)
213	248	return response
...	...	@@ -220,18 +255,26 @@ class _ByteStreamServicer(bytestream_pb2_grpc.ByteStreamServicer):
220	255
221	256
222	257	class _ContentAddressableStorageServicer(remote_execution_pb2_grpc.ContentAddressableStorageServicer):
223		- def __init__(self, cas, *, enable_push):
	258	+ def __init__(self, cas, cache_cleaner, *, enable_push):
224	259	super().__init__()
225	260	self.cas = cas
226	261	self.enable_push = enable_push
	262	+ self.cache_cleaner = cache_cleaner
227	263
228	264	def FindMissingBlobs(self, request, context):
229	265	response = remote_execution_pb2.FindMissingBlobsResponse()
230	266	for digest in request.blob_digests:
231		- if not _has_object(self.cas, digest):
232		- d = response.missing_blob_digests.add()
233		- d.hash = digest.hash
234		- d.size_bytes = digest.size_bytes
	267	+ objpath = self.cas.objpath(digest)
	268	+ try:
	269	+ os.utime(objpath)
	270	+ except OSError as e:
	271	+ if e.errno != errno.ENOENT:
	272	+ raise
	273	+ else:
	274	+ d = response.missing_blob_digests.add()
	275	+ d.hash = digest.hash
	276	+ d.size_bytes = digest.size_bytes
	277	+
235	278	return response
236	279
237	280	def BatchReadBlobs(self, request, context):
...	...	@@ -250,12 +293,12 @@ class _ContentAddressableStorageServicer(remote_execution_pb2_grpc.ContentAddres
250	293	try:
251	294	with open(self.cas.objpath(digest), 'rb') as f:
252	295	if os.fstat(f.fileno()).st_size != digest.size_bytes:
253		- blob_response.status.code = grpc.StatusCode.NOT_FOUND
	296	+ blob_response.status.code = code_pb2.NOT_FOUND
254	297	continue
255	298
256	299	blob_response.data = f.read(digest.size_bytes)
257	300	except FileNotFoundError:
258		- blob_response.status.code = grpc.StatusCode.NOT_FOUND
	301	+ blob_response.status.code = code_pb2.NOT_FOUND
259	302
260	303	return response
261	304
...	...	@@ -285,7 +328,7 @@ class _ContentAddressableStorageServicer(remote_execution_pb2_grpc.ContentAddres
285	328	continue
286	329
287	330	try:
288		- _clean_up_cache(self.cas, digest.size_bytes)
	331	+ self.cache_cleaner.clean_up(digest.size_bytes)
289	332
290	333	with tempfile.NamedTemporaryFile(dir=self.cas.tmpdir) as out:
291	334	out.write(blob_request.data)
...	...	@@ -328,6 +371,12 @@ class _ReferenceStorageServicer(buildstream_pb2_grpc.ReferenceStorageServicer):
328	371
329	372	try:
330	373	tree = self.cas.resolve_ref(request.key, update_mtime=True)
	374	+ try:
	375	+ self.cas.update_tree_mtime(tree)
	376	+ except FileNotFoundError:
	377	+ self.cas.remove(request.key, defer_prune=True)
	378	+ context.set_code(grpc.StatusCode.NOT_FOUND)
	379	+ return response
331	380
332	381	response.digest.hash = tree.hash
333	382	response.digest.size_bytes = tree.size_bytes
...	...	@@ -400,60 +449,79 @@ def _digest_from_upload_resource_name(resource_name):
400	449	return None
401	450
402	451
403		-def _has_object(cas, digest):
404		- objpath = cas.objpath(digest)
405		- return os.path.exists(objpath)
	452	+class _CacheCleaner:
406	453
	454	+ __cleanup_cache_lock = threading.Lock()
407	455
408		-# _clean_up_cache()
409		-#
410		-# Keep removing Least Recently Pushed (LRP) artifacts in a cache until there
411		-# is enough space for the incoming artifact
412		-#
413		-# Args:
414		-# cas: CASCache object
415		-# object_size: The size of the object being received in bytes
416		-#
417		-# Returns:
418		-# int: The total bytes removed on the filesystem
419		-#
420		-def _clean_up_cache(cas, object_size):
421		- # Determine the available disk space, in bytes, of the file system
422		- # which mounts the repo
423		- stats = os.statvfs(cas.casdir)
424		- buffer_ = int(2e9) # Add a 2 GB buffer
425		- free_disk_space = (stats.f_bfree * stats.f_bsize) - buffer_
426		- total_disk_space = (stats.f_blocks * stats.f_bsize) - buffer_
427		-
428		- if object_size > total_disk_space:
429		- raise ArtifactTooLargeException("Artifact of size: {} is too large for "
430		- "the filesystem which mounts the remote "
431		- "cache".format(object_size))
432		-
433		- if object_size <= free_disk_space:
434		- # No need to clean up
435		- return 0
436		-
437		- # obtain a list of LRP artifacts
438		- LRP_artifacts = cas.list_refs()
439		-
440		- removed_size = 0 # in bytes
441		- while object_size - removed_size > free_disk_space:
442		- try:
443		- to_remove = LRP_artifacts.pop(0) # The first element in the list is the LRP artifact
444		- except IndexError:
445		- # This exception is caught if there are no more artifacts in the list
446		- # LRP_artifacts. This means the the artifact is too large for the filesystem
447		- # so we abort the process
448		- raise ArtifactTooLargeException("Artifact of size {} is too large for "
449		- "the filesystem which mounts the remote "
450		- "cache".format(object_size))
	456	+ def __init__(self, cas, max_head_size, min_head_size=int(2e9)):
	457	+ self.__cas = cas
	458	+ self.__max_head_size = max_head_size
	459	+ self.__min_head_size = min_head_size
451	460
452		- removed_size += cas.remove(to_remove, defer_prune=False)
	461	+ def __has_space(self, object_size):
	462	+ stats = os.statvfs(self.__cas.casdir)
	463	+ free_disk_space = (stats.f_bavail * stats.f_bsize) - self.__min_head_size
	464	+ total_disk_space = (stats.f_blocks * stats.f_bsize) - self.__min_head_size
453	465
454		- if removed_size > 0:
455		- logging.info("Successfully removed {} bytes from the cache".format(removed_size))
456		- else:
457		- logging.info("No artifacts were removed from the cache.")
	466	+ if object_size > total_disk_space:
	467	+ raise ArtifactTooLargeException("Artifact of size: {} is too large for "
	468	+ "the filesystem which mounts the remote "
	469	+ "cache".format(object_size))
458	470
459		- return removed_size
	471	+ return object_size <= free_disk_space
	472	+
	473	+ # _clean_up_cache()
	474	+ #
	475	+ # Keep removing Least Recently Pushed (LRP) artifacts in a cache until there
	476	+ # is enough space for the incoming artifact
	477	+ #
	478	+ # Args:
	479	+ # object_size: The size of the object being received in bytes
	480	+ #
	481	+ # Returns:
	482	+ # int: The total bytes removed on the filesystem
	483	+ #
	484	+ def clean_up(self, object_size):
	485	+ if self.__has_space(object_size):
	486	+ return 0
	487	+
	488	+ with _CacheCleaner.__cleanup_cache_lock:
	489	+ if self.__has_space(object_size):
	490	+ # Another thread has done the cleanup for us
	491	+ return 0
	492	+
	493	+ stats = os.statvfs(self.__cas.casdir)
	494	+ target_disk_space = (stats.f_bavail * stats.f_bsize) - self.__max_head_size
	495	+
	496	+ # obtain a list of LRP artifacts
	497	+ LRP_objects = self.__cas.list_objects()
	498	+
	499	+ removed_size = 0 # in bytes
	500	+ last_mtime = 0
	501	+
	502	+ while object_size - removed_size > target_disk_space:
	503	+ try:
	504	+ last_mtime, to_remove = LRP_objects.pop(0) # The first element in the list is the LRP artifact
	505	+ except IndexError:
	506	+ # This exception is caught if there are no more artifacts in the list
	507	+ # LRP_artifacts. This means the the artifact is too large for the filesystem
	508	+ # so we abort the process
	509	+ raise ArtifactTooLargeException("Artifact of size {} is too large for "
	510	+ "the filesystem which mounts the remote "
	511	+ "cache".format(object_size))
	512	+
	513	+ try:
	514	+ size = os.stat(to_remove).st_size
	515	+ os.unlink(to_remove)
	516	+ removed_size += size
	517	+ except FileNotFoundError:
	518	+ pass
	519	+
	520	+ self.__cas.clean_up_refs_until(last_mtime)
	521	+
	522	+ if removed_size > 0:
	523	+ logging.info("Successfully removed {} bytes from the cache".format(removed_size))
	524	+ else:
	525	+ logging.info("No artifacts were removed from the cache.")
	526	+
	527	+ return removed_size

setup.py

---

...	...	@@ -337,7 +337,14 @@ setup(name='BuildStream',
337	337	install_requires=[
338	338	'setuptools',
339	339	'psutil',
340		- 'ruamel.yaml < 0.15.52',
	340	+ # According to ruamel.yaml's PyPI page, we are suppose to use
	341	+ # "<=0.15" in production until 0.15 becomes API stable.
	342	+ # However we need ruamel.yaml 0.15.41 or greater for Python 3.7.
	343	+ # We know that ruamel.yaml 0.15.52 breaks API in a way that
	344	+ # is incompatible with BuildStream.
	345	+ #
	346	+ # See issues #571 and #790.
	347	+ 'ruamel.yaml >= 0.15.41, < 0.15.52',
341	348	'pluginbase',
342	349	'Click',
343	350	'jinja2 >= 2.10',

tests/frontend/push.py

---

...	...	@@ -230,6 +230,8 @@ def test_artifact_expires(cli, datafiles, tmpdir):
230	230	# Create an artifact share (remote artifact cache) in the tmpdir/artifactshare
231	231	# Mock a file system with 12 MB free disk space
232	232	with create_artifact_share(os.path.join(str(tmpdir), 'artifactshare'),
	233	+ min_head_size=int(2e9),
	234	+ max_head_size=int(2e9),
233	235	total_space=int(10e9), free_space=(int(12e6) + int(2e9))) as share:
234	236
235	237	# Configure bst to push to the cache
...	...	@@ -313,6 +315,8 @@ def test_recently_pulled_artifact_does_not_expire(cli, datafiles, tmpdir):
313	315	# Create an artifact share (remote cache) in tmpdir/artifactshare
314	316	# Mock a file system with 12 MB free disk space
315	317	with create_artifact_share(os.path.join(str(tmpdir), 'artifactshare'),
	318	+ min_head_size=int(2e9),
	319	+ max_head_size=int(2e9),
316	320	total_space=int(10e9), free_space=(int(12e6) + int(2e9))) as share:
317	321
318	322	# Configure bst to push to the cache

tests/testutils/artifactshare.py

---

...	...	@@ -29,7 +29,11 @@ from buildstream._protos.build.bazel.remote.execution.v2 import remote_execution
29	29	#
30	30	class ArtifactShare():
31	31
32		- def __init__(self, directory, *, total_space=None, free_space=None):
	32	+ def __init__(self, directory, *,
	33	+ total_space=None,
	34	+ free_space=None,
	35	+ min_head_size=int(2e9),
	36	+ max_head_size=int(10e9)):
33	37
34	38	# The working directory for the artifact share (in case it
35	39	# needs to do something outside of its backend's storage folder).
...	...	@@ -50,6 +54,9 @@ class ArtifactShare():
50	54	self.total_space = total_space
51	55	self.free_space = free_space
52	56
	57	+ self.max_head_size = max_head_size
	58	+ self.min_head_size = min_head_size
	59	+
53	60	q = Queue()
54	61
55	62	self.process = Process(target=self.run, args=(q,))
...	...	@@ -74,7 +81,10 @@ class ArtifactShare():
74	81	self.free_space = self.total_space
75	82	os.statvfs = self._mock_statvfs
76	83
77		- server = create_server(self.repodir, enable_push=True)
	84	+ server = create_server(self.repodir,
	85	+ max_head_size=self.max_head_size,
	86	+ min_head_size=self.min_head_size,
	87	+ enable_push=True)
78	88	port = server.add_insecure_port('localhost:0')
79	89
80	90	server.start()
...	...	@@ -136,6 +146,15 @@ class ArtifactShare():
136	146
137	147	try:
138	148	tree = self.cas.resolve_ref(artifact_key)
	149	+ reachable = set()
	150	+ try:
	151	+ self.cas._reachable_refs_dir(reachable, tree, update_mtime=False)
	152	+ except FileNotFoundError:
	153	+ return None
	154	+ for digest in reachable:
	155	+ object_name = os.path.join(self.cas.casdir, 'objects', digest[:2], digest[2:])
	156	+ if not os.path.exists(object_name):
	157	+ return None
139	158	return tree
140	159	except CASError:
141	160	return None
...	...	@@ -167,8 +186,11 @@ class ArtifactShare():
167	186	# Create an ArtifactShare for use in a test case
168	187	#
169	188	@contextmanager
170		-def create_artifact_share(directory, *, total_space=None, free_space=None):
171		- share = ArtifactShare(directory, total_space=total_space, free_space=free_space)
	189	+def create_artifact_share(directory, *, total_space=None, free_space=None,
	190	+ min_head_size=int(2e9),
	191	+ max_head_size=int(10e9)):
	192	+ share = ArtifactShare(directory, total_space=total_space, free_space=free_space,
	193	+ min_head_size=min_head_size, max_head_size=max_head_size)
172	194	try:
173	195	yield share
174	196	finally:

—
View it on GitLab.
You're receiving this email because of your account on gitlab.com. If you'd like to receive fewer emails, you can adjust your notification settings.