mirror of
https://github.com/ansible/awx.git
synced 2026-10-31 14:36:17 -02:30
AAP-96085: Decouple collection cache from project cache_id (#16710)
* AAP-96085: Decouple collection cache from project cache_id Introduce a requirements-hash-keyed cache (reqcache_<hash>/) so that ansible-galaxy collection/role installs are reused across project commits when requirements files haven't changed. Previously, any commit to the project repo invalidated the entire cache, triggering a full reinstall of all collections — a single collection like community.general 11.0.0 (~3,500 files) adds 40-90s of I/O on SAN-backed storage. * Address review feedback: atomic cache publish, bounded eviction, path-aware hash - Atomic publish: _sync_requirements_cache() copies into a temp directory and renames into place, so a failed copytree cannot leave a partial reqcache directory that looks valid to subsequent cache-hit checks - Bounded eviction: clear_project_cache() evicts the oldest reqcache_* directories when the count exceeds max_reqcache (default 10), preventing unbounded storage growth from repeated requirements changes - Path-aware hash: requirements_hash now includes file paths alongside checksums via zip(), so swapping content between requirements files correctly invalidates the cache * Protect current reqcache from eviction during cleanup Read the requirements hash from stage/.requirements_hash and exclude the matching reqcache_* entry from age-based eviction. Without this, a cache hit on the oldest entry could be evicted before the job uses it. * AAP-96176: Cache tarball for make_local_copy() to reduce small-file I/O After populating reqcache_<hash>/, create a tarball of each subfolder (requirements_collections.tar, requirements_roles.tar). make_local_copy() prefers the tarball when available — copies and extracts one file instead of shutil.copytree() across potentially thousands of individual collection files. Falls back to copytree when no tarball exists (upgrade compatibility). On SAN-backed storage, this significantly reduces per-job copy I/O by eliminating per-file source reads. * Fix direct-cache tarball cross-reference and partial extraction fallback Remove reqcache tarball lookup on direct cache hits — the direct cache could diverge from the reqcache (e.g. galaxy server changes not in the requirements hash), serving stale content. The tarball optimization still works for reqcache fallback hits where cache_subpath + '.tar' resolves directly. Clean up partially extracted destination before copytree fallback so a failed tarball extraction doesn't leave behind a directory that causes FileExistsError. --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Dirk Julich
co-authored by
Claude Opus 4.6
Claude Opus 4.6
parent
918ba2ef36
commit
5613b5ac6c
+111
-3
@@ -9,6 +9,7 @@ import os
|
||||
from pathlib import Path
|
||||
import shutil
|
||||
import stat
|
||||
import tarfile
|
||||
import yaml
|
||||
import tempfile
|
||||
import traceback
|
||||
@@ -1616,18 +1617,97 @@ class RunProjectUpdate(BaseTask):
|
||||
shutil.copytree(awx_playbooks, os.path.join(private_data_dir, 'project'))
|
||||
|
||||
@staticmethod
|
||||
def clear_project_cache(cache_dir, keep_value):
|
||||
def clear_project_cache(cache_dir, keep_value, max_reqcache=10):
|
||||
if os.path.isdir(cache_dir):
|
||||
# Read the current requirements hash from stage so we don't evict it
|
||||
current_reqcache = None
|
||||
stage_hash_file = os.path.join(cache_dir, 'stage', '.requirements_hash')
|
||||
try:
|
||||
if os.path.exists(stage_hash_file):
|
||||
with open(stage_hash_file) as f:
|
||||
req_hash = f.read().strip()
|
||||
if req_hash:
|
||||
current_reqcache = f'reqcache_{req_hash}'
|
||||
except OSError:
|
||||
pass
|
||||
|
||||
reqcache_entries = []
|
||||
for entry in os.listdir(cache_dir):
|
||||
old_path = os.path.join(cache_dir, entry)
|
||||
if entry not in (keep_value, 'stage'):
|
||||
# invalidate, then delete
|
||||
if entry.startswith('reqcache_'):
|
||||
reqcache_entries.append(old_path)
|
||||
elif entry not in (keep_value, 'stage'):
|
||||
new_path = os.path.join(cache_dir, '.~~delete~~' + entry)
|
||||
try:
|
||||
os.rename(old_path, new_path)
|
||||
shutil.rmtree(new_path)
|
||||
except OSError:
|
||||
logger.warning(f"Could not remove cache directory {old_path}")
|
||||
if len(reqcache_entries) > max_reqcache:
|
||||
reqcache_entries.sort(key=lambda p: os.path.getmtime(p))
|
||||
for stale in reqcache_entries[: len(reqcache_entries) - max_reqcache]:
|
||||
if current_reqcache and os.path.basename(stale) == current_reqcache:
|
||||
continue
|
||||
try:
|
||||
shutil.rmtree(stale)
|
||||
logger.debug(f'Evicted stale requirements cache {stale}')
|
||||
except OSError:
|
||||
logger.warning(f"Could not remove stale requirements cache {stale}")
|
||||
|
||||
@staticmethod
|
||||
def _sync_requirements_cache(base_path, cache_path):
|
||||
"""After a successful install, sync the requirements hash cache.
|
||||
|
||||
If galaxy install ran (collections/roles exist in cache_path), save them
|
||||
to a hash-keyed reqcache directory for reuse across commits.
|
||||
If galaxy install was skipped (cache hit), the reqcache already has the content.
|
||||
"""
|
||||
hash_file = os.path.join(cache_path, '.requirements_hash')
|
||||
if not os.path.exists(hash_file):
|
||||
return
|
||||
|
||||
try:
|
||||
with open(hash_file) as f:
|
||||
req_hash = f.read().strip()
|
||||
except OSError:
|
||||
return
|
||||
|
||||
if not req_hash:
|
||||
return
|
||||
|
||||
reqcache_path = os.path.join(base_path, f'reqcache_{req_hash}')
|
||||
|
||||
for subfolder in ('requirements_collections', 'requirements_roles'):
|
||||
src = os.path.join(cache_path, subfolder)
|
||||
dst = os.path.join(reqcache_path, subfolder)
|
||||
if os.path.exists(src) and not os.path.exists(dst):
|
||||
tmp_path = tempfile.mkdtemp(dir=base_path, prefix=f'.reqcache_{req_hash}_tmp_')
|
||||
try:
|
||||
tmp_dst = os.path.join(tmp_path, subfolder)
|
||||
shutil.copytree(src, tmp_dst, symlinks=True)
|
||||
os.makedirs(reqcache_path, exist_ok=True)
|
||||
os.rename(tmp_dst, dst)
|
||||
logger.debug(f'Saved {subfolder} to requirements cache {reqcache_path}')
|
||||
except Exception:
|
||||
logger.warning(f'Failed to save {subfolder} to requirements cache, cleaning up')
|
||||
raise
|
||||
finally:
|
||||
shutil.rmtree(tmp_path, ignore_errors=True)
|
||||
|
||||
# Create a cached tarball for faster per-job copies (AAP-96176)
|
||||
tar_path = os.path.join(reqcache_path, f'{subfolder}.tar')
|
||||
if os.path.exists(dst) and not os.path.exists(tar_path):
|
||||
tmp_tar = tar_path + '.tmp'
|
||||
try:
|
||||
with tarfile.open(tmp_tar, 'w') as tar:
|
||||
tar.add(dst, arcname=subfolder)
|
||||
os.rename(tmp_tar, tar_path)
|
||||
logger.debug(f'Created tarball cache {tar_path}')
|
||||
except Exception:
|
||||
logger.warning(f'Failed to create tarball cache {tar_path}')
|
||||
if os.path.exists(tmp_tar):
|
||||
os.remove(tmp_tar)
|
||||
# Non-fatal: copytree fallback still works
|
||||
|
||||
@staticmethod
|
||||
def make_local_copy(project, job_private_data_dir):
|
||||
@@ -1649,8 +1729,34 @@ class RunProjectUpdate(BaseTask):
|
||||
subfolders.append('requirements_roles')
|
||||
for subfolder in subfolders:
|
||||
cache_subpath = os.path.join(cache_path, subfolder)
|
||||
if not os.path.exists(cache_subpath):
|
||||
# Fall back to requirements-hash-keyed cache (reqcache)
|
||||
hash_file = os.path.join(cache_path, '.requirements_hash')
|
||||
if os.path.exists(hash_file):
|
||||
try:
|
||||
with open(hash_file) as f:
|
||||
req_hash = f.read().strip()
|
||||
if req_hash:
|
||||
reqcache_subpath = os.path.join(project.get_cache_path(), f'reqcache_{req_hash}', subfolder)
|
||||
if os.path.exists(reqcache_subpath):
|
||||
cache_subpath = reqcache_subpath
|
||||
logger.debug(f'Using requirements cache for {subfolder} (hash {req_hash[:12]})')
|
||||
except OSError:
|
||||
pass
|
||||
if os.path.exists(cache_subpath):
|
||||
dest_subpath = os.path.join(job_private_data_dir, subfolder)
|
||||
# Prefer cached tarball for fewer file operations on SAN storage
|
||||
tar_path = cache_subpath + '.tar'
|
||||
if os.path.exists(tar_path):
|
||||
try:
|
||||
with tarfile.open(tar_path, 'r') as tar:
|
||||
tar.extractall(path=job_private_data_dir, filter='data')
|
||||
logger.debug('{0} {1} prepared {2} from tarball'.format(type(project).__name__, project.pk, dest_subpath))
|
||||
continue
|
||||
except Exception:
|
||||
logger.warning(f'Failed to extract tarball {tar_path}, falling back to copytree')
|
||||
if os.path.exists(dest_subpath):
|
||||
shutil.rmtree(dest_subpath)
|
||||
shutil.copytree(cache_subpath, dest_subpath, symlinks=True)
|
||||
logger.debug('{0} {1} prepared {2} from cache'.format(type(project).__name__, project.pk, dest_subpath))
|
||||
|
||||
@@ -1685,6 +1791,8 @@ class RunProjectUpdate(BaseTask):
|
||||
shutil.rmtree(cache_path)
|
||||
os.rename(stage_path, cache_path)
|
||||
logger.debug('{0} wrote to cache at {1}'.format(instance.log_format, cache_path))
|
||||
|
||||
self._sync_requirements_cache(base_path, cache_path)
|
||||
elif os.path.exists(stage_path):
|
||||
shutil.rmtree(stage_path) # cannot trust content update produced
|
||||
|
||||
|
||||
@@ -1,6 +1,7 @@
|
||||
# -*- coding: utf-8 -*-
|
||||
import json
|
||||
import os
|
||||
import tarfile
|
||||
from pathlib import Path
|
||||
|
||||
import fcntl
|
||||
@@ -1767,6 +1768,442 @@ def test_sync_and_copy_get_sync_needs_called_under_shared_lock(get_sync_needs, s
|
||||
assert call_order[1] == 'get_sync_needs'
|
||||
|
||||
|
||||
class TestClearProjectCachePreservesReqcache:
|
||||
def test_reqcache_entries_preserved(self, tmp_path):
|
||||
cache_dir = tmp_path / 'cache'
|
||||
cache_dir.mkdir()
|
||||
(cache_dir / 'abc123').mkdir()
|
||||
(cache_dir / 'reqcache_deadbeef').mkdir()
|
||||
(cache_dir / 'reqcache_cafebabe').mkdir()
|
||||
(cache_dir / 'old_revision').mkdir()
|
||||
(cache_dir / 'stage').mkdir()
|
||||
|
||||
jobs.RunProjectUpdate.clear_project_cache(str(cache_dir), 'abc123')
|
||||
|
||||
remaining = set(os.listdir(cache_dir))
|
||||
assert 'abc123' in remaining
|
||||
assert 'reqcache_deadbeef' in remaining
|
||||
assert 'reqcache_cafebabe' in remaining
|
||||
assert 'stage' in remaining
|
||||
assert 'old_revision' not in remaining
|
||||
|
||||
def test_no_reqcache_entries(self, tmp_path):
|
||||
cache_dir = tmp_path / 'cache'
|
||||
cache_dir.mkdir()
|
||||
(cache_dir / 'keep_me').mkdir()
|
||||
(cache_dir / 'delete_me').mkdir()
|
||||
|
||||
jobs.RunProjectUpdate.clear_project_cache(str(cache_dir), 'keep_me')
|
||||
|
||||
remaining = set(os.listdir(cache_dir))
|
||||
assert 'keep_me' in remaining
|
||||
assert 'delete_me' not in remaining
|
||||
|
||||
def test_evicts_oldest_reqcache_when_over_limit(self, tmp_path):
|
||||
cache_dir = tmp_path / 'cache'
|
||||
cache_dir.mkdir()
|
||||
(cache_dir / 'current').mkdir()
|
||||
|
||||
for i in range(12):
|
||||
entry = cache_dir / f'reqcache_hash{i:02d}'
|
||||
entry.mkdir()
|
||||
os.utime(str(entry), (1000 + i, 1000 + i))
|
||||
|
||||
jobs.RunProjectUpdate.clear_project_cache(str(cache_dir), 'current', max_reqcache=10)
|
||||
|
||||
remaining = set(os.listdir(cache_dir))
|
||||
assert 'reqcache_hash00' not in remaining
|
||||
assert 'reqcache_hash01' not in remaining
|
||||
assert 'reqcache_hash02' in remaining
|
||||
assert 'reqcache_hash11' in remaining
|
||||
assert len([e for e in remaining if e.startswith('reqcache_')]) == 10
|
||||
|
||||
def test_eviction_protects_current_reqcache(self, tmp_path):
|
||||
cache_dir = tmp_path / 'cache'
|
||||
cache_dir.mkdir()
|
||||
(cache_dir / 'current').mkdir()
|
||||
|
||||
stage_dir = cache_dir / 'stage'
|
||||
stage_dir.mkdir()
|
||||
(stage_dir / '.requirements_hash').write_text('hash00')
|
||||
|
||||
for i in range(12):
|
||||
entry = cache_dir / f'reqcache_hash{i:02d}'
|
||||
entry.mkdir()
|
||||
os.utime(str(entry), (1000 + i, 1000 + i))
|
||||
|
||||
jobs.RunProjectUpdate.clear_project_cache(str(cache_dir), 'current', max_reqcache=10)
|
||||
|
||||
remaining = set(os.listdir(cache_dir))
|
||||
# hash00 is the oldest but should be protected (current update's hash)
|
||||
assert 'reqcache_hash00' in remaining
|
||||
# hash01 is the next oldest and should be evicted
|
||||
assert 'reqcache_hash01' not in remaining
|
||||
# hash02+ should survive (only 1 non-protected entry evicted)
|
||||
assert 'reqcache_hash02' in remaining
|
||||
assert 'reqcache_hash11' in remaining
|
||||
|
||||
|
||||
class TestSyncRequirementsCache:
|
||||
def test_saves_collections_to_reqcache(self, tmp_path):
|
||||
base_path = tmp_path / 'cache'
|
||||
base_path.mkdir()
|
||||
cache_path = base_path / 'abc123'
|
||||
cache_path.mkdir()
|
||||
|
||||
# Simulate a galaxy install that produced collections
|
||||
collections_dir = cache_path / 'requirements_collections'
|
||||
collections_dir.mkdir()
|
||||
(collections_dir / 'ansible_collections').mkdir()
|
||||
(collections_dir / 'ansible_collections' / 'test.txt').write_text('test')
|
||||
|
||||
# Write the hash marker
|
||||
(cache_path / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
jobs.RunProjectUpdate._sync_requirements_cache(str(base_path), str(cache_path))
|
||||
|
||||
reqcache = base_path / 'reqcache_deadbeef1234' / 'requirements_collections'
|
||||
assert reqcache.exists()
|
||||
assert (reqcache / 'ansible_collections' / 'test.txt').read_text() == 'test'
|
||||
|
||||
def test_skips_when_no_hash_file(self, tmp_path):
|
||||
base_path = tmp_path / 'cache'
|
||||
base_path.mkdir()
|
||||
cache_path = base_path / 'abc123'
|
||||
cache_path.mkdir()
|
||||
|
||||
jobs.RunProjectUpdate._sync_requirements_cache(str(base_path), str(cache_path))
|
||||
|
||||
entries = os.listdir(base_path)
|
||||
assert not any(e.startswith('reqcache_') for e in entries)
|
||||
|
||||
def test_no_partial_cache_on_copytree_failure(self, tmp_path):
|
||||
base_path = tmp_path / 'cache'
|
||||
base_path.mkdir()
|
||||
cache_path = base_path / 'abc123'
|
||||
cache_path.mkdir()
|
||||
|
||||
collections_dir = cache_path / 'requirements_collections'
|
||||
collections_dir.mkdir()
|
||||
(collections_dir / 'test.txt').write_text('test')
|
||||
(cache_path / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
with mock.patch('shutil.copytree', side_effect=OSError('disk full')):
|
||||
try:
|
||||
jobs.RunProjectUpdate._sync_requirements_cache(str(base_path), str(cache_path))
|
||||
except OSError:
|
||||
pass
|
||||
|
||||
reqcache = base_path / 'reqcache_deadbeef1234'
|
||||
assert not reqcache.exists()
|
||||
tmp_dirs = [e for e in os.listdir(base_path) if e.startswith('.reqcache_')]
|
||||
assert len(tmp_dirs) == 0
|
||||
|
||||
def test_creates_tarball_after_sync(self, tmp_path):
|
||||
base_path = tmp_path / 'cache'
|
||||
base_path.mkdir()
|
||||
cache_path = base_path / 'abc123'
|
||||
cache_path.mkdir()
|
||||
|
||||
collections_dir = cache_path / 'requirements_collections'
|
||||
collections_dir.mkdir()
|
||||
(collections_dir / 'ansible_collections').mkdir()
|
||||
(collections_dir / 'ansible_collections' / 'test.txt').write_text('test')
|
||||
(cache_path / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
jobs.RunProjectUpdate._sync_requirements_cache(str(base_path), str(cache_path))
|
||||
|
||||
tar_path = base_path / 'reqcache_deadbeef1234' / 'requirements_collections.tar'
|
||||
assert tar_path.exists()
|
||||
with tarfile.open(str(tar_path), 'r') as tar:
|
||||
names = tar.getnames()
|
||||
assert 'requirements_collections/ansible_collections/test.txt' in names
|
||||
|
||||
def test_skips_when_reqcache_already_exists(self, tmp_path):
|
||||
base_path = tmp_path / 'cache'
|
||||
base_path.mkdir()
|
||||
cache_path = base_path / 'abc123'
|
||||
cache_path.mkdir()
|
||||
|
||||
# Pre-existing reqcache
|
||||
reqcache = base_path / 'reqcache_deadbeef1234' / 'requirements_collections'
|
||||
reqcache.mkdir(parents=True)
|
||||
(reqcache / 'existing.txt').write_text('original')
|
||||
|
||||
# New cache with same hash but different content
|
||||
collections_dir = cache_path / 'requirements_collections'
|
||||
collections_dir.mkdir()
|
||||
(collections_dir / 'new.txt').write_text('new')
|
||||
(cache_path / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
jobs.RunProjectUpdate._sync_requirements_cache(str(base_path), str(cache_path))
|
||||
|
||||
# Should not overwrite existing reqcache
|
||||
assert (reqcache / 'existing.txt').read_text() == 'original'
|
||||
assert not (reqcache / 'new.txt').exists()
|
||||
|
||||
|
||||
class TestMakeLocalCopyReqcacheFallback:
|
||||
def test_falls_back_to_reqcache(self, tmp_path):
|
||||
cache_base = tmp_path / 'cache'
|
||||
cache_base.mkdir()
|
||||
|
||||
# Create cache_id directory with hash pointer but no collections
|
||||
cache_id_dir = cache_base / 'abc123'
|
||||
cache_id_dir.mkdir()
|
||||
(cache_id_dir / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
# Create reqcache with collections
|
||||
reqcache_collections = cache_base / 'reqcache_deadbeef1234' / 'requirements_collections'
|
||||
reqcache_collections.mkdir(parents=True)
|
||||
(reqcache_collections / 'community').mkdir()
|
||||
(reqcache_collections / 'community' / 'general.txt').write_text('collection content')
|
||||
|
||||
# Create project path
|
||||
project_path = tmp_path / 'project'
|
||||
project_path.mkdir()
|
||||
(project_path / 'playbook.yml').write_text('- hosts: all')
|
||||
|
||||
# Create job private data dir
|
||||
job_dir = tmp_path / 'job'
|
||||
job_dir.mkdir()
|
||||
|
||||
project = mock.Mock()
|
||||
project.get_project_path.return_value = str(project_path)
|
||||
project.get_cache_path.return_value = str(cache_base)
|
||||
project.cache_id = 'abc123'
|
||||
project.pk = 1
|
||||
|
||||
with mock.patch('awx.main.tasks.jobs.settings') as mock_settings:
|
||||
mock_settings.AWX_COLLECTIONS_ENABLED = True
|
||||
mock_settings.AWX_ROLES_ENABLED = False
|
||||
jobs.RunProjectUpdate.make_local_copy(project, str(job_dir))
|
||||
|
||||
# Collections should be copied from reqcache
|
||||
dest = job_dir / 'requirements_collections' / 'community' / 'general.txt'
|
||||
assert dest.exists()
|
||||
assert dest.read_text() == 'collection content'
|
||||
|
||||
def test_uses_tarball_from_reqcache(self, tmp_path):
|
||||
cache_base = tmp_path / 'cache'
|
||||
cache_base.mkdir()
|
||||
|
||||
cache_id_dir = cache_base / 'abc123'
|
||||
cache_id_dir.mkdir()
|
||||
(cache_id_dir / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
# Create reqcache with collections directory AND tarball
|
||||
reqcache = cache_base / 'reqcache_deadbeef1234'
|
||||
reqcache.mkdir()
|
||||
reqcache_collections = reqcache / 'requirements_collections'
|
||||
reqcache_collections.mkdir()
|
||||
(reqcache_collections / 'community').mkdir()
|
||||
(reqcache_collections / 'community' / 'general.txt').write_text('collection content')
|
||||
|
||||
tar_path = reqcache / 'requirements_collections.tar'
|
||||
with tarfile.open(str(tar_path), 'w') as tar:
|
||||
tar.add(str(reqcache_collections), arcname='requirements_collections')
|
||||
|
||||
project_path = tmp_path / 'project'
|
||||
project_path.mkdir()
|
||||
(project_path / 'playbook.yml').write_text('- hosts: all')
|
||||
|
||||
job_dir = tmp_path / 'job'
|
||||
job_dir.mkdir()
|
||||
|
||||
project = mock.Mock()
|
||||
project.get_project_path.return_value = str(project_path)
|
||||
project.get_cache_path.return_value = str(cache_base)
|
||||
project.cache_id = 'abc123'
|
||||
project.pk = 1
|
||||
|
||||
with mock.patch('awx.main.tasks.jobs.settings') as mock_settings:
|
||||
mock_settings.AWX_COLLECTIONS_ENABLED = True
|
||||
mock_settings.AWX_ROLES_ENABLED = False
|
||||
jobs.RunProjectUpdate.make_local_copy(project, str(job_dir))
|
||||
|
||||
dest = job_dir / 'requirements_collections' / 'community' / 'general.txt'
|
||||
assert dest.exists()
|
||||
assert dest.read_text() == 'collection content'
|
||||
|
||||
def test_direct_cache_ignores_reqcache_tarball(self, tmp_path):
|
||||
"""Direct cache hit must use copytree, not a reqcache tarball that could have stale content."""
|
||||
cache_base = tmp_path / 'cache'
|
||||
cache_base.mkdir()
|
||||
|
||||
cache_id_dir = cache_base / 'abc123'
|
||||
cache_id_dir.mkdir()
|
||||
direct_collections = cache_id_dir / 'requirements_collections'
|
||||
direct_collections.mkdir()
|
||||
(direct_collections / 'direct.txt').write_text('direct content')
|
||||
(cache_id_dir / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
# Reqcache tarball with DIFFERENT content — must not be used
|
||||
reqcache = cache_base / 'reqcache_deadbeef1234'
|
||||
reqcache.mkdir()
|
||||
stale_dir = tmp_path / '_stale'
|
||||
stale_dir.mkdir()
|
||||
stale_collections = stale_dir / 'requirements_collections'
|
||||
stale_collections.mkdir()
|
||||
(stale_collections / 'direct.txt').write_text('stale tarball content')
|
||||
tar_path = reqcache / 'requirements_collections.tar'
|
||||
with tarfile.open(str(tar_path), 'w') as tar:
|
||||
tar.add(str(stale_collections), arcname='requirements_collections')
|
||||
|
||||
project_path = tmp_path / 'project'
|
||||
project_path.mkdir()
|
||||
(project_path / 'playbook.yml').write_text('- hosts: all')
|
||||
|
||||
job_dir = tmp_path / 'job'
|
||||
job_dir.mkdir()
|
||||
|
||||
project = mock.Mock()
|
||||
project.get_project_path.return_value = str(project_path)
|
||||
project.get_cache_path.return_value = str(cache_base)
|
||||
project.cache_id = 'abc123'
|
||||
project.pk = 1
|
||||
|
||||
with mock.patch('awx.main.tasks.jobs.settings') as mock_settings:
|
||||
mock_settings.AWX_COLLECTIONS_ENABLED = True
|
||||
mock_settings.AWX_ROLES_ENABLED = False
|
||||
jobs.RunProjectUpdate.make_local_copy(project, str(job_dir))
|
||||
|
||||
dest = job_dir / 'requirements_collections' / 'direct.txt'
|
||||
assert dest.exists()
|
||||
assert dest.read_text() == 'direct content'
|
||||
|
||||
def test_tarball_partial_extract_falls_back_to_copytree(self, tmp_path):
|
||||
"""If tarball extraction partially writes then fails, copytree fallback must still work."""
|
||||
cache_base = tmp_path / 'cache'
|
||||
cache_base.mkdir()
|
||||
|
||||
cache_id_dir = cache_base / 'abc123'
|
||||
cache_id_dir.mkdir()
|
||||
(cache_id_dir / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
reqcache = cache_base / 'reqcache_deadbeef1234'
|
||||
reqcache.mkdir()
|
||||
reqcache_collections = reqcache / 'requirements_collections'
|
||||
reqcache_collections.mkdir()
|
||||
(reqcache_collections / 'community').mkdir()
|
||||
(reqcache_collections / 'community' / 'general.txt').write_text('collection content')
|
||||
|
||||
# Create a valid tarball that we'll make fail during extraction
|
||||
tar_path = reqcache / 'requirements_collections.tar'
|
||||
with tarfile.open(str(tar_path), 'w') as tar:
|
||||
tar.add(str(reqcache_collections), arcname='requirements_collections')
|
||||
|
||||
project_path = tmp_path / 'project'
|
||||
project_path.mkdir()
|
||||
(project_path / 'playbook.yml').write_text('- hosts: all')
|
||||
|
||||
job_dir = tmp_path / 'job'
|
||||
job_dir.mkdir()
|
||||
|
||||
# Simulate partial extraction: create dest dir to mimic what a failed extractall leaves behind
|
||||
partial_dest = job_dir / 'requirements_collections'
|
||||
partial_dest.mkdir()
|
||||
|
||||
project = mock.Mock()
|
||||
project.get_project_path.return_value = str(project_path)
|
||||
project.get_cache_path.return_value = str(cache_base)
|
||||
project.cache_id = 'abc123'
|
||||
project.pk = 1
|
||||
|
||||
original_open = tarfile.open
|
||||
|
||||
def failing_tarfile_open(*args, **kwargs):
|
||||
tar = original_open(*args, **kwargs)
|
||||
|
||||
def raise_after_partial(*a, **kw):
|
||||
raise OSError("Simulated extraction failure")
|
||||
|
||||
tar.extractall = raise_after_partial
|
||||
return tar
|
||||
|
||||
with mock.patch('awx.main.tasks.jobs.settings') as mock_settings:
|
||||
mock_settings.AWX_COLLECTIONS_ENABLED = True
|
||||
mock_settings.AWX_ROLES_ENABLED = False
|
||||
with mock.patch('awx.main.tasks.jobs.tarfile.open', side_effect=failing_tarfile_open):
|
||||
jobs.RunProjectUpdate.make_local_copy(project, str(job_dir))
|
||||
|
||||
dest = job_dir / 'requirements_collections' / 'community' / 'general.txt'
|
||||
assert dest.exists()
|
||||
assert dest.read_text() == 'collection content'
|
||||
|
||||
def test_falls_back_to_copytree_when_no_tarball(self, tmp_path):
|
||||
cache_base = tmp_path / 'cache'
|
||||
cache_base.mkdir()
|
||||
|
||||
cache_id_dir = cache_base / 'abc123'
|
||||
cache_id_dir.mkdir()
|
||||
(cache_id_dir / '.requirements_hash').write_text('deadbeef1234')
|
||||
|
||||
# Reqcache with directory but NO tarball
|
||||
reqcache_collections = cache_base / 'reqcache_deadbeef1234' / 'requirements_collections'
|
||||
reqcache_collections.mkdir(parents=True)
|
||||
(reqcache_collections / 'community').mkdir()
|
||||
(reqcache_collections / 'community' / 'general.txt').write_text('collection content')
|
||||
|
||||
project_path = tmp_path / 'project'
|
||||
project_path.mkdir()
|
||||
(project_path / 'playbook.yml').write_text('- hosts: all')
|
||||
|
||||
job_dir = tmp_path / 'job'
|
||||
job_dir.mkdir()
|
||||
|
||||
project = mock.Mock()
|
||||
project.get_project_path.return_value = str(project_path)
|
||||
project.get_cache_path.return_value = str(cache_base)
|
||||
project.cache_id = 'abc123'
|
||||
project.pk = 1
|
||||
|
||||
with mock.patch('awx.main.tasks.jobs.settings') as mock_settings:
|
||||
mock_settings.AWX_COLLECTIONS_ENABLED = True
|
||||
mock_settings.AWX_ROLES_ENABLED = False
|
||||
jobs.RunProjectUpdate.make_local_copy(project, str(job_dir))
|
||||
|
||||
# Should still work via copytree fallback
|
||||
dest = job_dir / 'requirements_collections' / 'community' / 'general.txt'
|
||||
assert dest.exists()
|
||||
assert dest.read_text() == 'collection content'
|
||||
|
||||
def test_uses_direct_cache_when_available(self, tmp_path):
|
||||
cache_base = tmp_path / 'cache'
|
||||
cache_base.mkdir()
|
||||
|
||||
# Create cache_id directory with collections directly present
|
||||
cache_id_dir = cache_base / 'abc123'
|
||||
cache_id_dir.mkdir()
|
||||
direct_collections = cache_id_dir / 'requirements_collections'
|
||||
direct_collections.mkdir()
|
||||
(direct_collections / 'direct.txt').write_text('direct content')
|
||||
|
||||
# Create project path
|
||||
project_path = tmp_path / 'project'
|
||||
project_path.mkdir()
|
||||
(project_path / 'playbook.yml').write_text('- hosts: all')
|
||||
|
||||
job_dir = tmp_path / 'job'
|
||||
job_dir.mkdir()
|
||||
|
||||
project = mock.Mock()
|
||||
project.get_project_path.return_value = str(project_path)
|
||||
project.get_cache_path.return_value = str(cache_base)
|
||||
project.cache_id = 'abc123'
|
||||
project.pk = 1
|
||||
|
||||
with mock.patch('awx.main.tasks.jobs.settings') as mock_settings:
|
||||
mock_settings.AWX_COLLECTIONS_ENABLED = True
|
||||
mock_settings.AWX_ROLES_ENABLED = False
|
||||
jobs.RunProjectUpdate.make_local_copy(project, str(job_dir))
|
||||
|
||||
# Should use direct cache, not reqcache
|
||||
dest = job_dir / 'requirements_collections' / 'direct.txt'
|
||||
assert dest.exists()
|
||||
assert dest.read_text() == 'direct content'
|
||||
|
||||
|
||||
@pytest.mark.parametrize('injector_cls', [cls for cls in ManagedCredentialType.registry.values() if cls.injectors])
|
||||
def test_managed_injector_redaction(injector_cls):
|
||||
"""See awx.main.models.inventory.PluginFileInjector._get_shared_env
|
||||
|
||||
@@ -198,14 +198,15 @@
|
||||
name: Install content with ansible-galaxy command if necessary
|
||||
vars:
|
||||
galaxy_task_env: # configured in settings
|
||||
awx_cache_path: "{{ projects_root }}/.__awx_cache/{{ local_path }}"
|
||||
# additional_galaxy_env contains environment variables are used for installing roles and collections and will take precedence over items in galaxy_task_env
|
||||
additional_galaxy_env:
|
||||
# These paths control where ansible-galaxy installs collections and roles on top the filesystem
|
||||
ANSIBLE_COLLECTIONS_PATH: "{{ projects_root }}/.__awx_cache/{{ local_path }}/stage/requirements_collections"
|
||||
ANSIBLE_ROLES_PATH: "{{ projects_root }}/.__awx_cache/{{ local_path }}/stage/requirements_roles"
|
||||
ANSIBLE_COLLECTIONS_PATH: "{{ awx_cache_path }}/stage/requirements_collections"
|
||||
ANSIBLE_ROLES_PATH: "{{ awx_cache_path }}/stage/requirements_roles"
|
||||
# Put the local tmp directory in same volume as collection destination
|
||||
# otherwise, files cannot be moved accross volumes and will cause error
|
||||
ANSIBLE_LOCAL_TEMP: "{{ projects_root }}/.__awx_cache/{{ local_path }}/stage/tmp"
|
||||
ANSIBLE_LOCAL_TEMP: "{{ awx_cache_path }}/stage/tmp"
|
||||
tasks:
|
||||
- name: Check content sync settings
|
||||
when: not roles_enabled | bool and not collections_enabled | bool
|
||||
@@ -221,6 +222,49 @@
|
||||
- name: End play due to disabled content sync
|
||||
ansible.builtin.meta: end_play
|
||||
|
||||
- name: Compute requirements hash for cache reuse
|
||||
tags:
|
||||
- install_roles
|
||||
- install_collections
|
||||
block:
|
||||
- name: Find all requirements files
|
||||
ansible.builtin.stat:
|
||||
path: "{{ item }}"
|
||||
checksum_algorithm: sha256
|
||||
get_attributes: false
|
||||
get_mime: false
|
||||
register: req_files_stat
|
||||
loop:
|
||||
- "{{ project_path | quote }}/collections/requirements.yml"
|
||||
- "{{ project_path | quote }}/collections/requirements.yaml"
|
||||
- "{{ project_path | quote }}/roles/requirements.yml"
|
||||
- "{{ project_path | quote }}/roles/requirements.yaml"
|
||||
- "{{ project_path | quote }}/requirements.yml"
|
||||
- "{{ project_path | quote }}/requirements.yaml"
|
||||
|
||||
- name: Build combined requirements hash
|
||||
ansible.builtin.set_fact:
|
||||
requirements_hash: "{{ _existing | map(attribute='item') | zip(_existing | map(attribute='stat.checksum')) | list | to_json | hash('sha256') }}"
|
||||
has_requirements: "{{ _existing | length > 0 }}"
|
||||
vars:
|
||||
_existing: "{{ req_files_stat.results | selectattr('stat.exists', 'defined') | selectattr('stat.exists') | list }}"
|
||||
|
||||
- name: Check requirements hash cache
|
||||
ansible.builtin.stat:
|
||||
path: "{{ awx_cache_path }}/reqcache_{{ requirements_hash }}"
|
||||
register: reqcache_stat
|
||||
when: has_requirements | bool
|
||||
|
||||
- name: Write requirements hash marker for post-run cache management
|
||||
ansible.builtin.copy:
|
||||
content: "{{ requirements_hash }}"
|
||||
dest: "{{ awx_cache_path }}/stage/.requirements_hash"
|
||||
when: has_requirements | bool
|
||||
|
||||
- name: Set cache hit fact
|
||||
ansible.builtin.set_fact:
|
||||
requirements_cache_hit: "{{ has_requirements | bool and (reqcache_stat.stat.exists | default(false)) }}"
|
||||
|
||||
- block:
|
||||
- name: Fetch galaxy roles from roles/requirements.(yml/yaml)
|
||||
ansible.builtin.command:
|
||||
@@ -237,6 +281,7 @@
|
||||
when:
|
||||
- roles_enabled | bool
|
||||
- req_file | length > 0
|
||||
- not (requirements_cache_hit | default(false) | bool)
|
||||
tags:
|
||||
- install_roles
|
||||
|
||||
@@ -255,6 +300,7 @@
|
||||
when:
|
||||
- collections_enabled | bool
|
||||
- req_file | length > 0
|
||||
- not (requirements_cache_hit | default(false) | bool)
|
||||
tags:
|
||||
- install_collections
|
||||
|
||||
@@ -275,6 +321,7 @@
|
||||
- collections_enabled | bool
|
||||
- roles_enabled | bool
|
||||
- req_file | length > 0
|
||||
- not (requirements_cache_hit | default(false) | bool)
|
||||
tags:
|
||||
- install_collections
|
||||
- install_roles
|
||||
|
||||
Reference in New Issue
Block a user