bitbake: hashserv: Support read-only server

The -r/--readonly argument is added to the bitbake-hashserv app. If this
argument is given then clients may only perform read operations against
the server. The read-only mode is implemented by simply not installing
handlers for write operations, this keeps the permission model simple
and reduces the risk of accidentally allowing write operations.

As a sqlite database can be safely opened by multiple processes in
parallel, it's possible to start two hashserv instances against a single
database if you wish to export both a read-only port and a read-write
port.

(Bitbake rev: 492bb02eb0e071c792407ac3113f92492da1a9cc)

Signed-off-by: Paul Barker <pbarker@konsulko.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

4 files changed, 53 insertions, 12 deletions

diff --git a/bitbake/bin/bitbake-hashserv b/bitbake/bin/bitbake-hashserv
index 1bc1f91f38..2669bbd130 100755
--- a/bitbake/bin/bitbake-hashserv
+++ b/bitbake/bin/bitbake-hashserv
@@ -33,6 +33,7 @@ def main():
     parser.add_argument('--bind', default=DEFAULT_BIND, help='Bind address (default "%(default)s")')
     parser.add_argument('--database', default='./hashserv.db', help='Database file (default "%(default)s")')
     parser.add_argument('--log', default='WARNING', help='Set logging level')
+    parser.add_argument('-r', '--read-only', action='store_true', help='Disallow write operations from clients')
 
     args = parser.parse_args()
 
@@ -47,7 +48,7 @@ def main():
     console.setLevel(level)
     logger.addHandler(console)
 
-    server = hashserv.create_server(args.bind, args.database)
+    server = hashserv.create_server(args.bind, args.database, read_only=args.read_only)
     server.serve_forever()
     return 0
 
diff --git a/bitbake/lib/hashserv/__init__.py b/bitbake/lib/hashserv/__init__.py
index 55f48410d3..5f2e101e52 100644
--- a/bitbake/lib/hashserv/__init__.py
+++ b/bitbake/lib/hashserv/__init__.py
@@ -94,10 +94,10 @@ def chunkify(msg, max_chunk):
         yield "\n"
 
 
-def create_server(addr, dbname, *, sync=True, upstream=None):
+def create_server(addr, dbname, *, sync=True, upstream=None, read_only=False):
     from . import server
     db = setup_database(dbname, sync=sync)
-    s = server.Server(db, upstream=upstream)
+    s = server.Server(db, upstream=upstream, read_only=read_only)
 
     (typ, a) = parse_address(addr)
     if typ == ADDR_TYPE_UNIX:
diff --git a/bitbake/lib/hashserv/server.py b/bitbake/lib/hashserv/server.py
index 3ff4c51ccb..2770c23607 100644
--- a/bitbake/lib/hashserv/server.py
+++ b/bitbake/lib/hashserv/server.py
@@ -112,6 +112,9 @@ class Stats(object):
 class ClientError(Exception):
     pass
 
+class ServerError(Exception):
+    pass
+
 def insert_task(cursor, data, ignore=False):
     keys = sorted(data.keys())
     query = '''INSERT%s INTO tasks_v2 (%s) VALUES (%s)''' % (
@@ -138,7 +141,7 @@ class ServerClient(object):
     FAST_QUERY = 'SELECT taskhash, method, unihash FROM tasks_v2 WHERE method=:method AND taskhash=:taskhash ORDER BY created ASC LIMIT 1'
     ALL_QUERY =  'SELECT *                         FROM tasks_v2 WHERE method=:method AND taskhash=:taskhash ORDER BY created ASC LIMIT 1'
 
-    def __init__(self, reader, writer, db, request_stats, backfill_queue, upstream):
+    def __init__(self, reader, writer, db, request_stats, backfill_queue, upstream, read_only):
         self.reader = reader
         self.writer = writer
         self.db = db
@@ -149,15 +152,19 @@ class ServerClient(object):
 
         self.handlers = {
             'get': self.handle_get,
-            'report': self.handle_report,
-            'report-equiv': self.handle_equivreport,
             'get-stream': self.handle_get_stream,
             'get-stats': self.handle_get_stats,
-            'reset-stats': self.handle_reset_stats,
             'chunk-stream': self.handle_chunk,
-            'backfill-wait': self.handle_backfill_wait,
         }
 
+        if not read_only:
+            self.handlers.update({
+                'report': self.handle_report,
+                'report-equiv': self.handle_equivreport,
+                'reset-stats': self.handle_reset_stats,
+                'backfill-wait': self.handle_backfill_wait,
+            })
+
     async def process_requests(self):
         if self.upstream is not None:
             self.upstream_client = await create_async_client(self.upstream)
@@ -455,7 +462,10 @@ class ServerClient(object):
 
 
 class Server(object):
-    def __init__(self, db, loop=None, upstream=None):
+    def __init__(self, db, loop=None, upstream=None, read_only=False):
+        if upstream and read_only:
+            raise ServerError("Read-only hashserv cannot pull from an upstream server")
+
         self.request_stats = Stats()
         self.db = db
 
@@ -467,6 +477,7 @@ class Server(object):
             self.close_loop = False
 
         self.upstream = upstream
+        self.read_only = read_only
 
         self._cleanup_socket = None
 
@@ -510,7 +521,7 @@ class Server(object):
     async def handle_client(self, reader, writer):
         # writer.transport.set_write_buffer_limits(0)
         try:
-            client = ServerClient(reader, writer, self.db, self.request_stats, self.backfill_queue, self.upstream)
+            client = ServerClient(reader, writer, self.db, self.request_stats, self.backfill_queue, self.upstream, self.read_only)
             await client.process_requests()
         except Exception as e:
             import traceback
diff --git a/bitbake/lib/hashserv/tests.py b/bitbake/lib/hashserv/tests.py
index 77a19b8077..6f04e30d61 100644
--- a/bitbake/lib/hashserv/tests.py
+++ b/bitbake/lib/hashserv/tests.py
@@ -6,6 +6,7 @@
 #
 
 from . import create_server, create_client
+from .client import HashConnectionError
 import hashlib
 import logging
 import multiprocessing
@@ -29,7 +30,7 @@ class HashEquivalenceTestSetup(object):
 
     server_index = 0
 
-    def start_server(self, dbpath=None, upstream=None):
+    def start_server(self, dbpath=None, upstream=None, read_only=False):
         self.server_index += 1
         if dbpath is None:
             dbpath = os.path.join(self.temp_dir.name, "db%d.sqlite" % self.server_index)
@@ -38,7 +39,10 @@ class HashEquivalenceTestSetup(object):
             thread.terminate()
             thread.join()
 
-        server = create_server(self.get_server_addr(self.server_index), dbpath, upstream=upstream)
+        server = create_server(self.get_server_addr(self.server_index),
+                               dbpath,
+                               upstream=upstream,
+                               read_only=read_only)
         server.dbpath = dbpath
 
         server.thread = multiprocessing.Process(target=_run_server, args=(server, self.server_index))
@@ -242,6 +246,31 @@ class HashEquivalenceCommonTests(object):
         self.assertClientGetHash(side_client, taskhash4, unihash4)
         self.assertClientGetHash(self.client, taskhash4, None)
 
+    def test_ro_server(self):
+        (ro_client, ro_server) = self.start_server(dbpath=self.server.dbpath, read_only=True)
+
+        # Report a hash via the read-write server
+        taskhash = '35788efcb8dfb0a02659d81cf2bfd695fb30faf9'
+        outhash = '2765d4a5884be49b28601445c2760c5f21e7e5c0ee2b7e3fce98fd7e5970796f'
+        unihash = 'f46d3fbb439bd9b921095da657a4de906510d2cd'
+
+        result = self.client.report_unihash(taskhash, self.METHOD, outhash, unihash)
+        self.assertEqual(result['unihash'], unihash, 'Server returned bad unihash')
+
+        # Check the hash via the read-only server
+        self.assertClientGetHash(ro_client, taskhash, unihash)
+
+        # Ensure that reporting via the read-only server fails
+        taskhash2 = 'c665584ee6817aa99edfc77a44dd853828279370'
+        outhash2 = '3c979c3db45c569f51ab7626a4651074be3a9d11a84b1db076f5b14f7d39db44'
+        unihash2 = '90e9bc1d1f094c51824adca7f8ea79a048d68824'
+
+        with self.assertRaises(HashConnectionError):
+            ro_client.report_unihash(taskhash2, self.METHOD, outhash2, unihash2)
+
+        # Ensure that the database was not modified
+        self.assertClientGetHash(self.client, taskhash2, None)
+
 
 class TestHashEquivalenceUnixServer(HashEquivalenceTestSetup, HashEquivalenceCommonTests, unittest.TestCase):
     def get_server_addr(self, server_idx):