X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/9f4fd542a9fc94e9f48387e90fd70b614458c1f2..1efba8f3b728a3b8aa3c64c5aa09f441318ff2a8:/services/fuse/arvados_fuse/command.py

diff --git a/services/fuse/arvados_fuse/command.py b/services/fuse/arvados_fuse/command.py
index 5f0a1f80f6..9c607c7f0c 100644
--- a/services/fuse/arvados_fuse/command.py
+++ b/services/fuse/arvados_fuse/command.py
@@ -16,6 +16,7 @@ import signal
 import subprocess
 import sys
 import time
+import resource
 
 import arvados.commands._util as arv_cmd
 from arvados_fuse import crunchstat
@@ -83,13 +84,20 @@ class ArgumentParser(argparse.ArgumentParser):
                             type=str, metavar='PATH', action='append', default=[],
                             help="Create a new collection, mount it in read/write mode at mountpoint/PATH, and delete it when unmounting.")
 
+
         self.add_argument('--debug', action='store_true', help="""Debug mode""")
         self.add_argument('--logfile', help="""Write debug logs and errors to the specified file (default stderr).""")
         self.add_argument('--foreground', action='store_true', help="""Run in foreground (default is to daemonize unless --exec specified)""", default=False)
         self.add_argument('--encoding', type=str, help="Character encoding to use for filesystem, default is utf-8 (see Python codec registry for list of available encodings)", default="utf-8")
 
-        self.add_argument('--file-cache', type=int, help="File data cache size, in bytes (default 256MiB)", default=256*1024*1024)
-        self.add_argument('--directory-cache', type=int, help="Directory data cache size, in bytes (default 128MiB)", default=128*1024*1024)
+        self.add_argument('--file-cache', type=int, help="File data cache size, in bytes (default 8 GiB for disk-based cache or 256 MiB with RAM-only cache)", default=0)
+        self.add_argument('--directory-cache', type=int, help="Directory data cache size, in bytes (default 128 MiB)", default=128*1024*1024)
+
+        cachetype = self.add_mutually_exclusive_group()
+        cachetype.add_argument('--ram-cache', action='store_false', dest='disk_cache', help="Use in-memory caching only", default=True)
+        cachetype.add_argument('--disk-cache', action='store_true', dest='disk_cache', help="Use disk based caching (default)", default=True)
+
+        self.add_argument('--disk-cache-dir', type=str, help="Disk cache location (default ~/.cache/arvados/keep)", default=None)
 
         self.add_argument('--disable-event-listening', action='store_true', help="Don't subscribe to events on the API server", dest="disable_event_listening", default=False)
 
@@ -128,10 +136,38 @@ class Mount(object):
 
         try:
             self._setup_logging()
+        except Exception as e:
+            self.logger.exception("exception during setup: %s", e)
+            exit(1)
+
+        try:
+            nofile_limit = resource.getrlimit(resource.RLIMIT_NOFILE)
+
+            minlimit = 10240
+            if self.args.file_cache:
+                # Adjust the file handle limit so it can meet
+                # the desired cache size. Multiply by 8 because the
+                # number of 64 MiB cache slots that keepclient
+                # allocates is RLIMIT_NOFILE / 8
+                minlimit = int((self.args.file_cache/(64*1024*1024)) * 8)
+
+            if nofile_limit[0] < minlimit:
+                resource.setrlimit(resource.RLIMIT_NOFILE, (min(minlimit, nofile_limit[1]), nofile_limit[1]))
+
+            if minlimit > nofile_limit[1]:
+                self.logger.warning("file handles required to meet --file-cache (%s) exceeds hard file handle limit (%s), cache size will be smaller than requested", minlimit, nofile_limit[1])
+
+        except Exception as e:
+            self.logger.warning("unable to adjust file handle limit: %s", e)
+
+        nofile_limit = resource.getrlimit(resource.RLIMIT_NOFILE)
+        self.logger.info("file cache capped at %s bytes or less based on available disk (RLIMIT_NOFILE is %s)", ((nofile_limit[0]//8)*64*1024*1024), nofile_limit)
+
+        try:
             self._setup_api()
             self._setup_mount()
         except Exception as e:
-            self.logger.exception("arv-mount: exception during setup: %s", e)
+            self.logger.exception("exception during setup: %s", e)
             exit(1)
 
     def __enter__(self):
@@ -211,12 +247,32 @@ class Mount(object):
 
     def _setup_api(self):
         try:
+            # default value of file_cache is 0, this tells KeepBlockCache to
+            # choose a default based on whether disk_cache is enabled or not.
+
+            block_cache = arvados.keep.KeepBlockCache(cache_max=self.args.file_cache,
+                                                      disk_cache=self.args.disk_cache,
+                                                      disk_cache_dir=self.args.disk_cache_dir)
+
+            # If there's too many prefetch threads and you
+            # max out the CPU, delivering data to the FUSE
+            # layer actually ends up being slower.
+            # Experimentally, capping 7 threads seems to
+            # be a sweet spot.
+            prefetch_threads = min(max((block_cache.cache_max // (64 * 1024 * 1024)) - 1, 1), 7)
+
             self.api = arvados.safeapi.ThreadSafeApiCache(
                 apiconfig=arvados.config.settings(),
+                api_params={
+                    'num_retries': self.args.retries,
+                },
                 keep_params={
-                    'block_cache': arvados.keep.KeepBlockCache(self.args.file_cache),
+                    'block_cache': block_cache,
+                    'num_prefetch_threads': prefetch_threads,
                     'num_retries': self.args.retries,
-                })
+                },
+                version='v1',
+            )
         except KeyError as e:
             self.logger.error("Missing environment: %s", e)
             exit(1)