##// END OF EJS Templates
wireproto: add streams to frame-based protocol...
wireproto: add streams to frame-based protocol Previously, the frame-based protocol was just a series of frames, with each frame associated with a request ID. In order to scale the protocol, we'll want to enable the use of compression. While it is possible to enable compression at the socket/pipe level, this has its disadvantages. The big one is it undermines the point of frames being standalone, atomic units that can be read and written: if you add compression above the framing protocol, you are back to having a stream-based protocol as opposed to something frame-based. So in order to preserve frames, compression needs to occur at the frame payload level. Compressing each frame's payload individually will limit compression ratios because the window size of the compressor will be limited by the max frame size, which is 32-64kb as currently defined. It will also add CPU overhead, as it is more efficient for compressors to operate on fewer, larger blocks of data than more, smaller blocks. So compressing each frame independently is out. This means we need to compress each frame's payload as if it is part of a larger stream. The simplest approach is to have 1 stream per connection. This could certainly work. However, it has disadvantages (documented below). We could also have 1 stream per RPC/command invocation. (This is the model HTTP/2 goes with.) This also has disadvantages. The main disadvantage to one global stream is that it has the very real potential to create CPU bottlenecks doing compression. Networks are only getting faster and the performance of single CPU cores has been relatively flat. Newer compression formats like zstandard offer better CPU cycle efficiency than predecessors like zlib. But it still all too common to saturate your CPU with compression overhead long before you saturate the network pipe. The main disadvantage with streams per request is that you can't reap the benefits of the compression context for multiple requests. For example, if you send 1000 RPC requests (or HTTP/2 requests for that matter), the response to each would have its own compression context. The overall size of the raw responses would be larger because compression contexts wouldn't be able to reference data from another request or response. The approach for streams as implemented in this commit is to support N streams per connection and for streams to potentially span requests and responses. As explained by the added internals docs, this facilitates servers and clients delegating independent streams and compression to independent threads / CPU cores. This helps alleviate the CPU bottleneck of compression. This design also allows compression contexts to be reused across requests/responses. This can result in improved compression ratios and less overhead for compressors and decompressors having to build new contexts. Another feature that was defined was the ability for individual frames within a stream to declare whether that individual frame's payload uses the content encoding (read: compression) defined by the stream. The idea here is that some servers may serve data from a combination of caches and dynamic resolution. Data coming from caches may be pre-compressed. We want to facilitate servers being able to essentially stream bytes from caches to the wire with minimal overhead. Being able to mix and match with frames are compressed within a stream enables these types of advanced server functionality. This commit defines the new streams mechanism. Basic code for supporting streams in frames has been added. But that code is seriously lacking and doesn't fully conform to the defined protocol. For example, we don't close any streams. And support for content encoding within streams is not yet implemented. The change was rather invasive and I didn't think it would be reasonable to implement the entire feature in a single commit. For the record, I would have loved to reuse an existing multiplexing protocol to build the new wire protocol on top of. However, I couldn't find a protocol that offers the performance and scaling characteristics that I desired. Namely, it should support multiple compression contexts to facilitate scaling out to multiple CPU cores and compression contexts should be able to live longer than single RPC requests. HTTP/2 *almost* fits the bill. But the semantics of HTTP message exchange state that streams can only live for a single request-response. We /could/ tunnel on top of HTTP/2 streams and frames with HEADER and DATA frames. But there's no guarantee that HTTP/2 libraries and proxies would allow us to use HTTP/2 streams and frames without the HTTP message exchange semantics defined in RFC 7540 Section 8. Other RPC protocols like gRPC tunnel are built on top of HTTP/2 and thus preserve its semantics of stream per RPC invocation. Even QUIC does this. We could attempt to invent a higher-level stream that spans HTTP/2 streams. But this would be violating HTTP/2 because there is no guarantee that HTTP/2 streams are routed to the same server. The best we can do - which is what this protocol does - is shoehorn all request and response data into a single HTTP message and create streams within. At that point, we've defined a Content-Type in HTTP parlance. It just so happens our media type can also work as a standalone, stream-based protocol, without leaning on HTTP or similar protocol. Differential Revision: https://phab.mercurial-scm.org/D2907

File last commit:

r35104:795bfa2a default
r37304:9bfcbe4f default
Show More
error.py
307 lines | 9.7 KiB | text/x-python | PythonLexer
# error.py - Mercurial exceptions
#
# Copyright 2005-2008 Matt Mackall <mpm@selenic.com>
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
"""Mercurial exceptions.
This allows us to catch exceptions at higher levels without forcing
imports.
"""
from __future__ import absolute_import
# Do not import anything but pycompat here, please
from . import pycompat
def _tobytes(exc):
"""Byte-stringify exception in the same way as BaseException_str()"""
if not exc.args:
return b''
if len(exc.args) == 1:
return pycompat.bytestr(exc.args[0])
return b'(%s)' % b', '.join(b"'%s'" % pycompat.bytestr(a) for a in exc.args)
class Hint(object):
"""Mix-in to provide a hint of an error
This should come first in the inheritance list to consume a hint and
pass remaining arguments to the exception class.
"""
def __init__(self, *args, **kw):
self.hint = kw.pop(r'hint', None)
super(Hint, self).__init__(*args, **kw)
class RevlogError(Hint, Exception):
__bytes__ = _tobytes
class FilteredIndexError(IndexError):
__bytes__ = _tobytes
class LookupError(RevlogError, KeyError):
def __init__(self, name, index, message):
self.name = name
self.index = index
# this can't be called 'message' because at least some installs of
# Python 2.6+ complain about the 'message' property being deprecated
self.lookupmessage = message
if isinstance(name, bytes) and len(name) == 20:
from .node import short
name = short(name)
RevlogError.__init__(self, '%s@%s: %s' % (index, name, message))
def __bytes__(self):
return RevlogError.__bytes__(self)
def __str__(self):
return RevlogError.__str__(self)
class FilteredLookupError(LookupError):
pass
class ManifestLookupError(LookupError):
pass
class CommandError(Exception):
"""Exception raised on errors in parsing the command line."""
__bytes__ = _tobytes
class InterventionRequired(Hint, Exception):
"""Exception raised when a command requires human intervention."""
__bytes__ = _tobytes
class Abort(Hint, Exception):
"""Raised if a command needs to print an error and exit."""
__bytes__ = _tobytes
class HookLoadError(Abort):
"""raised when loading a hook fails, aborting an operation
Exists to allow more specialized catching."""
class HookAbort(Abort):
"""raised when a validation hook fails, aborting an operation
Exists to allow more specialized catching."""
class ConfigError(Abort):
"""Exception raised when parsing config files"""
class UpdateAbort(Abort):
"""Raised when an update is aborted for destination issue"""
class MergeDestAbort(Abort):
"""Raised when an update is aborted for destination issues"""
class NoMergeDestAbort(MergeDestAbort):
"""Raised when an update is aborted because there is nothing to merge"""
class ManyMergeDestAbort(MergeDestAbort):
"""Raised when an update is aborted because destination is ambiguous"""
class ResponseExpected(Abort):
"""Raised when an EOF is received for a prompt"""
def __init__(self):
from .i18n import _
Abort.__init__(self, _('response expected'))
class OutOfBandError(Hint, Exception):
"""Exception raised when a remote repo reports failure"""
__bytes__ = _tobytes
class ParseError(Hint, Exception):
"""Raised when parsing config files and {rev,file}sets (msg[, pos])"""
__bytes__ = _tobytes
class PatchError(Exception):
__bytes__ = _tobytes
class UnknownIdentifier(ParseError):
"""Exception raised when a {rev,file}set references an unknown identifier"""
def __init__(self, function, symbols):
from .i18n import _
ParseError.__init__(self, _("unknown identifier: %s") % function)
self.function = function
self.symbols = symbols
class RepoError(Hint, Exception):
__bytes__ = _tobytes
class RepoLookupError(RepoError):
pass
class FilteredRepoLookupError(RepoLookupError):
pass
class CapabilityError(RepoError):
pass
class RequirementError(RepoError):
"""Exception raised if .hg/requires has an unknown entry."""
class StdioError(IOError):
"""Raised if I/O to stdout or stderr fails"""
def __init__(self, err):
IOError.__init__(self, err.errno, err.strerror)
# no __bytes__() because error message is derived from the standard IOError
class UnsupportedMergeRecords(Abort):
def __init__(self, recordtypes):
from .i18n import _
self.recordtypes = sorted(recordtypes)
s = ' '.join(self.recordtypes)
Abort.__init__(
self, _('unsupported merge state records: %s') % s,
hint=_('see https://mercurial-scm.org/wiki/MergeStateRecords for '
'more information'))
class UnknownVersion(Abort):
"""generic exception for aborting from an encounter with an unknown version
"""
def __init__(self, msg, hint=None, version=None):
self.version = version
super(UnknownVersion, self).__init__(msg, hint=hint)
class LockError(IOError):
def __init__(self, errno, strerror, filename, desc):
IOError.__init__(self, errno, strerror, filename)
self.desc = desc
# no __bytes__() because error message is derived from the standard IOError
class LockHeld(LockError):
def __init__(self, errno, filename, desc, locker):
LockError.__init__(self, errno, 'Lock held', filename, desc)
self.locker = locker
class LockUnavailable(LockError):
pass
# LockError is for errors while acquiring the lock -- this is unrelated
class LockInheritanceContractViolation(RuntimeError):
__bytes__ = _tobytes
class ResponseError(Exception):
"""Raised to print an error with part of output and exit."""
__bytes__ = _tobytes
class UnknownCommand(Exception):
"""Exception raised if command is not in the command table."""
__bytes__ = _tobytes
class AmbiguousCommand(Exception):
"""Exception raised if command shortcut matches more than one command."""
__bytes__ = _tobytes
# derived from KeyboardInterrupt to simplify some breakout code
class SignalInterrupt(KeyboardInterrupt):
"""Exception raised on SIGTERM and SIGHUP."""
class SignatureError(Exception):
__bytes__ = _tobytes
class PushRaced(RuntimeError):
"""An exception raised during unbundling that indicate a push race"""
__bytes__ = _tobytes
class ProgrammingError(Hint, RuntimeError):
"""Raised if a mercurial (core or extension) developer made a mistake"""
__bytes__ = _tobytes
class WdirUnsupported(Exception):
"""An exception which is raised when 'wdir()' is not supported"""
__bytes__ = _tobytes
# bundle2 related errors
class BundleValueError(ValueError):
"""error raised when bundle2 cannot be processed"""
__bytes__ = _tobytes
class BundleUnknownFeatureError(BundleValueError):
def __init__(self, parttype=None, params=(), values=()):
self.parttype = parttype
self.params = params
self.values = values
if self.parttype is None:
msg = 'Stream Parameter'
else:
msg = parttype
entries = self.params
if self.params and self.values:
assert len(self.params) == len(self.values)
entries = []
for idx, par in enumerate(self.params):
val = self.values[idx]
if val is None:
entries.append(val)
else:
entries.append("%s=%r" % (par, val))
if entries:
msg = '%s - %s' % (msg, ', '.join(entries))
ValueError.__init__(self, msg)
class ReadOnlyPartError(RuntimeError):
"""error raised when code tries to alter a part being generated"""
__bytes__ = _tobytes
class PushkeyFailed(Abort):
"""error raised when a pushkey part failed to update a value"""
def __init__(self, partid, namespace=None, key=None, new=None, old=None,
ret=None):
self.partid = partid
self.namespace = namespace
self.key = key
self.new = new
self.old = old
self.ret = ret
# no i18n expected to be processed into a better message
Abort.__init__(self, 'failed to update value for "%s/%s"'
% (namespace, key))
class CensoredNodeError(RevlogError):
"""error raised when content verification fails on a censored node
Also contains the tombstone data substituted for the uncensored data.
"""
def __init__(self, filename, node, tombstone):
from .node import short
RevlogError.__init__(self, '%s:%s' % (filename, short(node)))
self.tombstone = tombstone
class CensoredBaseError(RevlogError):
"""error raised when a delta is rejected because its base is censored
A delta based on a censored revision must be formed as single patch
operation which replaces the entire base with new content. This ensures
the delta may be applied by clones which have not censored the base.
"""
class InvalidBundleSpecification(Exception):
"""error raised when a bundle specification is invalid.
This is used for syntax errors as opposed to support errors.
"""
__bytes__ = _tobytes
class UnsupportedBundleSpecification(Exception):
"""error raised when a bundle specification is not supported."""
__bytes__ = _tobytes
class CorruptedState(Exception):
"""error raised when a command is not able to read its state from file"""
__bytes__ = _tobytes
class PeerTransportError(Abort):
"""Transport-level I/O error when communicating with a peer repo."""
class InMemoryMergeConflictsError(Exception):
"""Exception raised when merge conflicts arose during an in-memory merge."""
__bytes__ = _tobytes