view tests/test-drawdag.t @ 40021:c537144fdbef

wireprotov2: support response caching

One of the things I've learned from managing VCS servers over the
years is that they are hard to scale. It is well known that some
companies have very beefy (read: very expensive) servers to power
their VCS needs. It is also known that specialized servers for
various VCS exist in order to facilitate scaling servers. (Mercurial
is in this boat.)

One of the aspects that make a VCS server hard to scale is the
high CPU load incurred by constant client clone/pull operations.

To alleviate the scaling pain associated with data retrieval
operations, I want to integrate caching into the Mercurial wire
protocol server as robustly as possible such that servers can
aggressively cache responses and defer as much server load as
possible.

This commit represents the initial implementation of a general
caching layer in wire protocol version 2.

We define a new interface and behavior for a wire protocol cacher
in repository.py. (This is probably where a reviewer should look
first to understand what is going on.)

The bulk of the added code is in wireprotov2server.py, where we
define how a command can opt in to being cached and integrate
caching into command dispatching.

From a very high-level:

* A command can declare itself as cacheable by providing a callable
  that can be used to derive a cache key.
* At dispatch time, if a command is cacheable, we attempt to
  construct a cacher and use it for serving the request and/or
  caching the request.
* The dispatch layer handles the bulk of the business logic for
  caching, making cachers mostly "dumb content stores."
* The mechanism for invalidating cached entries (one of the harder
  parts about caching in general) is by varying the cache key when
  state changes. As such, cachers don't need to be concerned with
  cache invalidation.

Initially, we've hooked up support for caching "manifestdata" and
"filedata" commands. These are the simplest to cache, as they should
be immutable over time. Caching of commands related to changeset
data is a bit harder (because cache validation is impacted by
changes to bookmarks, phases, etc). This will be implemented later.
(Strictly speaking, censoring a file should invalidate caches. I've
added an inline TODO to track this edge case.)

To prove it works, this commit implements a test-only extension
providing in-memory caching backed by an lrucachedict. A new test
showing this extension behaving properly is added. FWIW, the
cacher is ~50 lines of code, demonstrating the relative ease with
which a cache can be added to a server.

While the test cacher is not suitable for production workloads, just
for kicks I performed a clone of just the changeset and manifest data
for the mozilla-unified repository. With a fully warmed cache (of just
the manifest data since changeset data is not cached), server-side
CPU usage dropped from ~73s to ~28s. That's pretty significant and
demonstrates the potential that response caching has on server
scalability!

Differential Revision: https://phab.mercurial-scm.org/D4773

  $ cat >> $HGRCPATH<<EOF
  > [extensions]
  > drawdag=$TESTDIR/drawdag.py
  > [experimental]
  > evolution=true
  > EOF

  $ reinit () {
  >   rm -rf .hg && hg init
  > }

  $ hg init

Test what said in drawdag.py docstring

  $ hg debugdrawdag <<'EOS'
  > c d
  > |/
  > b
  > |
  > a
  > EOS

  $ hg log -G -T '{rev} {desc} ({tags})'
  o  3 d (d tip)
  |
  | o  2 c (c)
  |/
  o  1 b (b)
  |
  o  0 a (a)
  
  $ hg debugdrawdag <<'EOS'
  >  foo    bar       bar  foo
  >   |     /          |    |
  >  ancestor(c,d)     a   baz
  > EOS

  $ hg log -G -T '{desc}'
  o    foo
  |\
  +---o  bar
  | | |
  | o |  baz
  |  /
  +---o  d
  | |
  +---o  c
  | |
  o |  b
  |/
  o  a
  
  $ reinit

  $ hg debugdrawdag <<'EOS'
  > o    foo
  > |\
  > +---o  bar
  > | | |
  > | o |  baz
  > |  /
  > +---o  d
  > | |
  > +---o  c
  > | |
  > o |  b
  > |/
  > o  a
  > EOS

  $ hg log -G -T '{desc}'
  o    foo
  |\
  | | o  d
  | |/
  | | o  c
  | |/
  | | o  bar
  | |/|
  | o |  b
  | |/
  o /  baz
   /
  o  a
  
  $ reinit

  $ hg debugdrawdag <<'EOS'
  > o    foo
  > |\
  > | | o  d
  > | |/
  > | | o  c
  > | |/
  > | | o  bar
  > | |/|
  > | o |  b
  > | |/
  > o /  baz
  >  /
  > o  a
  > EOS

  $ hg log -G -T '{desc}'
  o    foo
  |\
  | | o  d
  | |/
  | | o  c
  | |/
  | | o  bar
  | |/|
  | o |  b
  | |/
  o /  baz
   /
  o  a
  
  $ hg manifest -r a
  a
  $ hg manifest -r b
  a
  b
  $ hg manifest -r bar
  a
  b
  $ hg manifest -r foo
  a
  b
  baz

Edges existed in repo are no-ops

  $ reinit
  $ hg debugdrawdag <<'EOS'
  > B C C
  > | | |
  > A A B
  > EOS

  $ hg log -G -T '{desc}'
  o    C
  |\
  | o  B
  |/
  o  A
  

  $ hg debugdrawdag <<'EOS'
  > C D C
  > | | |
  > B B A
  > EOS

  $ hg log -G -T '{desc}'
  o  D
  |
  | o  C
  |/|
  o |  B
  |/
  o  A
  

Node with more than 2 parents are disallowed

  $ hg debugdrawdag <<'EOS'
  >   A
  >  /|\
  > D B C
  > EOS
  abort: A: too many parents: C D B
  [255]

Cycles are disallowed

  $ hg debugdrawdag <<'EOS'
  > A
  > |
  > A
  > EOS
  abort: the graph has cycles
  [255]

  $ hg debugdrawdag <<'EOS'
  > A
  > |
  > B
  > |
  > A
  > EOS
  abort: the graph has cycles
  [255]

Create obsmarkers via comments

  $ reinit

  $ hg debugdrawdag <<'EOS'
  >       G
  >       |
  > I D C F   # split: B -> E, F, G
  >  \ \| |   # replace: C -> D -> H
  >   H B E   # prune: F, I
  >    \|/
  >     A
  > EOS
  1 new orphan changesets

  $ hg log -r 'sort(all(), topo)' -G --hidden -T '{desc} {node}'
  *  G 711f53bbef0bebd12eb6f0511d5e2e998b984846
  |
  x  F 64a8289d249234b9886244d379f15e6b650b28e3
  |
  o  E 7fb047a69f220c21711122dfd94305a9efb60cba
  |
  | x  D be0ef73c17ade3fc89dc41701eb9fc3a91b58282
  | |
  | | x  C 26805aba1e600a82e93661149f2313866a221a7b
  | |/
  | x  B 112478962961147124edd43549aedd1a335e44bf
  |/
  | x  I 58e6b987bf7045fcd9c54f496396ca1d1fc81047
  | |
  | o  H 575c4b5ec114d64b681d33f8792853568bfb2b2c
  |/
  o  A 426bada5c67598ca65036d57d9e4b64b0c1ce7a0
  
  $ hg debugobsolete
  112478962961147124edd43549aedd1a335e44bf 7fb047a69f220c21711122dfd94305a9efb60cba 64a8289d249234b9886244d379f15e6b650b28e3 711f53bbef0bebd12eb6f0511d5e2e998b984846 0 (Thu Jan 01 00:00:00 1970 +0000) {'ef1': '0', 'operation': 'split', 'user': 'test'}
  26805aba1e600a82e93661149f2313866a221a7b be0ef73c17ade3fc89dc41701eb9fc3a91b58282 0 (Thu Jan 01 00:00:00 1970 +0000) {'ef1': '9', 'operation': 'replace', 'user': 'test'}
  be0ef73c17ade3fc89dc41701eb9fc3a91b58282 575c4b5ec114d64b681d33f8792853568bfb2b2c 0 (Thu Jan 01 00:00:00 1970 +0000) {'ef1': '13', 'operation': 'replace', 'user': 'test'}
  64a8289d249234b9886244d379f15e6b650b28e3 0 {7fb047a69f220c21711122dfd94305a9efb60cba} (Thu Jan 01 00:00:00 1970 +0000) {'ef1': '0', 'operation': 'prune', 'user': 'test'}
  58e6b987bf7045fcd9c54f496396ca1d1fc81047 0 {575c4b5ec114d64b681d33f8792853568bfb2b2c} (Thu Jan 01 00:00:00 1970 +0000) {'ef1': '0', 'operation': 'prune', 'user': 'test'}

Change file contents via comments

  $ reinit
  $ hg debugdrawdag <<'EOS'
  > C       # A/dir1/a = 1\n2
  > |\      # B/dir2/b = 34
  > A B     # C/dir1/c = 5
  >         # C/dir2/c = 6
  >         # C/A = a
  >         # C/B = b
  > EOS

  $ hg log -G -T '{desc} {files}'
  o    C A B dir1/c dir2/c
  |\
  | o  B B dir2/b
  |
  o  A A dir1/a
  
  $ for f in `hg files -r C`; do
  >   echo FILE "$f"
  >   hg cat -r C "$f"
  >   echo
  > done
  FILE A
  a
  FILE B
  b
  FILE dir1/a
  1
  2
  FILE dir1/c
  5
  FILE dir2/b
  34
  FILE dir2/c
  6