hgwebdir: avoid systematic full garbage collection
Forcing a systematic full garbage collection upon each request
can serioulsy harm performance. This is reported as
https://bz.mercurial-scm.org/show_bug.cgi?id=6075
With this change we're performing the full collection according
to a new setting, `experimental.web.full-garbage-collection-rate`.
The default value is 1, which doesn't change the behavior and will
allow us to test on real use cases. If the value is 0, no full garbage
collection occurs.
Regardless of the value of the setting, a partial garbage collection
still occurs upon each request (not attempting to collect objects from
the oldest generation). This should be enough to take care of
reference cycles that have been created by the last request
(assessment of this requires changing the setting, not to be 1).
In my experience chasing memory leaks in Mercurial servers,
the full collection never reclaimed any memory, but this is with
Python 3 and biased towards small repositories.
On the other hand, as explained in the Python developer docs [1],
frequent full collections are very harmful in terms of performance if
lots of objects survive the collection, and hence stay in the
oldest generation. Note that `gc.collect()` is indeed trying to
collect the oldest generation [2]. This happens usually in two cases:
- unwanted lingering objects (i.e., an actual memory leak that
the GC cannot do anything about). Sadly, we have lots of those
these days.
- desireable long-term objects, typically in caches (not inner caches
carried by repositories, which should be collected with them). This
is a subject of interest for the Heptapod project.
In short, the flat rate that this change still permits is
probably a bad idea in most cases, and the default value can
be tweaked later on (or even be set to 0) according to experiments
in the wild.
The test is inspired from test-hgwebdir-paths.py
[1] https://devguide.python.org/garbage_collector/#collecting-the-oldest-generation
[2] https://docs.python.org/3/library/gc.html#gc.collect
Differential Revision: https://phab.mercurial-scm.org/D11204
$ cat <<EOF >> $HGRCPATH
> [ui]
> interactive = true
> EOF
$ hg init repo
$ cd repo
$ echo foo > foo
$ hg ci -qAm 'add foo'
$ echo >> foo
$ hg ci -m 'change foo'
$ hg up -qC 0
$ echo bar > bar
$ hg ci -qAm 'add bar'
$ hg log
changeset: 2:effea6de0384
tag: tip
parent: 0:bbd179dfa0a7
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: add bar
changeset: 1:ed1b79f46b9a
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: change foo
changeset: 0:bbd179dfa0a7
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: add foo
$ cd ..
don't show "(+1 heads)" message when pulling closed head
$ hg clone -q repo repo2
$ hg clone -q repo2 repo3
$ cd repo2
$ hg up -q 0
$ echo hello >> foo
$ hg ci -mx1
created new head
$ hg ci -mx2 --close-branch
$ cd ../repo3
$ hg heads -q --closed
2:effea6de0384
1:ed1b79f46b9a
$ hg pull --confirm << EOF
> n
> EOF
pulling from $TESTTMP/repo2
searching for changes
adding changesets
adding manifests
adding file changes
adding 2 changesets with 1 changes to 1 files
new changesets 8c900227dd5d:00cfe9073916
accept incoming changes (yn)? n
transaction abort!
rollback completed
abort: user aborted
[255]
$ hg pull --config pull.confirm=true << EOF
> n
> EOF
pulling from $TESTTMP/repo2
searching for changes
adding changesets
adding manifests
adding file changes
adding 2 changesets with 1 changes to 1 files
new changesets 8c900227dd5d:00cfe9073916
accept incoming changes (yn)? n
transaction abort!
rollback completed
abort: user aborted
[255]
$ hg pull --confirm << EOF
> y
> EOF
pulling from $TESTTMP/repo2
searching for changes
adding changesets
adding manifests
adding file changes
adding 2 changesets with 1 changes to 1 files
new changesets 8c900227dd5d:00cfe9073916
accept incoming changes (yn)? y
added 2 changesets with 1 changes to 1 files
new changesets 8c900227dd5d:00cfe9073916
(run 'hg update' to get a working copy)
$ hg heads -q --closed
4:00cfe9073916
2:effea6de0384
1:ed1b79f46b9a
pull--confirm config option should be ignored if HGPLAIN is set
$ HGPLAIN=1 hg pull --config pull.confirm=True
pulling from $TESTTMP/repo2
searching for changes
no changes found
$ cd ..
$ hg init copy
$ cd copy
Pull a missing revision:
$ hg pull -qr missing ../repo
abort: unknown revision 'missing'
[255]
Pull multiple revisions with update:
$ hg pull -qu -r 0 -r 1 ../repo
$ hg -q parents
0:bbd179dfa0a7
$ hg rollback
repository tip rolled back to revision -1 (undo pull)
working directory now based on revision -1
$ hg pull -qr 0 ../repo
$ hg log
changeset: 0:bbd179dfa0a7
tag: tip
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: add foo
$ hg pull -qr 1 ../repo
$ hg log
changeset: 1:ed1b79f46b9a
tag: tip
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: change foo
changeset: 0:bbd179dfa0a7
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: add foo
This used to abort: received changelog group is empty:
$ hg pull -qr 1 ../repo
Test race condition with -r and -U (issue4707)
We pull '-U -r <name>' and the name change right after/during the changegroup emission.
We use http because http is better is our racy-est option.
$ echo babar > ../repo/jungle
$ cat <<EOF > ../repo/.hg/hgrc
> [hooks]
> outgoing.makecommit = hg ci -Am 'racy commit'; echo committed in pull-race
> EOF
$ hg serve -R ../repo -p $HGPORT2 -d --pid-file=../repo.pid
$ cat ../repo.pid >> $DAEMON_PIDS
$ hg pull --rev default --update http://localhost:$HGPORT2/
pulling from http://localhost:$HGPORT2/
searching for changes
adding changesets
adding manifests
adding file changes
added 1 changesets with 1 changes to 1 files (+1 heads)
new changesets effea6de0384
2 files updated, 0 files merged, 0 files removed, 0 files unresolved
$ hg log -G
@ changeset: 2:effea6de0384
| tag: tip
| parent: 0:bbd179dfa0a7
| user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| summary: add bar
|
| o changeset: 1:ed1b79f46b9a
|/ user: test
| date: Thu Jan 01 00:00:00 1970 +0000
| summary: change foo
|
o changeset: 0:bbd179dfa0a7
user: test
date: Thu Jan 01 00:00:00 1970 +0000
summary: add foo
$ cd ..
$ killdaemons.py