Increase the safety factor to 2.5x for extra safety if max of measured
timestamp resolution and measured minimal racy threshold is < 100ms, use
1.25 otherwise since for large filesystem resolution values the
influence of finite resolution of the system clock should be negligible.
Before, not yet using the newly introduced minRacyThreshold measurement,
the threshold was 1.1x FS resolution, and we could issue the
following sequence of events,
start
create-file
read-file (currentTime)
end
which had the following timestamps:
create-file 1564589081998
start 1564589082002
read 1564589082003
end 1564589082004
In this case, the difference between create-file and read is 5ms,
which exceeded the 4ms FS resolution, even though the events together
took just 2ms of runtime.
Reproduce with:
bazel test --runs_per_test=100 \
//org.eclipse.jgit.test:org_eclipse_jgit_internal_storage_file_FileSnapshotTest
The file system timestamp resolution is 4ms in this case.
This code assumes that the kernel and the JVM use the same clock that
is synchronized with the file system clock. This seems plausible,
given the resolution of System.currentTimeMillis() and the latency for
a gettimeofday system call (typically ~1us), but it would be good to
justify this with specifications.
Also cover a source of flakiness: if the test runs under extreme load,
then we could have
start
create-file
<long delay>
read
end
which would register as an unmodified file. Avoid this by skipping the
test if end-start is too big.
[msohn]:
- downported from master to stable-5.1
- skip test if resolution is below 10ms
- adjust safety factor to 1.25 for resolutions above 100ms
Change-Id: I87d2cf035e01c44b7ba8364c410a860aa8e312ef
Signed-off-by: Han-Wen Nienhuys <hanwen@google.com>
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Cache FileStoreAttributeCache entries since looking up FileStore for a
file may be expensive on some platforms.
Implement a simple LRU cache based on ConcurrentHashMap using a simple
long counter to order access to cache entries.
Change-Id: I4881fa938ad2f17712c05da857838073a2fc4ddb
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Signed-off-by: Marc Strapetz <marc.strapetz@syntevo.com>
Also-By: Marc Strapetz <marc.strapetz@syntevo.com>
To enable persisting the minimal racy threshold per FileStore add a
new config option to the user global git configuration:
- Config section is "filesystem"
- Config subsection is concatenation of
- Java vendor (system property "java.vendor")
- Java version (system property "java.version")
- FileStore's name, on Windows we use the attribute volume:vsn instead
since the name is not necessarily unique.
- separated by '|'
e.g.
"AdoptOpenJDK|1.8.0_212-b03|/dev/disk1s1"
The same prefix is used as for filesystem timestamp resolution, so
both values are stored in the same config section
- The config key for minmal racy threshold is "minRacyThreshold" as a
time value, supported time units are those supported by
DefaultTypedConfigGetter#getTimeUnit
- measure for 3 seconds to limit runtime which depends on hardware, OS
and Java version being used
If the minimal racy threshold is configured for a given FileStore the
configured value is used instead of measuring it.
When the minimal racy threshold was measured it is persisted in the user
global git configuration.
Rename FileStoreAttributeCache to FileStoreAttributes since this class
is now declared public in order to enable exposing all attributes in one
object.
Example:
[filesystem "AdoptOpenJDK|11.0.3|/dev/disk1s1"]
timestampResolution = 7000 nanoseconds
minRacyThreshold = 3440 microseconds
Change-Id: I22195e488453aae8d011b0a8e3276fe3d99deaea
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Also-By: Marc Strapetz <marc.strapetz@syntevo.com>
By running FileSnapshotTest#detectFileModified we found that the sum of
measured filesystem timestamp resolution and measured clock resolution
may yield a too small interval after a file has been modified which we
need to consider racily clean. In our tests we didn't find this behavior
on all systems we tested on, e.g. on MacOS using APFS and Java 8 and 11
this effect was not observed.
On Linux (SLES 15, kernel 4.12.14-150.22-default) we collected the
following test results using Java 8 and 11:
In 23-98% of 10000 test runs (depending on filesystem type and Java
version) the test failed, which means the effective interval which needs
to be considered racily clean after a file was modified is larger than
the measured file timestamp resolution.
"delta" is the observed interval after a file has been modified but
FileSnapshot did not yet detect the modification:
"resolution" is the measured sum of file timestamp resolution and clock
resolution seen in Java.
Java version filesystem failures resolution min delta max delta
1.8.0_212-b04 btrfs 98.6% 1 ms 3.6 ms 6.6 ms
1.8.0_212-b04 ext4 82.6% 3 ms 1.1 ms 4.1 ms
1.8.0_212-b04 xfs 23.8% 4 ms 3.7 ms 3.9 ms
1.8.0_212-b04 zfs 23.1% 3 ms 4.8 ms 5.0 ms
11.0.3+7 btrfs 98.1% 3 us 0.7 ms 4.7 ms
11.0.3+7 ext4 98.1% 6 us 0.7 ms 4.7 ms
11.0.3+7 xfs 98.5% 7 us 0.1 ms 8.0 ms
11.0.3+7 zfs 98.4% 7 us 0.7 ms 5.2 ms
Mac OS
1.8.0_212 APFS 0% 1 s
11.0.3+7 APFS 0% 6 us
The observed delta is not distributed according to a normal gaussian
distribution but rather random in the observed range between "min delta"
and "max delta".
Run this test after measuring file timestamp resolution in
FS.FileAttributeCache to auto-configure JGit since it's unclear what
mechanism is causing this effect.
In FileSnapshot#isRacyClean use the maximum of the measured timestamp
resolution and the measured "delta" as explained above to decide if a
given FileSnapshot is to be considered racily clean. Add a 30% safety
margin to ensure we are on the safe side.
Change-Id: I1c8bb59f6486f174b7bbdc63072777ddbe06694d
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Repeat the test 10000 times to get statistics if measured
fsTimestampResolution is working in practice to detect racy git
situations.
Add a class to compute statistics for this test. Log delta between
lastModified and time when FileSnapshot failed to detect modification.
This happens if the racy git limit determined by measuring filesystem
timestamp resolution and clock resolution is too small. If it would be
correct FileSnapshot would always detect modification or mark it
modified if time since modification is smaller than the racy git limit.
Change-Id: Iabe7af1a7211ca58480f8902d4fa4e366932fc77
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
This should help to detect if measured fsTimeResolution is too small.
Change-Id: Id1f54dbdedb52b17859904e47776fa3a5887b8be
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
When filesystem timestamp resolution is very high some tests don't work
since runtime of the test setup is too long to reach a racily clean
FileSnapshot. Hence skip these tests when timestamp resolution is higher
than 10 millisecond.
Change-Id: Ie47dd10eda22037b5c1ebff6b6becce0654ea807
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
This helps to avoid some time critical tests can't prepare the test
fixture intended since measuring timestamp resolution takes time.
Change-Id: Ib34023e682a106070ca97e98ef16789a4dfb97b4
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
- use Path instead of File
- create test directories, files and output stream using Files methods
- delete unused list "files"
Change-Id: I8c5c601eca9f613efb5618d33b262277df92a06a
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
This enables higher file timestamp resolution on filesystems like ext4,
Mac APFS (1ns) or NTFS (100ns) providing high timestamp resolution on
filesystem level.
Note:
- on some OSes Java 8,9 truncate milliseconds, see
https://bugs.openjdk.java.net/browse/JDK-8177809, fixed in Java 10
- UnixFileAttributes truncates timestamp resolution to microseconds when
converting the internal representation to FileTime exposed in the API,
see https://bugs.openjdk.java.net/browse/JDK-8181493
- WindowsFileAttributes also provides only microsecond resolution
Change-Id: I25ffff31a3c6f725fc345d4ddc2f26da3b88f6f2
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Checking lastModified is time critical hence debug trace is the only way
to analyze issues since debugging is impractical.
Also add configuration for buffering of log4j output to reduce runtime
impact when debug trace is on. Limit buffer to 1MiB and comment this
configuration out since we may not always want to use buffering.
Change-Id: Ib1a0537b67c8dc3fac994a77b42badd974ce6c97
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
EolRepositoryTest and GcCommitSelectionTest timed out frequently when
running unit tests using bazel with the default timeout "moderate"
(300s). Increase timeout of these tests to "long" (900s).
Change-Id: I43588cf950f55b50f868d9fe9c66d22bd428a54c
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
This change is fixing confusing name warning: [1].
./org.eclipse.jgit.test/tests.bzl:12: confusing-name:
Never use 'l', 'I', or 'O' as names (they're too easily confused
with 'I', 'l', or '0').
And is also fixing: "All calls to rules or macros should pass arguments
by keyword position argument" warning: [2].
./org.eclipse.jgit.test/BUILD:42: positional-args: All calls to rules
or macros should pass arguments by keyword (arg_name=value) syntax.
[1] https://github.com/bazelbuild/buildtools/blob/master/WARNINGS.md#confusing-name
[2] https://github.com/bazelbuild/buildtools/blob/master/WARNINGS.md#positional-args
Change-Id: If5c28ec8a1ddc1d1b1035bd07b838a2a564aea4f
Signed-off-by: David Ostrovsky <david@ostrovsky.org>
Dependencies on commons-compress, xz, and jgit-archive are required
for the build to succeed.
Change-Id: I42f3721078a240ad93b8dcab909e66b9bfff0b56
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
FileTreeIteratorWithTimeControl was deleted in a024759, but was
not removed from the BUILD file, thus causing the bazel build to
fail.
Change-Id: I892c0ffcac947298d0d6009374ee2c5d9afefb66
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
The only usage of this test iterator was removed in df637928d. Hence
delete this iterator and associated test.
Change-Id: I47710133ec3edc675c21db210960c024982668c6
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Add a unittest.
In commit I5485db55 ("Fix FileSnapshot's consideration of file size"),
the special casing of UNKNOWN_SIZE was forgotten.
This change, together with I493f3b57b ("Measure file timestamp
resolution used in FileSnapshot") introduced a regression that would
occasionally surface in Gerrit integration tests marked UseLocalDisk,
with the symptom that creating the Admin user in NoteDb failed with a
LOCK_FAILURE.
Signed-off-by: Han-Wen Nienhuys <hanwen@google.com>
Change-Id: I7ffd972581f815c144f810481103c7985af5feb0
This test case assumed file system timestamp resolution of 1 second. On
filesystems with a finer resolution this test fails since the index
entry is only smudged if the file index entry's lastModified and the
lastModified of the git index itself are within the same filesystem
timer tick. Fix this by ensuring that these timestamps are identical
which should work for any filesystem timer resolution.
Bug: 548188
Change-Id: Id84d59e1cfeb48fa008f8f27f2f892c4f73985de
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Signed-off-by: Christian Halstrick <christian.halstrick@sap.com>
By using File#setLastModified, we can create a racy git situation
stably.
Tested with --runs_per_test=100
Bug: 526111
Change-Id: I60b3632d353e19f335668325aa603640be423f58
Signed-off-by: Masaya Suzuki <masayasuzuki@google.com>
Archives created by the ArchiveCommand didn't produce deterministic
archive hashes. For RevCommits RevWalk.parseTree returns the root tree
instead of the RevCommit hence retrieving the commit's timestamp didn't
work. Instead use RevWalk.parseAny and extract the tree manually.
Archive entries store timestamps with 1 second resolution hence we need
to wait longer when creating the same archive twice and compare archive
hashes. Otherwise hash comparison in tests wouldn't fail without this
patch.
Bug: 548312
Change-Id: I437d515de51cf68265584d28a8446cebe6341b79
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
update Maven plugins
- jacoco-maven-plugin to 0.8.4
- japicmp-maven-plugin to 0.14.1
- maven-compiler-plugin to 3.8.1
- maven-deploy-plugin to 3.0.0-M1
- maven-enforcer-plugin to 3.0.0-M2
- maven-install-plugin to 3.0.0-M1
- maven-jar-plugin to 3.1.2
- maven-javadoc-plugin to 3.1.0
- maven-jxr-plugin to 3.0.0
- maven-pmd-plugin to 3.12.0
- maven-resources-plugin to 3.1.0
- maven-shade-plugin to 3.2.1
- maven-source-plugin to 3.1.0
- maven-surefire-plugin to 3.0.0-M3
- spotbugs-maven-plugin to 3.1.12
- tycho to 1.3.0
- tycho-pack200a-plugin to 1.3.0
- tycho-pack200b-plugin to 1.3.0
Cleanup Maven warnings
- pin version of all used Maven plugins
- remove deprecated way to declare minimum Maven version
Change-Id: If23e2e2bb03e5e1e7b1eb9d4924a8faa0aa3704e
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
As reported by Error Prone:
An inner class should be static unless it references members of its
enclosing class. An inner class that is made non-static unnecessarily
uses more memory and does not make the intent of the class clear.
See https://errorprone.info/bugpattern/ClassCanBeStatic
Change-Id: Ib99d120532630dba63cf400cc1c61c318286fc41
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
(cherry picked from commit ee40efcea4)
In a delete-modify conflict with the deletion as "ours" there may be
no stage 2 in the index. Add appropriate null checks. Add a new test
for this case, and verify that the file gets added with a single LF
after conflict resolution with core.autocrlf=true. This matches the
behavior of canonical git for this case.
Bug: 547724
Change-Id: I1bafdb83d9b78bf85294c78325e818e72fae53bc
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
The only remaining code where the return value is ignored is in tests.
Update them to store the value and perform a basic assertion.
Change-Id: I29ef5bd5dd0648aac3490f9e47ecc74544109652
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
EolRepositoryTest and GcCommitSelectionTest timed out frequently when
running unit tests using bazel with the default timeout "moderate"
(300s). Increase timeout of these tests to "long" (900s).
Change-Id: I43588cf950f55b50f868d9fe9c66d22bd428a54c
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
A merging pull on an unborn branch was already supported. But a
rebasing pull failed. If the user has pull.rebase = true in his
user config, the pull would try to rebase. Rebasing needs a parent
commit, though. Native git handles this case:
git init
git remote add origin <URI>
git pull --rebase origin master
Check up front in PullCommand for the unborn head and just do a
checkout in this case. MergeCommand already has similar code.
Bug: 544965
Change-Id: I1277e1ac0b0364b4623fd791f3d6b07bd5f58fca
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Canonical git treats CR-LF in config files as LF.[1][2] JGit does so,
too, except when escaped as a line continuation. Correct this and
treat the sequence \-CR-LF as a line continuation.
[1] https://github.com/git/git/commit/db2c075d9
[2] https://github.com/git/git/blob/v2.21.0/config.c#L485
Bug: 545850
Change-Id: I51e7378a22c21b3baa3701163c423d04c900af5a
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Otherwise tags may fail to match if their name contains slashes.
Canonical git also uses its wildcard matcher in glob mode.[1]
[1] https://github.com/git/git/blob/v2.21.0/builtin/describe.c#L182
Bug: 546703
Change-Id: I122c7959974fa1fc6a53dfc65837e4314a8badd4
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Commit messages of length 1 were not read. 'lineEnd' is the offset
of the last character in the line before the terminating LF or CR-LF,
and 'nextSpace' is actually the offset of the character _after_ the
next space. With a one-character commit message, nextSpace == lineEnd.
The code also assumes the commit message to be optional, but actually
failed in that case because it read beyond the line ending. Fix that,
too.
Add a test case for reading a todo file.
Bug: 546245
Change-Id: I368d63615930ea2398a6230e756442fd88870654
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Replace "size() > 0" with "!isEmpty()" where appropriate.
In the Status implementation we can drop the check; the subsequent
loop will only execute when the list is non-empty anyway.
Change-Id: I355aff551a603373e702a9d44304f087b476263c
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
The bazel build fails due to NetscapeCookieFileTest's internal class not
being visible to TransportHttpTest.
Split the file out to its own class in the util package, so it's visible
to both.
Change-Id: I69236026eecb9d08a9a66e51752a80ea522b0c6a
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
The git config entries "http.cookieFile" and
"http.saveCookies" are correctly evaluated.
Bug: 488572
Change-Id: Icfeeea95e1a5bac3fa4438849d4ac2306d7d5562
Signed-off-by: Konrad Windszus <konrad_w@gmx.de>
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Test that JGit detects that packfiles have changed even if they are
repacked multiple times in one tick of the filesystem timer.
Test that this detection works also when repacking doesn't change the
length or the filekey of the packfile. In this case where a modified
file can't be detected by looking at file metadata JGit should still
detect too fast modification by racy git checks and trigger rescanning
the pack list and consequently rereading of packfile content.
Change-Id: I67682cfb807c58afc6de9375224ff7489d6618fb
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
If the attributes of FileSnapshot don't detect modification of a
packfile read the packfile's checksum and compare it against the
checksum cached in the loaded packfile.
Since reading the checksum needs less IO than reloading the complete
packfile this may help to reduce the overhead to detect modficiation
when a gc completes while ObjectDirectory scans for packfiles in another
thread.
Bug: 546891
Change-Id: I9811b497eb11b8a85ae689081dc5d949ca8c4be5
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
PacketLineIn.END is only referenced in tests. Replace most of those
with a new package visible end() method.
Remaining usages of PacketLineIn.END are in the form:
while ((line = pckIn.readString()) != PacketLineIn.END) {
and are not trivial replacements, hence are not touched in this change.
Change-Id: Id77c5321ddcad127130b246bde8f08736e60e1ea
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
Deprecate DELIM with the intention of making it private in a future
release.
Callers that want to test if a packet line string is the delimiter
should use the isDelimiter(String) method.
The only other references to DELIM in the JGit code are in tests. For
those, introduce a package visible delimiter() method.
Change-Id: I21e8bbac0ffb9ef710c9753e23435416b09a4891
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
Replace reference comparisons of PacketLineIn's DELIM and END strings
with usage of the helper methods isDelimiter() and isEnd().
Change-Id: I52dcfc4ee9097f1bd6970601c716701847d9eebd
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
Han-Wen Nienhuys
Matthias Sohn
David Ostrovsky
David Pursehouse
Masaya Suzuki
Thomas Wolf
Konrad Windszus
Christian Halstrick