the default encoding of ceph::real_time truncates seconds to uint32_t,
so stores the wrong timestamp for object lock enforcement
Fixes: https://tracker.ceph.com/issues/63537
Signed-off-by: Casey Bodley <cbodley@redhat.com>
the default encodings for chrono types were made to be
backward-compatible with utime_t, so truncated seconds to 32 bits
adds new functions that encode these chrono types using their underlying
representation, which for ceph::real_time and ceph::timespan is
'nanoseconds as uint64_t'
Signed-off-by: Casey Bodley <cbodley@redhat.com>
instead of using a virtual `allocate_refresh_handler()` to create an
object to call the virtual `AsyncRefreshHandler::init_fetch()` on, just
provide one virtual `init_refresh()` to start the operation. this avoids
the need for `AsyncRefreshHandler` altogether
Signed-off-by: Casey Bodley <cbodley@redhat.com>
instead of relying on async_refresh_response/fail to manage ref counts
manually, just stash an intrusive_ptr in AsyncRefreshHandler for the
duration of the async operation
Signed-off-by: Casey Bodley <cbodley@redhat.com>
`rgw::sal::Bucket::read_stats_async()` had returned stats as
`map<RGWObjCategory, RGWStorageStats>`, but the only caller in
rgw_quota.cc just summed up the categories into a single instance
of `RGWStorageStats`. moving this summation up into rgw_rados'
`RGWGetBucketStatsContext` allows us to unify these sal callbacks
around `RGWStorageStats` alone
the `RGWGetUserStats_CB`/`RGWGetBucketStats_CB` callbacks had inherited
from `RefCountedObject` and required manual reference counting. switch
to `boost::intrusive_ptr` for scope-based shared ownership
Signed-off-by: Casey Bodley <cbodley@redhat.com>
the base class doesn't need member variables. the derived class in
rgw_quota.cc accepts stats by const-ref
Signed-off-by: Casey Bodley <cbodley@redhat.com>
Fix warnings printed by the command "flake8 --select=F,E9
--exclude=venv,.tox src/pybind/mgr/volumes/".
While working with mgr/vol, syntax errors are not printed anywhere. Any
attempt to run vstart.sh after such faulty patch causes vstart.sh to go
in a infinite loop. And running "ceph fs volume" command prints "no such
command exists". This doesn't tell the actual issue and causes
confusion.
When flake8 is run, the issue is not immediately apparent due to so many
warnings. Therefore, fix these warnings so that it becomes easier to
spot such critical issues here onwards.
Note: mypy uses comments like "#type: Dict" for type checking and
therefore counts 'from typechecking import Dict' as not unsued. But
flake8 doesn't recognize type hints embedded in comment. Therfore,
switch to actually using type hints instead of adding type hints to
comments.
Signed-off-by: Rishabh Dave <ridave@redhat.com>
Writing guest keyring to CWD's file named "keyring" will over-write
build/keyring on developer's machine which will make the cluster
inoperatable and also fail the test.
Fixes: https://tracker.ceph.com/issues/63506
Signed-off-by: Rishabh Dave <ridave@redhat.com>
in some cases the RGW may publish the notifications before the kafka
consumer started reading the events from the topic
Fixes: https://tracker.ceph.com/issues/62136
Signed-off-by: Yuval Lifshitz <ylifshit@redhat.com>
Format definition lists in the section "Do I need to restart a monitor
to adjust debug levels?" in
doc/rados/troubleshooting/troubleshooting-mon.rst.
Signed-off-by: Zac Dover <zac.dover@proton.me>
some of rocksdb's dependencies may not have built by the time its
ExternalProject starts, so it can fail with missing headers or
libraries. for example, `uring::uring` may itself be an ExternalProject,
and its include directory won't exist until it starts building:
```
[89/1345] Performing configure step for 'rocksdb_ext'
FAILED: src/rocksdb_ext-prefix/src/rocksdb_ext-stamp/rocksdb_ext-configure build/src/rocksdb_ext-prefix/src/rocksdb_ext-stamp/rocksdb_ext-configure
...
CMake Error in CMakeLists.txt:
Imported target "uring::uring" includes non-existent path
"build/src/liburing/src/include"
in its INTERFACE_INCLUDE_DIRECTORIES. Possible reasons include:
* The path was deleted, renamed, or moved to another location.
* An install or uninstall procedure did not complete successfully.
* The installation package was faulty and references files it does not
provide.
...
[91/1345] Performing download step (git clone) for 'liburing_ext'
Cloning into 'liburing'...
```
use `add_dependencies(rocksdb_ext)` to make sure all of its dependencies
are available before starting the build
Signed-off-by: Casey Bodley <cbodley@redhat.com>
It's not obvious where the clone creation is performed. It's especially
obfuscated because we call the function/method for cloning indirectly.
Therefore, add comments where necessary and mark it with heavy rating.
Same goes for all other "subvolume snapshot clone" operations.
Signed-off-by: Rishabh Dave <ridave@redhat.com>
Edit the section "Client can't connect or mount" in the file
doc/rados/troubleshooting/troubleshooting-mon.rst.
Signed-off-by: Zac Dover <zac.dover@proton.me>
The Windows Universal C Runtime (ucrt) "_creat" function is no
longer POSIX compatible and requires Windows specific mode flags.
We got admin socket test failures after switching from msvcrt to
uscrt.
We'll address the issue with some platform checks.
Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>
Windows CLI arguments use either ANSI (main()) or UTF-16 (wmain()).
Meanwhile, Ceph libraries expect UTF-8 and raise exceptions when
trying to use Unicode CLI arguments or log Unicode output:
rbd.exe create test_unicode_șțăâ --size=32M
terminate called after throwing an instance of 'std::runtime_error'
what(): invalid utf8
We'll use a Windows application manifest, setting the "activeCodePage"
property [1][2]. This enables the Windows UCRT UTF-8 mode so that
functions that receive char* arguments will expect UTF-8 instead of ANSI,
including main(). One exception is CreateProcess, which will need the
UTF-16 form (CreateProcessW).
Despite the locale being set to utf-8, we'll have to explicitly set
the console output to utf-8 using SetConsoleOutputCP(CP_UTF8).
In order to use the UTF-8 locale, we'll have to switch the mingw-llvm
runtime from msvcrt to ucrt.
This also fixes ceph-dokan crashes that currently occur when non-ANSI
paths are logged.
[1] https://learn.microsoft.com/en-us/windows/win32/sbscs/application-manifests#activecodepage
[2] https://learn.microsoft.com/en-us/windows/apps/design/globalizing/use-utf8-code-page
Signed-off-by: Lucian Petrut <lpetrut@cloudbasesolutions.com>
Edit the section "Recovering a monitor's broken monmap" in
doc/rados/troubleshooting/troubleshooting-mon.rst.
Co-authored-by: Anthony D'Atri <anthony.datri@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>
[161/715] Building CXX object src/crush/CMakeFiles/crush_objs.dir/CrushTester.cc.o
ceph/src/crush/CrushTester.cc:478:7: warning: variable 'num_devices_active' set but not used [-Wunused-but-set-variable]
int num_devices_active = 0;
^
1 warning generated.
[165/715] Building CXX object src/crush/CMakeFiles/crush_objs.dir/CrushWrapper.cc.o
ceph/src/crush/CrushWrapper.cc:1579:9: warning: variable 'local_changed' set but not used [-Wunused-but-set-variable]
int local_changed = 0;
^
Signed-off-by: Casey Bodley <cbodley@redhat.com>
when called on a versioned object, prepare() may follow olh and look up
a different object instance
but when called on a multipart part, we should not overwrite the
original object name with the part's object name (of the form
mymultipart.2~_XLFNqOW0NuiALg7q4-Hi_7hdtAkZUH.1)
Signed-off-by: Casey Bodley <cbodley@redhat.com>
add an overload to expose the manifest storage to callers of
get_obj_state(). the existing RGWObjState+RGWObjManifest overload
just calls the RGWObjStateManifest one
Signed-off-by: Casey Bodley <cbodley@redhat.com>
