| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
| |
* Fix formatting of stat pages
|
| | |
|
| | |
|
| |
|
|
|
|
| |
mode (#745)
* only check file size for requested range in asio TransmitFileAsync in debug mode
* only do Transmit file for whole files
|
| |\
| |
| | |
Run Windows service as an interactive session, so we correctly respond to RequestApplicationExit when requested to stop by the service manager
|
| | |
| |
| |
| |
| |
| | |
respond to RequestApplicationExit when requested to stop by the service manager"
This reverts commit b380c1bac4b738f49cd412697e0a051402e6d171.
|
| | |
| |
| |
| | |
non-interactive sessions
|
| | |
| |
| |
| | |
RequestApplicationExit when requested to stop by the service manager
|
| | |
| |
| |
| | |
* replace http router AddPattern with AddMatcher
* fix scrub logging
|
| | |
| |
| |
| |
| | |
* implemented chunking for TransmitFile path, to ensure payloads exceeding the TransmitFile API limit of 4GB can be handled
* also fixes similar issue with memory path
|
| |\ \
| | |
| | | |
Restrict filesystem reads in snapshot to paths under project root
|
| | | | |
|
| | | |
| | |
| | |
| | | |
project root
|
| | | |
| | |
| | |
| | | |
out of loop
|
| | |/ |
|
| | |
| |
| | |
* reduce held locks while performing scrub operation
|
| | | |
|
| | | |
|
| | |
| |
| |
| |
| | |
* reduce maximum size per chunk to read to reduce disk contention
* increase timeout before warning on slow shut down of zenserver
* reduce default window size for blockstore chunk iteration
|
| | |
| |
| | |
* don't do full cb-object validation on cache records when read from disk
|
| | | |
|
| | |
| |
| |
| | |
avoid integer vs float conversion issues (#735)
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
- Feature: `zen builds download` now supports `--download-spec-path` to determine what content to download from a build
- The unstructured format expects one line per file relative to the root with '/' as a path delimiter
- The structured format uses JSon format and the `--download-spec-path` must have extension `.json` to enable structured input
{
"parts": {
"default" : {
"files": [
"foo/bar",
"baz.exe"
]
},
"symbols": {
"files": [
"baz.pdb"
]
}
}
}
|
| | |
| |
| |
| |
| |
| | |
* add ability to override scheduling mode in ParallelWork
* Don't increase buffering size when copying from local cache with --boost-worker-memory enabled
* Rework scheduling writes of downloaded data to reduce memory usage
|
| | |
| |
| |
| | |
- Feature: Added `--chunking-cache-path` option to `zen builds upload` and `zen builds diff`
- Path to cache for chunking information of scanned files. Default is empty resulting in no caching
|
| | | |
|
| | |
| |
| | |
* Fix incorrect oplog navigation symbols
|
| | |
| |
| |
| | |
* make sure we properly convert command line args for zenserver as well
* make sure we *add* wildcards/excludes in addition to defaults
|
| | |
| |
| |
| |
| | |
Initial implementation of zenserver "hub" mode. This is an experimental feature.
zenserver can be started in hub mode by specifying `hub` as the first argument to zenserver
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Feature: `zen builds upload` now support structure manifest input for `--manifest-path` when the path has a `.json` extension
- The structured manifest supports splitting a build into multiple parts
{
"parts": {
"default" : {
"partId": "f939f3939939fff3f3202", # optional - used to control the id of each part
"files": [
"foo/bar",
"baz.exe"
]
},
"symbols": {
"files": [
"baz.pdb"
]
}
}
}
|
| |
|
|
|
|
|
| |
* add system metrics output to top command
* removed unnecessary xmake directives
* file system API/comment tweaks
* fixed out-of-range access in httpserver test
* updated ZenServer base API to allow customization by mode
|
| |
|
|
|
|
|
| |
This refactor aims to improve the `ZenServerProcess` classes by making them useful for managing child zenserver instances in more scenarios than just automated tests. This involves changing some functions to not talk about "test directory" and instead use "data directory" etc
As a consequence of the API changes, some tests have changed accordingly.
The code includes som reference to the "hub" mode but there is not yet any other code using this mode, it's just included in this PR to simplify future merges.
|
| |
|
| |
initially we had ZENCORE_API macros to potentially allow for DLL linkage. It turns out that this is not useful and the macros just contribute noise, so this change removes them completely.
|
| |
|
|
|
|
|
| |
This PR brings over some changes made to avoid performing setup for otel instrumentation if we are not sending otel information anywhere anyway.
It also adds the ability to configure an OTLP endpoint on the command line using `--otlp-endpoint=<URI>`.
Bear in mind that OTLP support is still not officially supported so this should not be used in production at this stage.
|
| |
|
|
|
|
|
|
| |
* this adds a consul package which can be used to fetch a consul binary
* it also adds a `ConsulProcess` helper which can be used to spawn and manage a consul service instance
* zencore dependencies brought across:
- `except_fmt.h` for easer generation of formatted exception messages
- `process.h/cpp` changes (adds `Kill` operation and process group support on Windows)
- `string.h` changes to allow generic use of `WideToUtf8()`
|
| |\
| |
| | |
Implement final changes required for daemon mode on Mac
|
| | | |
|
| | | |
|
| | | |
|
| | |
| |
| |
| | |
* throw error on bad oidctoken-exe-path
|
| | |
| |
| |
| |
| |
| | |
string (#712)
* set utf8 locale, only set LC_CTYPE as we don't want to affect how decimal point or sorting orders are handled
* set language/region explicitly so we can use LC_ALL
|
| | | |
|
| | |
| |
| |
| | |
* make ResolveBuildStore respect Verbose flag
* add structured output to zen builds ls command
|
| | |
| |
| |
| |
| |
| |
| | |
This change primarily introduces improved logic for dealing with sending data from file references.
This is intended to reduce the amount of memory-mapping we end up doing when sending data from files. Windows now uses `TransmitFile` to send file data more efficiently using kernel-side I/O, but Linux/Mac basically behaves as before since they don't offer any true async file I/O support via asio. This should be implemented separately using a background I/O thread pool.
This PR also includes improved memory management for http/asio with reduced allocation counts, and a fix for a potential use-after-free in very high load scenarios.
|
| | |
| |
| |
| |
| | |
* remove unreferenced local in projectstore_cmd
* fix minor atomic memory issue in RotatingFileSink
|
| | |
| |
| |
| |
| |
| |
| |
| | |
commands (#706)
* added `--exclude-folders` to `zen upload`, `zen download` and `zen diff`
added `--exclude-extensions` to `zen upload` and `zen diff`
excluded folder names are now matched by folder name in subfolders in addition to root level folders
* allow multiple token separators
|
| | | |
|
| | |
| |
| |
| | |
sending error to Sentry (#705)
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
- Improvement: Validate chunk hashes when dechunking files in oplog import
- Improvement: Use stream decompression when dechunking files
- Improvement: When assembling blocks for oplog export, make sure we keep under/at block size limit
- Improvement: Make cancelling of oplog import more responsive
- Improvement: Use decompress to composite to avoid allocating a new memory buffer for uncompressed chunks during oplog import
- Improvement: Reduce memory buffer size and allocate it on demand when writing multiple chunks to block store
- Improvement: Reduce lock contention when fetching/checking existence of chunks in block store
|
| | | |
|