Feature branch sync with master #5937
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Adds `with_child_trace` fuction that only creates a span only if there exists a parent. Signed-off-by: Gabriel Buica <danutgabriel.buica@cloud.com>
Intruments the following path of an http request: - `handle_one`, - `callback`, - `callback1`, - `dispatch`, - `jsoncallback`, - `response`. Signed-off-by: Gabriel Buica <danutgabriel.buica@cloud.com>
Sometimes it takes the underlying cluster stack (corosync) some time to return a consistent view of the quorum. For example, it may be that the membership information correctly reflects the new members after a membership change, while the quorum field is still out of date. Add a delay here to make sure that the information from corosync represents a consistent snapshot of the current cluster state. Signed-off-by: Vincent Liu <shuntian.liu2@cloud.com>
Intruments the request reading loop. Adds new functionality to the tracing library to update a span with a new parent. This way we can retroactively set the parent span with the correnct one. Signed-off-by: Gabriel Buica <danutgabriel.buica@cloud.com>
Fixes warning of unused records fields. These fields are part of the opentelementry spec but we are currently not using them in `xapi`. Signed-off-by: Gabriel Buica <danutgabriel.buica@cloud.com>
Increases the default maximum number of spans inside a trace from `1000` to `2500`. Now that we instrumented the internal http calls, with all components enabled, the number of spans inside `xapi` component for a `vm-start` operations is slightly greater than `1000`. This causes spans to be leaked, they are removed from the ongoing span table but never added in the finished tabled. Therefore, they are lost unless the limit is change in `/etc/xapi.conf`. This should fix the issue until we implement a better abstraction for the span hashtables. Signed-off-by: Gabriel Buica <danutgabriel.buica@cloud.com>
….Request` Like the title says, this commit adds `Http.Request.with_tracing` to `Http.Request`. This should enable updating the traceparent field of a request while we process it. Signed-off-by: Gabriel Buica <danutgabriel.buica@cloud.com>
This change augments all parameter structures emitted by the JSON backend to include a "required" member that specifies whether the parameter is optional or not. Parameters are considered optional in the IDL if a default value is provided for them. The simplest example is session.login_with_password. This message permits two extra parameters, "version" and "originator" - but these are not required to successfully invoke the function. Signed-off-by: Colin James <colin.barr@cloud.com>
…/CP-50444 CP-50444: Instrument http svr with dt
This has ended up in our public PyPI package description: https://pypi.org/project/XenAPI/ However it is irrelevant to a user of the package on PyPI, because they'd already get the "built" version, that has the version number inside, so they don't need OCaml or 'make' to use it. Signed-off-by: Edwin Török <edwin.torok@cloud.com>
Output if parameter is required in JSON backend
…ysdk Python SDK: Move "Packaging" section out of the public docs
Signed-off-by: Colin James <colin.barr@cloud.com>
```
File "ocaml/libs/clock/dune", line 27, characters 19-33:
27 | (names test_date test_timer_run)
^^^^^^^^^^^^^^ qcheck random seed: 423397317 Testing `Timer'. This run has ID `D009HP55'. �[<35;18;6M
F.............
┌──────────────────────────────────────────────────────────────────────────────┐
│ [FAIL] Timer 0 Timer.remaining. │
└──────────────────────────────────────────────────────────────────────────────┘
Sleeping for 0.150000 seconds...
Sleeping for 0.000500 seconds...
test `Timer.remaining` failed on ≥ 1 cases:
1ms
Expected to have spare time, but got excess: 1.91μs. Duration: 1ms, actual: 999μs, timer:
elapsed: 1.03ms
duration: 1ms
```
Here we asked for a sleep of 0.5ms, but got woken up twice as late.
And the quickcheck property was expecting that the actual wake up time won't be twice as wrong.
In reality it can wake up by arbitrary amounts of time later based on how busy the OS is,
but we do need to check that the wakeups are not completely wrong.
So skip the check on very short durations, but keep it on the longer 100ms and 300ms tests.
Signed-off-by: Edwin Török <edwin.torok@cloud.com>
Allow remediation commits for DCO
jemalloc was updated to version 5.3.0: - Provides libjemalloc.so.2 instead of libjemalloc.so.1 - No longer supported `lg_dirty_mult` option Signed-off-by: Stephen Cheng <stephen.cheng@cloud.com>
…persist tmp file after reboot - Modified xapi_globs.ml to include this change. Signed-off-by: Ashwinh <ashwin.h@cloud.com>
On behalf of Colin <colin.barr@cloud.com>, I, Colin James <colin.barr@cloud.com>, hereby add my Signed-off-by to this commit: 8337fa9 Signed-off-by: Colin James <colin.barr@cloud.com>
Retroactively sign off 8337fa9
Ensure that the updated data and metadata are written before the headers are updated otherwise xcp-rrdd might start reading the data block before all the data is populated and thus run off the end of the data. Signed-off-by: Mark Syms <mark.syms@cloud.com>
…tephenche/CA-397171 CA-397171: Replace libjemalloc.so.1 with libjemalloc.so.2
…leepy CI: fix spurious failure on busy system
…A-392685 CA-392685:/tmp is mounted with type tmpfs which cause the files to be cleaned after host reboot in XS rawhide.
…oro-wait CA-396635: Wait for corosync to update its info
In the commit 62db5cb, the device name validation was consolidated with the function which converts name to device number. Particularly, the function "Device_number.of_string" is used for both the validation and the conversion. But one issue was introduced in the validation is that the "None" value returned from "Device_number.of_string" is considered as valid. This causes the error "the device name is invalid". This commit just fixes this issue. Signed-off-by: Ming Lu <ming.lu@cloud.com>
Signed-off-by: Ming Lu <ming.lu@cloud.com>
|
Current CI check because new python code in master does not have necessary unittest, |
lindig
approved these changes
Aug 14, 2024
Vincent-lau
approved these changes
Aug 14, 2024
Correct, opened #5938 to test (and thereby cover) the merged changes for the sync of Please just close and re-open this PR (#5937) after #5938 is merged to |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.