docs: Fix a few typos (DataDog#8963)

- Should read `accessed` rather than `acessed`.
- Should read `throttling` rather than `throtling`.
- Should read `standard` rather than `standars`.
- Should read `occurrences` rather than `occurences`.
- Should read `minimum` rather than `minumum`.
- Should read `inadvertently` rather than `inadvertedly`.
- Should read `foreign` rather than `foreing`.
- Should read `enqueuing` rather than `enqueing`.
- Should read `definitions` rather than `defintions`.

Author: timgates42

Dockerfiles/manifests/README.md

@@ -12,4 +12,4 @@ Several examples are provided:
 * [`cluster-agent-datadogMetrics`](cluster-agent) Contains the agent DaemonSet as well as the cluster agent with DatadogMetric CRD support;
 * [`cluster-checks-runners`](cluster-checks-runners) Contains the agent DaemonSet as well as the cluster agent and the cluster check runners.
 
-**NOTE:** Manifests are generated in the `default` namespace. You will need to modify `namespace: default` occurences if you are installing in another namespace.
+**NOTE:** Manifests are generated in the `default` namespace. You will need to modify `namespace: default` occurrences if you are installing in another namespace.

docs/dev/caveats.md

@@ -9,7 +9,7 @@ This document provides a list of known development caveats
 The COM concurrency model may be set in different ways, it also has to be called for every thread that might indeed interact with the COM library. Furthermore, once a concurrency model is set for a thread, it cannot be changed unless the thread is `CoUnitilialize()`d. This poses an issue for us for a variety of reasons:
 1. Third party libraries like `gopsutil`, which initializes the concurrency model by setting it to the multi-threaded model, will fail in its calls as the model is different.
 2. We also have python integrations that employ the COM library (ie. WMI, SQLserver, ...) that ultimately rely on `pythoncom` for this. `pythoncom`, in fact, initializes the COM library to the single-threaded model by default, but doesn't really care about the concurrency model and will not fail if a different model has been previously set. 
-3. Because the actual *loading* of the integrations will import `pythoncom` the concurrency model might be inadvertedly and implicitly be set to the default (single-threaded) concurrency model meaning that any subsequent call to an affected `gopsutil` function would fail as the concurrency model would already be set. 
+3. Because the actual *loading* of the integrations will import `pythoncom` the concurrency model might be inadvertently and implicitly be set to the default (single-threaded) concurrency model meaning that any subsequent call to an affected `gopsutil` function would fail as the concurrency model would already be set. 
 4. Due to go's concurrency model we can assume nothing about what goroutine is running on what thread at any given time, so it's not trivial to tell what concurrency model a thread's COM library was initialized to. 
 
 

pkg/network/ebpf/c/http.h

@@ -75,7 +75,7 @@ static __always_inline void http_enqueue(http_transaction_t *http) {
     // It seems that support for this type of access range by the verifier was added later on:
     // https://patchwork.ozlabs.org/project/netdev/patch/[email protected]/
     //
-    // What is unfortunate about this is not only that enqueing a HTTP transaction is O(HTTP_BATCH_SIZE),
+    // What is unfortunate about this is not only that enqueuing a HTTP transaction is O(HTTP_BATCH_SIZE),
     // but also that we can't really increase the batch/page size at the moment because that blows up the eBPF *program* size
 #pragma unroll
     for (int i = 0; i < HTTP_BATCH_SIZE; i++) {

pkg/network/ebpf/c/sockfd.h

@@ -5,7 +5,7 @@
 #include <linux/types.h>
 
 // This map is used to to temporarily store function arguments (sockfd) for
-// sockfd_lookup_light function calls, so they can be acessed by the corresponding kretprobe.
+// sockfd_lookup_light function calls, so they can be accessed by the corresponding kretprobe.
 // * Key is the pid_tgid;
 // * Value the socket FD;
 struct bpf_map_def SEC("maps/sockfd_lookup_args") sockfd_lookup_args = {

pkg/network/ebpf/c/tracer-maps.h

@@ -137,7 +137,7 @@ struct bpf_map_def SEC("maps/telemetry") telemetry = {
 
 // This map is used to to temporarily store function arguments (the struct sock*
 // mapped to the given fd_out) for do_sendfile function calls, so they can be
-// acessed by the corresponding kretprobe.
+// accessed by the corresponding kretprobe.
 // * Key is pid_tgid (u64)
 // * Value is (struct sock*)
 struct bpf_map_def SEC("maps/do_sendfile_args") do_sendfile_args = {

pkg/util/retry/README.md

@@ -36,7 +36,7 @@ methods from `retrier.go` on that class:
 - passing the error to `Retry.IsErrWillRetry()` and `Retry.IsErrPermaFail()`
 will tell you whether it's necessary to retry again, or just give up
 initialising this object
-- retry throtling is implemented in case several users try to use a class.
+- retry throttling is implemented in case several users try to use a class.
 Calling `NextRetry()` will tell you when the next retry is possible. Before
 that time, all calls to `TriggerRetry()` will return a `FailWillRetry` error.
 **The retry will not automatically run when that time is reached, you have

rtloader/README.md

@@ -5,7 +5,7 @@ CPython embedding/extending backend with support for multiple Python versions.
 ## Concepts
 
 RtLoader is a C++ wrapper around the CPython API with a C89-compatible public API
-that can be used by foreing languages like Go. In order to provide support for
+that can be used by foreign languages like Go. In order to provide support for
 multiple Python versions, RtLoader fully abstracts Python in order to decouple client
 applications and CPython. Which Python version to use can be decided at runtime,
 RtLoader will `dlopen` the proper backend libraries accordingly.

rtloader/common/builtins/datadog_agent.c

@@ -225,7 +225,7 @@ PyObject *get_config(PyObject *self, PyObject *args)
 }
 
 /*! \fn PyObject *headers(PyObject *self, PyObject *args, PyObject *kwargs)
-    \brief This function provides a standars set of HTTP headers the caller might want to
+    \brief This function provides a standard set of HTTP headers the caller might want to
     use for HTTP requests.
     \param self A PyObject* pointer to the `datadog_agent` module.
     \param args A PyObject* pointer to the `agentConfig`, but not expected to be used.

tasks/go.py

@@ -404,7 +404,7 @@ def generate_licenses(ctx, filename='LICENSE-3rdparty.csv', verbose=False):
 @task
 def generate_protobuf(ctx):
     """
-    Generates protobuf defintions in pkg/proto
+    Generates protobuf definitions in pkg/proto
     """
     base = os.path.dirname(os.path.abspath(__file__))
     repo_root = os.path.abspath(os.path.join(base, ".."))

tools/dep_tree_resolver/README.md

@@ -6,7 +6,6 @@ ones specified) when the code is being run. While no method is perfect at this
 task, we can use this helper script to generate our best-guess at the resulting
 dependency tree.
 
-
 ## Caveats
 
 ### Isolated module dependencies

tools/windows/install-help/cal/customaction.h

@@ -6,7 +6,7 @@
 #define MIN_PASS_LEN 16         /* minimum length of password to generate */
 #define MAX_PASS_LEN 20         /* maximum length of password to generate */
 #define MIN_NUM_LOWER_CHARS 2   /* minimum allowable number of lowercase chars */
-#define MIN_NUM_UPPER_CHARS 2   /* minumum allowable number of uppercase chars */
+#define MIN_NUM_UPPER_CHARS 2   /* minimum allowable number of uppercase chars */
 #define MIN_NUM_NUMBER_CHARS 2  /* minimum allowable number of numeric chars */
 #define MIN_NUM_SPECIAL_CHARS 2 /* minimum number of special characters */
 #include "SID.h"