docs: fix broken links

The link checker found various broken links.

The main table of content gets simplified because the top-level entry
didn't actually link to anything.

README.md

@@ -109,13 +109,12 @@ Docker\* Hub. Those images contain license texts under
 
 ## Content
 
-- [PMEM-CSI for Kubernetes](#pmem-csi-for-kubernetes)
-    - [Supported Kubernetes versions](#supported-kubernetes-versions)
-    - [Design and architecture](docs/design.md)
-    - [Installation and Usage](docs/install.md)
-       - [Prerequisites](docs/install.md#prerequisites)
-       - [Installation and setup](docs/install.md#installation-and-setup)
-       - [Filing issues and contributing](docs/install.md#filing-issues-and-contributing)
-    - [Develop and contribute](docs/DEVELOPMENT.md)
-    - [Automated testing](docs/autotest.md)
-    - [Application examples](examples/readme.rst)
+- [Supported Kubernetes versions](#supported-kubernetes-versions)
+- [Design and architecture](docs/design.md)
+- [Installation and Usage](docs/install.md)
+   - [Prerequisites](docs/install.md#prerequisites)
+   - [Installation and setup](docs/install.md#installation-and-setup)
+   - [Filing issues and contributing](docs/install.md#filing-issues-and-contributing)
+- [Develop and contribute](docs/DEVELOPMENT.md)
+- [Automated testing](docs/autotest.md)
+- [Application examples](examples/readme.rst)

conf.json

@@ -18,6 +18,11 @@
         "sphinx_md",
         "sphinx_copybutton"
     ],
+    "linkcheck_anchors_ignore": [
+        "^!",
+        "^L[0-9]+-L[0-9]+$",
+        "^discussion_r[0-9]+$"
+    ],
     "copybutton_prompt_text": "$ ",
     "html_theme": "sphinx_rtd_theme",
     "project": "PMEM-CSI",

docs/DEVELOPMENT.md

@@ -4,7 +4,7 @@
 
 ### Build PMEM-CSI
 
-1.  Use `make build-images` to produce Docker\* container images.
+1.  Use `make build-images` to produce Docker\* container images.g180
 
 2.  Use `make push-images` to push Docker container images to a Docker image
     registry. The default is to push to a local
@@ -177,7 +177,7 @@ Network ports are opened as configured in manifest files:
 
 - metrics endpoint: typical port values 10010 (PMEM-CSI) and 10011
   (external-provisioner)
-- webhook endpoint: disabled by default, port chosen when [enabling the scheduler extensions](../README.md#enable-scheduler-extensions)
+- webhook endpoint: disabled by default, port chosen when [enabling the scheduler extensions](install.md#enable-scheduler-extensions)
 
 ### Local sockets
 
@@ -191,8 +191,8 @@ The Kubernetes CSI API is used over a local socket inside the same host.
 ### Command line arguments
 
 See the `main.go` files of the
-[pmem-csi-driver](./pkg/pmem-csi-driver/main.go) and
-the [pmem-csi-operator](./pkg/pmem-csi-operator/main.go) commands.
+[pmem-csi-driver](/pkg/pmem-csi-driver/main.go) and
+the [pmem-csi-operator](/pkg/pmem-csi-operator/main.go) commands.
 
 ### Environment variables
 
@@ -248,7 +248,7 @@ make test_e2e TEST_E2E_FOCUS=lvm-production.*late.binding.*stress.test
 ```
 
 Alternatively, one can run the
-[`hack/stress-driver.sh`](hack/stress-driver.sh)
+[`hack/stress-driver.sh`](/hack/stress-driver.sh)
 helper script to generate load on the driver
 ```console
 ROUNDS=500 VOL_COUNT=5 ./hack/stress-driver.sh

docs/autotest.md

@@ -82,7 +82,7 @@ the master node.
 
 After `make start`, PMEM-CSI is *not* installed yet. Either install
 manually as [described for a normal
-cluster](#run-pmem-csi-on-kubernetes) or use the
+cluster](install.md#installation-and-setup) or use the
 [setup-deployment.sh](/test/setup-deployment.sh) script.
 
 ## Configuration options

docs/design.md

@@ -278,7 +278,7 @@ volumes](https://kubernetes.io/docs/concepts/storage/ephemeral-volumes/#generic-
 which are an alpha feature in Kubernetes 1.19 and supported by
 PMEM-CSI because they use the normal volume provisioning process.
 
-See [exposing persistent and cache volumes](install.md#expose-persistent-and-cache-volumes-to-applications) for configuration information.
+See [volume parameters](install.md#volume-parameters) for configuration information.
 
 ## Volume Size
 

docs/install.md

@@ -274,8 +274,8 @@ PMEM-CSI operator or by using reference yaml files provided in source code.
 
 #### Install using the operator
 
-[PMEM-CSI operator](./design.md#operator) facilitates deploying and managing
-the [PMEM-CSI driver](https://github.com/intel/pmem-csi) on a Kubernetes cluster.
+The [PMEM-CSI operator](./design.md#pmem-csi-operator) facilitates deploying and managing
+the PMEM-CSI driver on a Kubernetes cluster.
 
 ##### Installing the operator from Operator Hub
 
@@ -942,7 +942,7 @@ This is the original implementation of ephemeral inline volumes for
 CSI drivers in Kubernetes. It is currently available as a beta feature
 in Kubernetes.
 
-Volume requests [embedded in the pod spec](https://kubernetes-csi.github.io/docs/ephemeral-local-volumes.html#example-of-inline-csi-pod-spec) are provisioned as
+Volume requests [embedded in the pod spec with the `csi` field](https://kubernetes-csi.github.io/docs/ephemeral-local-volumes.html) are provisioned as
 ephemeral volumes. The volume request could use below fields as
 [`volumeAttributes`](https://kubernetes.io/docs/concepts/storage/volumes/#csi):
 
@@ -1564,7 +1564,7 @@ deployment is in the `Failed` state, then one can look into the event(s) using
 > has to take extreme care of such deployments by ensuring that not more than
 > one driver ends up running on the same node(s). Nodes on which a PMEM-CSI
 > driver could run can be configured by using `nodeSelector` property of
-> [`DeploymentSpec`](#deployment-crd-api).
+> [`DeploymentSpec`](#pmem-csi-deployment-crd).
 
 #### Operator metrics data
 

examples/gce.md

@@ -8,7 +8,7 @@ form](https://docs.google.com/forms/d/e/1FAIpQLSeX1tN6Qt-aQUK2iVVioClFX5N-061jqO
 
 This document explains how to bring up Kubernetes on those machines
 with the [scripts from
-Kubernetes](https://kubernetes.io/docs/setup/production-environment/turnkey/gce/)
+Kubernetes](https://github.com/kubernetes/kubernetes/blob/master/cluster/kube-up.sh)
 and how to install PMEM-CSI.
 
 It was written for and tested with the v0.5.0 release of PMEM-CSI, but
@@ -157,7 +157,7 @@ $ qemu-system-x86_64 -enable-kvm -machine accel=kvm,usb=off  -nographic \
 As for the `debian-9` image on a GCE VM, this machine will have a `/dev/pmem0`.
 
 Now build a normal image ("base" or "dev") and [create a GCE image
-from it](https://cloud.google.com/container-optimized-os/docs/how-to/building-from-open-source#running_on_compute_engine).
+from it](https://cloud.google.com/container-optimized-os/docs/how-to/building-from-open-source).
 
 ### Starting and stopping Kubernetes
 
@@ -206,7 +206,7 @@ To stop the cluster, use the same env variables for the
 After the previous step, `kubectl` works and is configured to use the
 new cluster. What follows next are the steps explained in more details
 in the top-level README's [Run PMEM-CSI on
-Kubernetes](../docs/install.md#run-pmem-csi-on-kubernetes) section.
+Kubernetes](../docs/install.md#install-pmem-csi-driver) section.
 
 First the worker nodes need to be labeled:
 

examples/memcached.md

@@ -184,7 +184,7 @@ deployment](/deploy/kustomize/memcached/persistent/memcached-persistent.yaml)
 uses a stateful set because that can automatically create persistent
 volumes for each instance. The shell wrapper around memcached
 translates the normal [`SIGTERM` shutdown
-signal](https://kubernetes.io/docs/concepts/workloads/pods/pod/#termination-of-pods)
+signal](https://kubernetes.io/docs/concepts/workloads/pods/pod-lifecycle/#pod-termination)
 into `SIGUSR1`.
 
 Deploying like that becomes less flexible because the memcached pods