profile
viewpoint
Guillaume Rose guillaumerose Red Hat Paris Software engineer

guillaumerose/bind-http-json 11

HTTP backend for Bind9

guillaumerose/bind-mongodb 7

MongoDB for Bind9

guillaumerose/dhcp-http-json 6

HTTP backend for DHCP

guillaumerose/beaglebone-tools 3

Quelques outils pour démarrer avec une BeagleBone

guillaumerose/dhcp-mongodb 3

MongoDB for DHCP

guillaumerose/commutateurs 2

A simple gem to manage network switches

guillaumerose/api-browser 1

Browse any API

guillaumerose/arduino-bootloader 1

For Arduino Diecimila

PR opened code-ready/snc

WIP: move cvo override between `create manifest` and `create cluster`

It works but I'm not sure this is the right path to follow.

Pros:

  • Unwanted operators never run
  • It's what we aim with cluster profile

Cons:

  • Still leave broken contents (ns with services but not the operator, etc.)
  • CRDs created by operators are not present (PrometheusRule and ServiceMonitor for example)
  • It doesn't speed up snc
  • Override list is hard to maintain.
+274 -47

0 comment

3 changed files

pr created time in 11 hours

push eventguillaumerose/snc

Praveen Kumar

commit sha 9088e0e0a37904df34515dba2265ecb48c10e9f0

Use threads option for xz compression Using the threads option for xz we can reduce the bundle creation time significantly. This will also help us to stop timeout for CI. ``` $ time xz --keep --threads=0 crc.qcow2 real 4m36.255s user 54m17.698s sys 0m3.416s $ time xz --keep crc.qcow2 real 34m48.959s user 34m46.237s sys 0m2.710s $ time xz --threads=0 crc_libvirt_4.4.8.tar real 15m8.081s user 107m18.482s sys 0m18.774s ```

view details

Guillaume Rose

commit sha 14e2f4e3ba91f3214d8f315ba3bad30d72d2608f

Add OWNERS

view details

Guillaume Rose

commit sha 73f9a076bc285c0eb58f1715d8cfabbbb143058e

Use `oc scale` instead of `oc patch` for scaling down ingress controller According to the docs, it's the correct way to scale reduce the number of routers. `oc patch` still left a broken pod, `oc scale` correctly remove it. See https://github.com/openshift/cluster-ingress-operator#managing.

view details

Guillaume Rose

commit sha 41262978ec0634d5eebdd0565d572d51ddd99ea2

Update cvo overrides before the cluster starts

view details

Guillaume Rose

commit sha 5eaa77019a7081c715b20b0bd56a9e931babba21

Dont delete machine-config-operator. This operator is responsible of the tagging of the master node.

view details

Guillaume Rose

commit sha 03b1ec6da3df347bc9ed0812cd3a87522561d13c

Disable monitoring

view details

Guillaume Rose

commit sha a741d77d7df6dd3bcdbf51fef393e0c111a1ae0a

Re-enable machine-api operator

view details

Guillaume Rose

commit sha 5cb62031af836142164ed3fee2176d449f1c500e

Disable some cluster operators

view details

Guillaume Rose

commit sha 021ec0f47d441f558c4977d17b24737c99ed9c8d

Remove unecessary cleanup code

view details

push time in 11 hours

issue commentcode-ready/snc

Installer stuck: netplugin failed with no error message

I'm using 4.5.0-rc.6

guillaumerose

comment created time in 16 hours

push eventcode-ready/snc

Guillaume Rose

commit sha 73f9a076bc285c0eb58f1715d8cfabbbb143058e

Use `oc scale` instead of `oc patch` for scaling down ingress controller According to the docs, it's the correct way to scale reduce the number of routers. `oc patch` still left a broken pod, `oc scale` correctly remove it. See https://github.com/openshift/cluster-ingress-operator#managing.

view details

push time in 17 hours

PR merged code-ready/snc

Use `oc scale` instead of `oc patch` for scaling down ingress controller lgtm

According to the docs, it's the correct way to scale reduce the number of routers. See https://github.com/openshift/cluster-ingress-operator#managing.

For me, oc patch still left a broken pod, oc scale correctly remove it.

+1 -1

2 comments

1 changed file

guillaumerose

pr closed time in 17 hours

issue commentcode-ready/snc

Installer stuck: netplugin failed with no error message

In the ovs pod in openshift-sdn namespace:

2020-07-06T10:58:05.863Z|00024|jsonrpc|WARN|unix#1080: send error: Broken pipe
2020-07-06T10:58:05.863Z|00025|reconnect|WARN|unix#1080: connection dropped (Broken pipe)
2020-07-06T11:01:25.520Z|00217|bridge|WARN|could not open network device veth86d28bf4 (No such device)
2020-07-06T11:01:25.544Z|00218|bridge|WARN|could not open network device vethc7bbd048 (No such device)
2020-07-06T11:01:25.571Z|00219|bridge|WARN|could not open network device vethea3a4197 (No such device)
2020-07-06T11:01:25.600Z|00220|bridge|WARN|could not open network device veth136237a3 (No such device)
2020-07-06T11:01:25.627Z|00221|bridge|WARN|could not open network device veth5b07b5c5 (No such device)
2020-07-06T11:01:25.657Z|00222|bridge|WARN|could not open network device vethaec2ac98 (No such device)
2020-07-06T11:01:25.687Z|00223|bridge|WARN|could not open network device vethbdbb5504 (No such device)
guillaumerose

comment created time in 18 hours

issue commentcode-ready/snc

Installer stuck: netplugin failed with no error message

Events:
  Type     Reason                  Age        From                         Message
  ----     ------                  ----       ----                         -------
  Normal   Scheduled               <unknown>  default-scheduler            Successfully assigned openshift-marketplace/marketplace-operator-7d858b7747-7fc48 to crc-p4sm2-master-0
  Warning  FailedCreatePodSandBox  16m        kubelet, crc-p4sm2-master-0  Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create pod network sandbox k8s_marketplace-operator-7d858b7747-7fc48_openshift-marketplace_54657dca-e0c1-4d1e-a991-4b726b7042c7_0(cccb1387ee48090b9bb2b9ca9879642cb09ffbc46a1e552157c2bbb2e09deb83): netplugin failed with no error message
  Warning  FailedCreatePodSandBox  13m        kubelet, crc-p4sm2-master-0  Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create pod network sandbox k8s_marketplace-operator-7d858b7747-7fc48_openshift-marketplace_54657dca-e0c1-4d1e-a991-4b726b7042c7_0(d7ecbb199c93c47aac5d82fb6e382a3a533dad9d2cc628d62eda9e6fa66d592f): netplugin failed with no error message
  Warning  FailedCreatePodSandBox  9m49s      kubelet, crc-p4sm2-master-0  Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create pod network sandbox k8s_marketplace-operator-7d858b7747-7fc48_openshift-marketplace_54657dca-e0c1-4d1e-a991-4b726b7042c7_0(2232c7d9706a5a4abb4771c934e8245507929ec2d7e51fb50be6c9be87570c0a): netplugin failed with no error message
  Warning  FailedCreatePodSandBox  6m38s      kubelet, crc-p4sm2-master-0  Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create pod network sandbox k8s_marketplace-operator-7d858b7747-7fc48_openshift-marketplace_54657dca-e0c1-4d1e-a991-4b726b7042c7_0(629c6a96a2373d61f990a0a2d7b3787d99fd210246601ae68ed10cdad35c6d7a): netplugin failed with no error message
  Warning  FailedCreatePodSandBox  4m3s       kubelet, crc-p4sm2-master-0  Failed to create pod sandbox: rpc error: code = Unknown desc = failed to create pod network sandbox k8s_marketplace-operator-7d858b7747-7fc48_openshift-marketplace_54657dca-e0c1-4d1e-a991-4b726b7042c7_0(1aa6a2c681df4b6c45ca5ae523042a0d6258e509a02b2d0d0443103e5c3184c5): netplugin failed with no error message
  Normal   AddedInterface          3m46s      multus                       Add eth0 [10.128.0.94/23]
  Normal   Pulling                 3m45s      kubelet, crc-p4sm2-master-0  Pulling image "quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:643978bd7cfc6056aad9c0c8be01962987d85324ff3275a6baa1263a1a4bcf26"
  Normal   Pulled                  2m50s      kubelet, crc-p4sm2-master-0  Successfully pulled image "quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:643978bd7cfc6056aad9c0c8be01962987d85324ff3275a6baa1263a1a4bcf26"
  Normal   Created                 2m49s      kubelet, crc-p4sm2-master-0  Created container marketplace-operator
  Normal   Started                 2m49s      kubelet, crc-p4sm2-master-0  Started container marketplace-operator

About 13min stuck.

guillaumerose

comment created time in 18 hours

Pull request review commentcode-ready/snc

Add OWNERS

+approvers:+  - cfergeau+  - praveenkumar+  - anjannath+  - guillaumerose+  - gbraad

I took the example from https://github.com/tektoncd/cli/blob/master/OWNERS. It seems to work fine for them.

guillaumerose

comment created time in 19 hours

push eventguillaumerose/snc

Guillaume Rose

commit sha 666255f612540ef13fce5b1c2fd8d60d06404e18

Dont delete machine-config-operator. This operator is responsible of the tagging of the master node.

view details

Guillaume Rose

commit sha 8f71ba70c9d091aede46daf2ab3847b70b916bbb

Disable monitoring

view details

Guillaume Rose

commit sha 84f6560c2eb3f08b5146de2399b7542923597d6a

Re-enable machine-api operator

view details

Guillaume Rose

commit sha 90b7129a4a27aa1acd2bfd961216fac1e24055ae

Disable some cluster operators

view details

push time in 19 hours

issue openedcode-ready/snc

Installer stuck: netplugin failed with no error message

Sometimes when running snc, the cluster doesn't progress anymore and around 5-10 containers are waiting in the state "ContainerCreating".

Looking at logs, they are all waiting for a network.

crc-8tfgf-master-0 crio[1425]: time="2020-07-06 07:41:06.471577450Z" level=error msg="Error while adding pod to CNI network \"multus-cni-network\": netplugin failed with no error message"

I had to restart all pods in openshift-sdn namespace to unlock everything. Sadly, these containers don't report any errors in their logs.

created time in 21 hours

create barnchguillaumerose/snc

branch : cvoover

created branch time in 4 days

PR opened code-ready/snc

Use `oc scale` instead of `oc patch` for scaling down ingress controller

According to the docs, it's the correct way to scale reduce the number of routers. See https://github.com/openshift/cluster-ingress-operator#managing.

For me, oc patch still left a broken pod, oc scale correctly remove it.

+1 -1

0 comment

1 changed file

pr created time in 4 days

create barnchguillaumerose/snc

branch : scalepatch

created branch time in 4 days

Pull request review commentcode-ready/snc

Misc changes to accommodate Ben's changes so bundle can be used as is.

 metadata:   name: crc networking:   clusterNetwork:-  - cidr: 10.128.0.0/14+  - cidr: 10.116.0.0/14     hostPrefix: 23   machineNetwork:   - cidr: 192.168.126.0/24   networkType: OpenShiftSDN   serviceNetwork:-  - 172.30.0.0/16+  - 172.25.0.0/16 platform:

we could choose a single IP block (large enough) and say explicitly in the documentation that this IP block is reserved for crc. Inside this block, we can do what we want.

praveenkumar

comment created time in 4 days

PR opened code-ready/snc

Add OWNERS

Required to please the openshift-ci bot.

+6 -0

0 comment

1 changed file

pr created time in 4 days

create barnchguillaumerose/snc

branch : owners

created branch time in 4 days

PR opened code-ready/snc

Pin master to openshift release candidate

master was pointing to 4.4.x. It's now 4.5.x.

+1 -1

0 comment

1 changed file

pr created time in 4 days

create barnchguillaumerose/snc

branch : pin45

created branch time in 4 days

PR opened code-ready/snc

Pin branch 4.4 to openshift stable-4.4
+1 -1

0 comment

1 changed file

pr created time in 4 days

push eventcode-ready/snc

Guillaume Rose

commit sha ac70a9d2b1354eac3803e61b0d5a96ed14fbdcf7

Add oc and yq to gitignore

view details

Guillaume Rose

commit sha bd0cd8f7220753294ad0d03b0fbca64b27a68a56

Delete temporary file after use

view details

push time in 4 days

create barnchguillaumerose/snc

branch : pin44

created branch time in 5 days

push eventguillaumerose/snc

Guillaume Rose

commit sha 612fd420c169c05aa569b132571e669080048535

Delete temporary file after use

view details

push time in 5 days

Pull request review commentcode-ready/snc

Ensure git status is clean after running ./snc.sh

 function create_pvs() { function renew_certificates() {     # Get the cli image from release payload and update it to bootstrap-cred-manager resource     cli_image=$(${OC} adm release -a ${OPENSHIFT_PULL_SECRET_PATH} info ${OPENSHIFT_INSTALL_RELEASE_IMAGE_OVERRIDE} --image-for=cli)-    ${YQ} write kubelet-bootstrap-cred-manager-ds.yaml.in spec.template.spec.containers[0].image ${cli_image} >kubelet-bootstrap-cred-manager-ds.yaml+    ${YQ} write kubelet-bootstrap-cred-manager-ds.yaml.in spec.template.spec.containers[0].image ${cli_image} > kubelet-bootstrap-cred-manager-ds.yaml

yes you're right!

guillaumerose

comment created time in 5 days

PR opened code-ready/snc

Ensure git status is clean after running ./snc.sh
+4 -1

0 comment

2 changed files

pr created time in 5 days

create barnchguillaumerose/snc

branch : gitignore

created branch time in 5 days

push eventguillaumerose/cluster-version-operator

Vadim Rutkovsky

commit sha 17d47a098e2eb9ddfe65eb5b302c82ff8e7a3cc6

hack/build-go: Use OS_GIT_VERSION if available This would update the build script, so that OCP could be built using a custom semver string from OS_GIT_VERSION instead of `git describe` output

view details

Jack Ottofaro

commit sha dadbf9e0df59b7e73ecfc9b8ed609c77695b6c3d

Add an alert to CVO that fires when cluster is not upgradeable Added alert ClusterNotUpgradeable that fires when metric cluster_operator_conditions{name="version"} reports condition Upgradeable as false for 60 minutes or more. Alert reports the reason cluster is not upgradeable and console URL to get more information.

view details

OpenShift Merge Robot

commit sha 62ff8e3e51cd4bfe6cd3b8cb6400a28ada1b183d

Merge pull request #354 from vrutkovs/nonsemver-override Bug 1826349: hack/build-go: Use OS_GIT_VERSION if available

view details

W. Trevor King

commit sha 6044c9c85627d3e2c4529f27afbcde6be2c0a3f7

pkg/verify: Refactor signature fetching only use a Store interface This consolidates around abstract signature retrieval, removing the split between stores and locations which we grew in 9bbf366e1c (verify: Refactor the verify package to have common code, 2019-12-09, #279). This will make it easier to make future changes like parallel HTTP(S) signature retrieval retrieval [1]. [1]: https://bugzilla.redhat.com/show_bug.cgi?id=1840343

view details

W. Trevor King

commit sha a068ee368a11cebdf5c7a5af62c1e3e702fee708

pkg/verify: Parallelize HTTP(S) signature retrieval Fixing [1,2]: $ grep 'unable to load signature' cvo.log I0526 13:13:36.123153 1 verify.go:404] unable to load signature: Get https://storage.googleapis.com/openshift-release/official/signatures/openshift/release/sha256=baa687f29b0ac155d8f4c6914056d36d68f343feb9c1e82b46eef95819d00be5/signature-1: dial tcp 172.217.4.48:443: connect: connection timed out I0526 13:15:47.195128 1 verify.go:404] unable to load signature: Get https://storage.googleapis.com/openshift-release/official/signatures/openshift/release/sha256=baa687f29b0ac155d8f4c6914056d36d68f343feb9c1e82b46eef95819d00be5/signature-2: dial tcp 172.217.9.80:443: connect: connection timed out I0526 13:17:10.718027 1 verify.go:404] unable to load signature: Get https://storage.googleapis.com/openshift-release/official/signatures/openshift/release/sha256=baa687f29b0ac155d8f4c6914056d36d68f343feb9c1e82b46eef95819d00be5/signature-3: context deadline exceeded I0526 13:19:44.764143 1 verify.go:404] unable to load signature: Get https://storage.googleapis.com/openshift-release/official/signatures/openshift/release/sha256=baa687f29b0ac155d8f4c6914056d36d68f343feb9c1e82b46eef95819d00be5/signature-1: dial tcp 172.217.4.48:443: connect: connection timed out I0526 13:21:55.835063 1 verify.go:404] unable to load signature: Get https://storage.googleapis.com/openshift-release/official/signatures/openshift/release/sha256=baa687f29b0ac155d8f4c6914056d36d68f343feb9c1e82b46eef95819d00be5/signature-2: dial tcp 172.217.4.48:443: connect: connection timed out I0526 13:23:18.233801 1 verify.go:404] unable to load signature: Get https://storage.googleapis.com/openshift-release/official/signatures/openshift/release/sha256=baa687f29b0ac155d8f4c6914056d36d68f343feb9c1e82b46eef95819d00be5/signature-3: context deadline exceeded In this case, that was because a restricted network blocked access to storage.googleapis.com, but you'd get similar behavior if storage.googleapis.com itself was slow. We want to hit other signature sources (like our mirrors [3]) before giving up on signatures entirely, and with this commit one HTTP(S) store no longer blocks another. The local ConfigMap store still comes first with a serial store, because we don't want to involve external stores before we've exhausted that local-cluster store. [1]: https://bugzilla.redhat.com/show_bug.cgi?id=1840343 [2]: https://bugzilla.redhat.com/show_bug.cgi?id=1838497#c10 [3]: https://github.com/openshift/cluster-update-keys/blob/cca4ce696383e70ae669e770bd63265a9540b721/manifests.rhel/0000_90_cluster-update-keys_configmap.yaml#L5

view details

OpenShift Merge Robot

commit sha 692a8a37b52c2bc00c45fb84622ba2d5ed3125bb

Merge pull request #393 from wking/parallel-http-signature-stores Bug 1840343: pkg/verify: Parallelize HTTP(S) signature retrieval

view details

OpenShift Merge Robot

commit sha 9b265475d94190387845145adf6bbb0e2a699a7e

Merge pull request #384 from jottofar/ota-189-alert-upgradeable-false Add an alert to CVO that fires when cluster is not upgradeable

view details

Guillaume Rose

commit sha 93eeecf01ad2550d3f3f20a2d4ccf1bab6c731b7

Use profile from configmap

view details

push time in 5 days

push eventguillaumerose/cluster-version-operator

Guillaume Rose

commit sha ba6a21ab3ce916a86516c2e243cd11dd6fae7314

Use profile from configmap

view details

push time in 5 days

push eventcode-ready/crc

Guillaume Rose

commit sha 3ab66412b5067834d13874e2ea80090f5576e3a3

Bump clicumber

view details

push time in 5 days

PR merged code-ready/crc

Reviewers
Bump clicumber approved lgtm

It adds stdout and stderr in the error message when a command exits with the wrong exit code.

+6 -6

2 comments

4 changed files

guillaumerose

pr closed time in 5 days

PR opened code-ready/crc

Bump clicumber

It adds stdout and stderr in the error message when a command exits with the wrong exit code.

+6 -6

0 comment

4 changed files

pr created time in 5 days

create barnchguillaumerose/crc

branch : clicumber

created branch time in 5 days

pull request commentcode-ready/snc

Misc changes to accommodate Ben's changes so bundle can be used as is.

It might worth a release note when released with crc.

/lgtm

praveenkumar

comment created time in 5 days

pull request commentcode-ready/machine-driver-libvirt

Upgrade dependencies

I updated the PR. The scope is smaller without the change of version of libvirt now.

guillaumerose

comment created time in 5 days

push eventguillaumerose/machine-driver-libvirt

Guillaume Rose

commit sha a4c0bf2fd2e0cffb67bfdfa0f9d81609aa792d52

Prepare for next release. Bump version.

view details

Guillaume Rose

commit sha 398d53289b533d089f825d6d209c2f039f2eea70

Bump dependencies and base images

view details

push time in 5 days

PR closed openshift/installer

Reviewers
WIP: Add a ConfigMap with cluster profile do-not-merge/hold do-not-merge/work-in-progress

This PR is a spike for https://github.com/openshift/enhancements/pull/200

install-config.yaml has a new parameter: clusterProfile. It adds a ConfigMap within the openshift-config namespace with the current profile.

+119 -29

4 comments

10 changed files

guillaumerose

pr closed time in 5 days

pull request commentopenshift/installer

WIP: Add a ConfigMap with cluster profile

Thanks for the feedback. I will close this PR for now to remove the noise.

Sorry for the inconvenience.

guillaumerose

comment created time in 5 days

push eventcode-ready/crc

Guillaume Rose

commit sha 4dda6faf8e8753de99635c2bbeeb56ffbf1643b6

Add testdata for xz format

view details

Guillaume Rose

commit sha 3e6fb618a53af30b7f476640e55b46cd4fd07055

untar: call defer file.Close() after each file deferred calls are only called when the current function ends, which means untar() is currently holding all files open until it has finished processing the whole archive. By introducing `untarFile()` and `unzipFile()` helpers, we can ensure each file is closed as soon as it's fully uncompressed.

view details

Guillaume Rose

commit sha 336b399d88688f2997317e23880cd7c5e863c51a

Remove unused method

view details

Guillaume Rose

commit sha e42127f4fb6ccda01cc511c05adc6d64bddbaaf6

Add progress bar for bundle extraction The progress bar is displayed only for files larger than 100MB.

view details

push time in 5 days

PR merged code-ready/crc

Reviewers
Add progress bar for bundle extraction approved lgtm

This PR adds multiple things:

  • testdata for the xz format
  • a fix for a defer file.Close() in a loop
  • a progress bar for bundle extraction (in a terminal only)

Output:

INFO Checking file permissions for /etc/resolver/testing
INFO Extracting bundle: crc_hyperkit_4.4.5.crcbundle ...
crc_hyperkit_4.4.5/kubeadmin-password: 23 B / 23 B [-----------------------------------------------------------------------------------------] 100.00% ? p/s 0s
crc_hyperkit_4.4.5/kubeconfig: 11.55 KiB / 11.55 KiB [---------------------------------------------------------------------------------------] 100.00% ? p/s 0s
crc_hyperkit_4.4.5/id_rsa_crc: 2.53 KiB / 2.53 KiB [-----------------------------------------------------------------------------------------] 100.00% ? p/s 0s
crc_hyperkit_4.4.5/crc.qcow2: 490.22 MiB / 9.15 GiB [--->_______________________________________________________________________] 5.23% 34.99 MiB p/s ETA 4m13s

Tell me what you think!

+51156 -37068

11 comments

280 changed files

guillaumerose

pr closed time in 5 days

push eventguillaumerose/machine-driver-libvirt

Guillaume Rose

commit sha 88aef573829804ffea6427caf26799e64130753b

Bump dependencies and base images

view details

Guillaume Rose

commit sha a826ea9e8fe6669a9afa0525015197c5d8b8fa9b

Prepare for next release. Bump version.

view details

push time in 5 days

Pull request review commentcode-ready/machine-driver-libvirt

Upgrade dependencies

 module github.com/code-ready/machine-driver-libvirt  require (-	github.com/code-ready/machine v0.0.0-20190904103148-0fde37b5d95b-	github.com/libvirt/libvirt-go v3.4.0+incompatible+	github.com/code-ready/machine v0.0.0-20191122132905-c31e0b90623d+	github.com/libvirt/libvirt-go v6.4.0+incompatible

Thanks! I will stay on 3.4.

guillaumerose

comment created time in 5 days

pull request commentcode-ready/crc

Add progress bar for bundle extraction

Any idea why prow doesn't merge the PR ? https://github.com/kubernetes/community/blob/master/contributors/guide/owners.md#the-code-review-process It should do it automatically.

guillaumerose

comment created time in 5 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 4697a0be38d93d63666a7b115b255f2a490cc031

Add progress bar for bundle extraction The progress bar is displayed only for files larger than 100MB.

view details

push time in 6 days

PR opened code-ready/machine-driver-libvirt

Upgrade dependencies

Tested with latest crc. It works fine.

+14 -19

0 comment

6 changed files

pr created time in 6 days

push eventguillaumerose/machine-driver-libvirt

Guillaume Rose

commit sha 9b081a083ee94b631c68ed53ee82ac559009fd80

Bump dependencies and base images

view details

Guillaume Rose

commit sha a03deff4a60aaaacf143e8fb4b95f966bebaebae

Prepare for next release. Bump version.

view details

push time in 6 days

create barnchguillaumerose/machine-driver-libvirt

branch : upgrade

created branch time in 6 days

Pull request review commentcode-ready/crc

Add progress bar for bundle extraction

 func unzipFile(path string, file *zip.File) error { 	} 	defer targetFile.Close() +	var reader io.Reader+	reader = fileReader+	if showProgress && file.FileInfo().Size() >= minSizeForProgressBar {+		bar := pb.Simple.Start64(file.FileInfo().Size())+		bar.Set("prefix", fmt.Sprintf("%s: ", filepath.Base(file.Name)))+		defer bar.Finish()+		reader = bar.NewProxyReader(fileReader)+	}

Good catch! I refactored a bit more and now I have https://github.com/code-ready/crc/pull/1343/files#diff-106940100fecd5f362aaeae069041830R198 The two functions are very similar but calling one from the other makes me introduce more complexity than required. WDYT ?

guillaumerose

comment created time in 6 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 757b0e2591ab02dc852fc8bc3cb9e23b82fcda8d

Add progress bar for bundle extraction The progress bar is displayed only for files larger than 100MB.

view details

push time in 6 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 7ae98747273550281c3b3cafe2530be8ebd0d8c0

Add progress bar for bundle extraction The progress bar is displayed only for files larger than 100MB.

view details

push time in 6 days

fork guillaumerose/machine-driver-libvirt

Machine driver to interact wth KVM using libvirt for use with CodeReady Containers

fork in 6 days

PR closed code-ready/crc

Reviewers
Use os.Args[0] instead of crc in usage hints approved

As a developer, I often run a custom version of crc. This PR helps to have the correct eval string.

$ crc oc-env
export PATH="/home/guillaumerose/.crc/bin/oc:$PATH"
# Run this command to configure your shell:
# eval $(crc oc-env)

$ ./out/linux-amd64/crc oc-env
export PATH="/home/guillaumerose/.crc/bin/oc:$PATH"
# Run this command to configure your shell:
# eval $(./out/linux-amd64/crc oc-env)

$ cd out/linux-amd64 
$ ./crc oc-env
export PATH="/home/guillaumerose/.crc/bin/oc:$PATH"
# Run this command to configure your shell:
# eval $(./crc oc-env)

+6 -2

3 comments

2 changed files

guillaumerose

pr closed time in 6 days

pull request commentcode-ready/crc

Use os.Args[0] instead of crc in usage hints

For Windows, it's maybe better I suppose. It will certainly print crc.exe.

However, I can't find a good and easy way to escape args0. Only 1-person-maintainer libraries.

It doesn't bring anything for the user, closing it.

guillaumerose

comment created time in 6 days

pull request commentcode-ready/crc

Add progress bar for bundle extraction

I tried to add a max width to the progress bar but it messes up with carriage return when 2 progress bars are shown. I tried to add newline, etc. but it's not pretty. I'm back with a fullscreen progress bar.

guillaumerose

comment created time in 6 days

push eventguillaumerose/crc

Guillaume Rose

commit sha b83ee9dc0f66ddb2a13a182642c83dff80d1ac37

Add progress bar for bundle extraction The progress bar is displayed only for files larger than 100MB.

view details

push time in 6 days

Pull request review commentcode-ready/crc

Add progress bar for bundle extraction

 func untarFile(tarReader *tar.Reader, header *tar.Header, path string) error { 	} 	defer file.Close() +	var reader io.Reader+	reader = tarReader+	if isatty.IsTerminal(os.Stdout.Fd()) && header.FileInfo().Size() >= 1_000_000_000 {

done. In crcbundle, only the qcow2 file is larger than 100MB, so still only one progress bar.

guillaumerose

comment created time in 6 days

Pull request review commentcode-ready/crc

Add progress bar for bundle extraction

 func untarFile(tarReader *tar.Reader, header *tar.Header, path string) error { 	} 	defer file.Close() +	var reader io.Reader+	reader = tarReader

I added support for zip file also.

guillaumerose

comment created time in 6 days

Pull request review commentcode-ready/crc

Add progress bar for bundle extraction

 func untarFile(tarReader *tar.Reader, header *tar.Header, path string) error { 	} 	defer file.Close() +	var reader io.Reader+	reader = tarReader+	if isatty.IsTerminal(os.Stdout.Fd()) && header.FileInfo().Size() >= 1_000_000_000 {+		bar := pb.Simple.Start64(header.FileInfo().Size())+		bar.Set("prefix", fmt.Sprintf("%s: ", filepath.Base(header.Name)))

I set the max width parameter to 80. It will not fill the terminal anymore.

guillaumerose

comment created time in 6 days

Pull request review commentcode-ready/crc

Add progress bar for bundle extraction

 import ( 	"path/filepath" 	"strings" +	"github.com/cheggaaa/pb/v3" 	"github.com/code-ready/crc/pkg/crc/logging"+	"github.com/mattn/go-isatty" 	"github.com/xi2/xz" ) -func UncompressWithFilter(tarball, targetDir string, fileFilter func(string) bool) ([]string, error) {-	return uncompress(tarball, targetDir, fileFilter)+func UncompressWithFilter(tarball, targetDir string, showProgress bool, fileFilter func(string) bool) ([]string, error) {+	return uncompress(tarball, targetDir, fileFilter, showProgress) } -func Uncompress(tarball, targetDir string) ([]string, error) {-	return uncompress(tarball, targetDir, nil)+func Uncompress(tarball, targetDir string, showProgress bool) ([]string, error) {+	return uncompress(tarball, targetDir, nil, showProgress) } -func uncompress(tarball, targetDir string, fileFilter func(string) bool) ([]string, error) {+func uncompress(tarball, targetDir string, fileFilter func(string) bool, showProgress bool) ([]string, error) {

fixed, thanks!

guillaumerose

comment created time in 6 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 9c4b48ec89f94a4f554d9c01c27f0de71e916253

Add progress bar for bundle extraction The progress bar is displayed only for files larger than 100MB.

view details

push time in 6 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 59d82d8b338d2c2d4a737fdadb1f43c082522b73

untar: call defer file.Close() after each file deferred calls are only called when the current function ends, which means untar() is currently holding all files open until it has finished processing the whole archive. By introducing `untarFile()` and `unzipFile()` helpers, we can ensure each file is closed as soon as it's fully uncompressed.

view details

Guillaume Rose

commit sha 53c991f6934d0a3c42914753fd41d922e4614eac

Remove unused method

view details

Guillaume Rose

commit sha c2861b65b4593651c0f1510458dfddd175c1c535

Add progress bar for bundle extraction The progress bar is displayed only for files larger than 1GB.

view details

push time in 6 days

PR opened code-ready/crc

Use os.Args[0] instead of crc in usage hints

As a developer, I often run a custom version of crc. This PR helps to have the correct eval string.

$ crc oc-env
export PATH="/home/guillaumerose/.crc/bin/oc:$PATH"
# Run this command to configure your shell:
# eval $(crc oc-env)

$ ./out/linux-amd64/crc oc-env
export PATH="/home/guillaumerose/.crc/bin/oc:$PATH"
# Run this command to configure your shell:
# eval $(./out/linux-amd64/crc oc-env)

$ cd out/linux-amd64 
$ ./crc oc-env
export PATH="/home/guillaumerose/.crc/bin/oc:$PATH"
# Run this command to configure your shell:
# eval $(./crc oc-env)

+6 -2

0 comment

2 changed files

pr created time in 7 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 44a03e04e68e895ad51521a991455e79b0026e74

Use os.Args[0] instead of crc in usage hints

view details

push time in 7 days

create barnchguillaumerose/crc

branch : args0

created branch time in 7 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 76f4dc32d8505f8b859ec319c6df7719df6f63ab

Add progress bar for bundle extraction (only in a terminal)

view details

push time in 7 days

pull request commentcode-ready/crc

Add progress bar for bundle extraction

Other try, less chatty:

INFO Checking if NetworkManager service is running 
INFO Checking if /etc/NetworkManager/conf.d/crc-nm-dnsmasq.conf exists 
INFO Checking if /etc/NetworkManager/dnsmasq.d/crc.conf exists 
INFO Extracting bundle: crc_libvirt_4.4.8.crcbundle ... 
crc.qcow2: 445.91 MiB / 9.40 GiB [---->______________________________________________________________________________________] 4.63%

Without a pty:

level=info msg="Checking if NetworkManager service is running"
level=info msg="Checking if /etc/NetworkManager/conf.d/crc-nm-dnsmasq.conf exists"
level=info msg="Checking if /etc/NetworkManager/dnsmasq.d/crc.conf exists"
level=info msg="Extracting bundle: crc_libvirt_4.4.8.crcbundle ..."
guillaumerose

comment created time in 7 days

pull request commentcode-ready/crc

Include libmachine in crc repo

I moved the commits related to the cleanup to https://github.com/code-ready/machine/pull/35 and created this PR by just two copy and two search and replace.

guillaumerose

comment created time in 7 days

push eventguillaumerose/crc

Guillaume Rose

commit sha b40827d3ccab19fec0edac239a7159dba43e79b2

Include libmachine

view details

push time in 7 days

push eventguillaumerose/machine

Guillaume Rose

commit sha 7cb914b3f91a43079c46e4efaa910e4bc89763c0

Remove dead code. Fix unit tests.

view details

push time in 7 days

PR opened code-ready/machine

Remove code not useful for crc

Fix the linter and go test in libmachine and drivers packages.

+59 -5046

0 comment

56 changed files

pr created time in 7 days

push eventguillaumerose/machine

Christophe Fergeau

commit sha e149ff68f0482ab3377bb59a9590164e2dab1037

Merge pull request #29 from cfergeau/removed-modules Remove use of removed modules

view details

olivbour

commit sha 3b7015b22740452246993a13c2bfb39391386006

Fix HyperV Administrator check IsInRole() parameter should be either a group name [String] with potential I18N issues (the group name may change), or an SID [SecurityIdentifier]. Using a SID as a String doesn't work.

view details

Gerard Braad

commit sha b284f794e91062d5f72511eb89f3d13c7a0bbc93

Merge pull request #30 from praveenkumar/master Fix HyperV Administrator check

view details

Zeeshan Ali

commit sha f6dbb5449ec8980cf766aae445e9ebd9963ec857

Add required go version to modules file This is so that go 1.13 stops modifying the file to add version info.

view details

Christophe Fergeau

commit sha c31e0b90623d75fede28fb2911cc354f34208502

Merge pull request #31 from zeenix/go-version Add required go version to modules file

view details

Guillaume Rose

commit sha 41a887538b3aa90f5ee4325a7d03614a0fe2f6d9

hyperv package changed a lot. Fields are not the same. Removing test.

view details

Guillaume Rose

commit sha 89a8dba70ff6b6786cad2aa31f2fa9b938c0e41a

Remove dead code. Fix unit tests. Fix linter errors Remove dead code Lint

view details

push time in 7 days

create barnchguillaumerose/machine

branch : cleanup

created branch time in 7 days

push eventguillaumerose/crc

Praveen Kumar

commit sha f91d978eb2a225a8cd70baddd75f3f7ea5a8b2ca

Issue #1060 Add admin context to oc runner If a user by mistake do following then the kubeconfig file which is part of the bundle gets another entry for context and also current context sets to a different cluster. ``` $ export KUBECONFIG=~/.crc/cache/<bundle>/kubeconfig $ run kind <= way to deploy a single cluster -- Check the contexts of the kubeconfig file -- $ oc config get-contexts --kubeconfig ~/.crc/cache/<bundle>/kubeconfig CURRENT NAME CLUSTER AUTHINFO NAMESPACE admin crc admin default/api-crc-testing:6443/kube:admin api-crc-testing:6443 kube:admin/api-crc-testing:6443 default * kind-kind kind-kind kind-kind ``` With this kubeconfig file, if a user tries to run the crc cluster then he might get something following. ``` Unable to connect to the server: x509: certificate signed by unknown authority (possibly because of \"crypto/rsa: verification error\" while trying to verify candidate authority certificate \"kubernetes\")" ``` This patch make sure each oc command run use `admin` context which is default for crc cluster.

view details

Gerard Braad

commit sha 0c23b4dbddd158560a9f63fa7e0980bc59eac497

Issue #607 Disable dynamic memory (Hyper-V)

view details

MengZeLee

commit sha 6eab68aafd4ae02870c51636ea79ce67e12e6c00

fix typo fix typo

view details

Praveen Kumar

commit sha 08c905b64323705356dd00162083d41a492118cb

Add ClusterStatus struct to capture current state of cluster Till now, `GetClusterOperatorStatus` returned a bool if all the operators are in available state. With this addition now it return the if some operator are in different state and can be used for further showing the cluster status.

view details

Praveen Kumar

commit sha 960a6a2909ac9b187e6d8ac533cc27f5dd0a0a02

Issue #1041 Add intermediate `Starting` status for the cluster When starting CRC, some of the operators don't available even after 3 min wait. A user who execute `crc status` sees openshift status as stopped which is confusing since cluster is not completely stopped but waiting for some of operators to be available. After this change, user can see `Starting` as status. ``` $crc status CRC VM: Running OpenShift: Starting Disk Usage: 10.11GB of 32.2GB (Inside the CRC VM) Cache Usage: 22.99GB Cache Directory: /home/prkumar/.crc/cache ```

view details

Kevin Owen

commit sha dea3f321a8bc836d8c97918ebb67a6274754cac6

Issue #1127 Remove Windows 10 Home docs as supported edition Hyper-V is not available on this edition of Windows.

view details

Praveen Kumar

commit sha a144faedc3af8b7a50db0558b6e9d692c1be8132

Update bundle from 4.3.1 to 4.3.8

view details

Praveen Kumar

commit sha 0a318dc9335a1e7cc24c5b19b5aa383ec619f9c4

cut v1.8.0

view details

Guillaume Rose

commit sha 87f1453de9ee8f14e424bc2f993634e4bcfbe206

Add configuration file for golangci-lint. Enable goimports. Signed-off-by: Guillaume Rose <guillaume.rose@docker.com>

view details

Gerard Braad

commit sha 291385e32678ce3d6f0b75df71185d4a4a814f44

Remove issue numbers from PR template

view details

Christophe Fergeau

commit sha 27c3e22248afa24f4952a994331f49d48a2ddace

doc: Update Odo doc link to latest This was still pointing to 4.2 documentation. I don't think much changed for basic Odo use between these 2 releases, but better to point to the latest one. This replaces '4.2' with 'latest' in the doc URL so that it does not need to be updated in the future.

view details

Praveen Kumar

commit sha 51fe265365183a3e067674ae4930254b531885e4

Add cleanup feature as part of command set The 'cleanup' methods should ignore as many errors as possible when doing their job, to cope with both the nominal case, and cases where some of the checks are disabled/skipped. Which would mean we don't need to check the config options, as the 'cleanup' method will be appropriate in both cases. As of now, cleanup command for linux is partial since we are not sure if virt installed packages as part of setup command or by the user. Also user added himself as part of libvirt group or it happened by setup command.

view details

Kevin Owen

commit sha 3a25ed287113574a4b34f78e077384273c3be0ee

Issue #1147 Move docs modules to accessible directory Per modular documentation guidelines, modules should exist in an accessible location for potential use in multiple guides.

view details

Anjan Nath

commit sha 48f8a05f9a93515eae5fa8894f619b58b84bcc0f

Issue #1140 Improve AgentRunning function When an agents plist file is not loaded `launchctl list` won't show any output and the AgentRunning check was wrongly notifying that the agent is running, now we explicitly look if a PID is present in the output of `launchctl list` to determine if an agent is running.

view details

Anjan Nath

commit sha 22cbdac2fc3685bcb6840cbf082308bb8d8db47e

Issue #1140 Load plist file before trying to start a launchd agent Earlier we didn't test a scenario where an agent was removed from launchd, and expected setup to start the daemon and the tray, but to start an agent using its label, their plist has to be loaded before using `launchctl load`.

view details

Anjan Nath

commit sha c99bcd2b7efc88fc58366cbd6fbbb17cc983dc9a

Issue #1141 Force daemon plist creation with current executable's path crc setup can be ran from any location in the users computer, and we need to update the plist file with the path of the crc binary which was used to run setup

view details

Anjan Nath

commit sha 20d32e725098acf94d9032f0799e43224b9b82f5

Add more debug logs during tray setup process

view details

Kevin Owen

commit sha 60261b444c6c4a3b5567f7a5146a015ad019fe92

Issue #1156 Apply IBM style guide to documentation

view details

Praveen Kumar

commit sha 16515f6ad97c904a139b3c4f98d7a15f58f649b9

Issue #1152 Add file mode as argument for file creation As of now if a user uses umask as `0077` which is serve security level then any file created by default not readable by any different user. With this setup if that user run `crc setup` it will pass without any issue but then `crc start` fails with following error. ``` FATA Error opening file: /etc/NetworkManager/conf.d/crc-nm-dnsmasq.conf: open /etc/NetworkManager/conf.d/crc-nm-dnsmasq.conf: permission denied ``` This is happen since we create those file using `sudo` and don't provide any file-mode which makes those file only readable by root user and thus this failure.

view details

Praveen Kumar

commit sha 598780d4fcd686bafd6d69a22a3fec6387595c99

Issue #507 Implement cleanup feature for macos

view details

push time in 7 days

pull request commentcode-ready/crc

Add progress bar for bundle extraction

Yes there is too much details. The issue is that we don't have the total length of the uncompressed data. I can change the template and remove information.

What if I only display the bar for object larger than 1 GB ?

guillaumerose

comment created time in 7 days

PR opened code-ready/crc

Add progress bar for bundle extraction

This PR adds multiple things:

  • testdata for the xz format
  • a fix for a defer file.Close() in a loop
  • a progress bar for bundle extraction (in a terminal only)

Output:

INFO Checking file permissions for /etc/resolver/testing
INFO Extracting bundle: crc_hyperkit_4.4.5.crcbundle ...
crc_hyperkit_4.4.5/kubeadmin-password: 23 B / 23 B [-----------------------------------------------------------------------------------------] 100.00% ? p/s 0s
crc_hyperkit_4.4.5/kubeconfig: 11.55 KiB / 11.55 KiB [---------------------------------------------------------------------------------------] 100.00% ? p/s 0s
crc_hyperkit_4.4.5/id_rsa_crc: 2.53 KiB / 2.53 KiB [-----------------------------------------------------------------------------------------] 100.00% ? p/s 0s
crc_hyperkit_4.4.5/crc.qcow2: 490.22 MiB / 9.15 GiB [--->_______________________________________________________________________] 5.23% 34.99 MiB p/s ETA 4m13s

Tell me what you think!

+51115 -37046

0 comment

280 changed files

pr created time in 7 days

create barnchguillaumerose/crc

branch : compress

created branch time in 7 days

pull request commentcode-ready/crc

Include libmachine in crc repo

The approval is due to the new workflow. I guess it's because I'm in the owners list. If you add /lgtm I believe the bot will merge automatically. We can then avoid to add merge rights for everyone.

guillaumerose

comment created time in 7 days

PR opened openshift/installer

WIP: Add a ConfigMap with cluster profile

This PR is a spike for https://github.com/openshift/enhancements/pull/200

install-config.yaml has a new parameter: clusterProfile. It adds a ConfigMap within the openshift-config namespace with the current profile.

+119 -29

0 comment

10 changed files

pr created time in 8 days

create barnchguillaumerose/installer

branch : profile

created branch time in 8 days

pull request commentcode-ready/clicumber

Log stdout and stderr when a command is not running as expected.

Yes but it's already in an other log. It's just more convenient.

guillaumerose

comment created time in 8 days

PR opened code-ready/clicumber

Log stdout and stderr when a command is not running as expected.

The intent is to add more details when a command fails.

Example: https://deck-ci.apps.ci.l2s4.p1.openshiftapps.com/view/gs/origin-ci-test/pr-logs/pull/code-ready_crc/1297/pull-ci-code-ready-crc-master-e2e-crc/1274295022597967872

  Scenario: Push local image to OpenShift image registry # features/story_registry.feature:24
    When executing "sudo podman push hello:test default-route-openshift-image-registry.apps-crc.testing/testproj-img/hello:test --tls-verify=false" succeeds # features/story_registry.feature:28
      Error: command 'sudo podman push hello:test default-route-openshift-image-registry.apps-crc.testing/testproj-img/hello:test --tls-verify=false', expected to succeed, exited with exit code: 125
+2 -2

0 comment

1 changed file

pr created time in 8 days

create barnchguillaumerose/clicumber

branch : moredetails

created branch time in 8 days

fork guillaumerose/clicumber

Ready-to-go cucumber steps for testing CLI tools

fork in 8 days

push eventguillaumerose/crc

Guillaume Rose

commit sha ad8a35d3edab7c5723e18bff7cc6279bb1150cb2

Remove virtualbox serial driver

view details

Guillaume Rose

commit sha a527cbe4befe60b42eca75810201c82c3c413f83

Remove dead code

view details

Guillaume Rose

commit sha 537a8b9c51d542f5bccde9efff31a3d180127db1

Fix linter

view details

push time in 8 days

PR opened code-ready/crc

Include libmachine in crc repo

I had to fix tests and the linter. I removed couple of unused packages.

+8170 -506

0 comment

108 changed files

pr created time in 10 days

push eventguillaumerose/crc

Guillaume Rose

commit sha ac1f137951be0c6fce4d9333ec6fcf9cc3aef612

Remove dead code

view details

Guillaume Rose

commit sha 47d47defb5453fb282a99ff58c1fc99eaced67e6

Fix linter in pkg/drivers

view details

push time in 10 days

push eventguillaumerose/crc

Guillaume Rose

commit sha 094ba6980a15e11deb864e1765162d948193b5ad

Remove virtualbox serial driver

view details

Guillaume Rose

commit sha 598902ff024c85a9f838e128c52139985e6dd957

Remove dead code

view details

push time in 10 days

create barnchguillaumerose/crc

branch : libmachine

created branch time in 10 days

issue openedcode-ready/snc

Hyperkit: package a raw image instead of qcow2

crc will need a raw image instead of qcow2 image for https://github.com/code-ready/crc/issues/1341

@cfergeau suggested this change: https://gist.github.com/guillaumerose/87209539fd2e4d07fde2b2a2def477c8

created time in 11 days

issue openedcode-ready/crc

Hyperkit: use raw disk format instead of qcow2

Few years ago, Apple started to ship macOS with APFS. It supports sparse files allowing us to use raw file format.

According to Docker Desktop release notes, we can expect a factor 2 for disk performance:

For systems running APFS on SSD on High Sierra, use raw format VM disks by default. This improves disk throughput (from 320MiB/sec to 600MiB/sec in dd on a 2015 MacBook Pro) and disk space handling.

created time in 11 days

issue openedcode-ready/crc

Bundle compression/decompression takes too much time

The crcbundle is currently compressed using xz. This format is not natively supported by Golang and we have to use github.com/xi2/xz library.

On my machine, it takes 3m47s to decompress a bundle using xi2/xz. With xz command line tool, it takes 2m20s. xi2/xz documentation says:

// On the author's Intel Ivybridge i5, decompression speed is about
// half that of the standard XZ Utils (tested with a recent linux
// kernel tarball).

I also tried with https://github.com/ulikunitz/xz and it doesn't finish in less than 5min.

Given that the compression step in snc takes also a huge amount of time (about 30min by bundle if I'm correct), maybe we should investigate an other format.

crc_libvirt_4.4.6.crcbundle
xz: 2.0GB
zip (default level): 2.8GB
gzip (default level): 2.8GB
zstd: 2.6GB

created time in 11 days

create barnchguillaumerose/cluster-version-operator

branch : profile

created branch time in 13 days

push eventcfergeau/installer

Guillaume Rose

commit sha ea2fac222b3db2bb91763ea9bbc260de33124709

Fix unit tests

view details

push time in 13 days

push eventcfergeau/cluster-version-operator

Guillaume Rose

commit sha 02184e256ab00cb0602f5a4662f5974840f9a859

Fix unit tests Signed-off-by: Guillaume Rose <gurose@redhat.com>

view details

push time in 13 days

push eventguillaumerose/crc

Praveen Kumar

commit sha 16d015a9e978c4a2b63b1b22b8e8677298afe52a

Use deployment instead deploymentconfigs (dc) for rollout status With newer version of oc binary when you use a mapping github repo for your app deployment which have deploymentconfig as template then it uses deployment instead dc. ``` $ ~/.crc/bin/oc/oc version Client Version: 4.5.0-202005291417-9933eb9 $ ~/.crc/bin/oc/oc new-app centos/httpd-24-centos7~https://github.com/sclorg/httpd-ex [...] deployment.apps "httpd-ex" created $ oc version Client Version: 4.5.0-20200326-010450 $ oc new-app centos/httpd-24-centos7~https://github.com/sclorg/httpd-ex && oc rollout status dc/httpd-ex [...] deploymentconfig.apps.openshift.io "httpd-ex" created ```

view details

Praveen Kumar

commit sha 6e1f05516e9f62b522c5478f72178cb4afff88b9

Bump to dev version

view details

Praveen Kumar

commit sha 459acbb88823faf099cde9519a954ea872d27c1a

Fix integration test for registry image deployment With new oc version if you create an app then app have only deployment instead of deployment config.

view details

Anjan Nath

commit sha 21ee90f1e58d3d5ae24941390edeb469a484e930

Issue #1275 Extract the tray binary in correct location

view details

Anjan Nath

commit sha 97b193aa735906fdb6b75e929e0e8d9fff50c22b

Try to stop running tray when version check fails Move kill running tray process to version fix function it should have been here all along, we can only expect a running tray when there is one available, and only when it is available and the versions don't match we need to update the version in which case we'll have to stop the running tray, in normal case when fixTrayBinaryExists is called, it'll most probably not have a running tray, as that binary doesn't exist yet

view details

Anjan Nath

commit sha 4e83b0620a2883d982a1bef51769ecb2477011f4

Add helper to read utf-16 encoded files in windows Windows ini files are UTF-16LE encoded

view details

Anjan Nath

commit sha 0eeb47908fff6875faf36f1148655d0713c53d4e

Add package secpol to modify windows user rights security policy

view details

Anjan Nath

commit sha 38b19dc46d13d7d4480b79747513e6d2b349beae

Issue #1286 Add check, fix and cleanup for log on as service We want to add log on as service permission for the user to run the daemon service, we do not use the `LocalSystem` account to run the daemon service, as it provides unnecessary elevated privilages But to use the current user account it needs log as service user right which is not enabled by default.

view details

Anjan Nath

commit sha 7117b5b95600ab19c8521c55f6391e4df1d2f906

Do not return errors from cleanup functions Only log errors that occur during cleanup error should not be returned

view details

Anjan Nath

commit sha 71591914efc84991b7d1571a3e7577e5a583af58

Check if a tray exists and running before trying to remove it During setup we want to force the installation of daemon service so that it always run the crc binary which was used in the setup but previously we were not checking if a service with the same name exists or running, this patch adds that check before proceeding with the kill

view details

Anjan Nath

commit sha cc6c989b8713f147d0c676691ddae72514a3dd27

Remove log-level debug flag from the daemon command When log-level debug is passed, machine sends its logs to stdout but running from the daemon those gets discarded, when log level is not debug machine packages sends the logs to the log file, this should now put logs in the ~/.crc/crcd.log file

view details

Anjan Nath

commit sha 8e7cc284640e1066cd9282b823a6c6603a068ec3

Use package os/user to get the current username We want the logged in user's name in domain_name/user_name format, the output of whoami.exe in systems having cygwin installed doesn't provide it in the required format. Also user.Username seems to have it in correct format

view details

Anjan Nath

commit sha e3e4703e6e668aaed7a805d18838831352bc4a42

Check first if daemon service is installed during cleanup With having a check if the service is installed before trying to stop and delete it will reduce unnecessary UAC prompts, as stop and delete are privilege operations but checking if a service is installed doesn't need elevated privileges

view details

Praveen Kumar

commit sha 92ed9664632cbe08a6cd2c7b96d9fa074c6a6601

Update bundle version from 4.4.6 to 4.4.8

view details

Praveen Kumar

commit sha 6710affb0ca3e1d01fdf32b5583e29cf1b598437

cut v1.12.0

view details

Guillaume Rose

commit sha 508c0fdc6ef84fa38dd56b5277d2c9c76f508f64

Separate Start() from others commands in machine package

view details

push time in 14 days

PR closed code-ready/machine

Move hyperkit and libvirt drivers inside machine

Prepare to simplify the code around machine.

+21342 -4482

2 comments

127 changed files

guillaumerose

pr closed time in 15 days

pull request commentcode-ready/machine

Move hyperkit and libvirt drivers inside machine

Drivers should land directly in crc.

guillaumerose

comment created time in 15 days

PR closed code-ready/crc

RFC Move all drivers inside crc binary

Not finished yet.

Pros:

  • No more drivers. 2 binaries removed.
  • No more RPC between crc and drivers.
  • Allow us to delete more machine code and slowly diverge.

Cons:

  • crc is not statically compiled on linux anymore because of libvirt-go
+21337 -2051

1 comment

136 changed files

guillaumerose

pr closed time in 15 days

more