(bugfix): `run bundle` and `run bundle-upgrade` - update registry pod index directory by everettraven · Pull Request #5894 · operator-framework/operator-sdk

everettraven · 2022-06-22T17:15:44Z

The registry pod index directory now includes the CatalogSource name to prevent conflicts between indexes in the same directory

Signed-off-by: Bryce Palmer bpalmer@redhat.com

Description of the change:
Update the registry pod index directory to include the CatalogSource name.

Motivation for the change:
Potential fix for #5870

run bundle-upgrade was not working properly when upgrading a traditionally (OLM) installed operator for FBC indexes. It seems this issue is due to multiple indexes being in the same directory that have repeating package definitions.

This change makes it so that the directory indexes are stored in correlate to the CatalogSource that is created/used for the upgrade. This seems to have resolved the issue.

(This change also affects run bundle as it will now use the same directory logic for storing indexes)

Checklist

If the pull request includes user-facing changes, extra documentation is required:

Add a new changelog fragment in changelog/fragments (see changelog/fragments/00-template.yaml)
Add or update relevant sections of the docs website in website/content/en/docs

The registry pod index directory now includes the CatalogSource name to prevent conflicts between indexes in the same directory Signed-off-by: Bryce Palmer <bpalmer@redhat.com>

everettraven · 2022-06-22T17:16:16Z

/hold

needs more testing and discussion

rashmigottipati · 2022-06-22T17:17:23Z

Change looks good to me after testing some scenarios. Will approve after testing with some more images.

Signed-off-by: Bryce Palmer <bpalmer@redhat.com>

everettraven · 2022-06-23T15:37:51Z

I performed some fairly extensive tests to verify that these changes work as expected and resolve the bug. From what I can tell, these changes do resolve the bug and work as expected.

For more details on the tests expand this dropdown (warning this is a lot of stuff, ~400 lines of Markdown)

run make build before testing to ensure using latest changes

`run bundle`

kind create cluster
./build/operator-sdk olm install
./build/operator-sdk run bundle docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.1
./build/operator-sdk run bundle docker.io/vramaraj/memcached-operator-bundle:v0.0.1
./build/operator-sdk run bundle docker.io/vramaraj/nginx-operator-bundle:v0.0.1
kind delete cluster

RESULTS:
kubectl get pods:

NAME                                                              READY   STATUS      RESTARTS      AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70becdgmrn   0/1     Completed   0             3m
d5c45fcd9b236bdd5984dc447d51e75bcea50fbcd17f0bb582621d99fbnv2j5   0/1     Completed   0             42s
d9a67cc68f7ece68aed6e49e9bac8847d783b08992b6ae214797011696x9v5s   0/1     Completed   0             100s
docker-io-bpalmer-hello-kubernetes-operator-bundle-v0-0-1         1/1     Running     0             3m7s
docker-io-vramaraj-memcached-operator-bundle-v0-0-1               1/1     Running     0             105s
docker-io-vramaraj-nginx-operator-bundle-v0-0-1                   1/1     Running     0             48s
hello-kubernetes-controller-manager-c5ccf8c5f-6fjmg               2/2     Running     1 (38s ago)   2m44s
memcached-operator-controller-manager-54f5f89956-m2rvs            2/2     Running     0             89s
nginx-operator-controller-manager-769c985b56-mxwzq                2/2     Running     0             26s

kubectl get csv:

NAME                        DISPLAY            VERSION   REPLACES   PHASE
hello-kubernetes.v0.0.1     hello-kubernetes   0.0.1                Succeeded
memcached-operator.v0.0.1   memcached          0.0.1                Succeeded
nginx-operator.v0.0.1       nginx              0.0.1                Succeeded

kubectl get sub:

NAME                            PACKAGE              SOURCE                       CHANNEL
hello-kubernetes-v0-0-1-sub     hello-kubernetes     hello-kubernetes-catalog     alpha
memcached-operator-v0-0-1-sub   memcached-operator   memcached-operator-catalog   alpha
nginx-operator-v0-0-1-sub       nginx-operator       nginx-operator-catalog       alpha

`run bundle` with `--index-image`

kind create cluster
./build/operator-sdk olm install
./build/operator-sdk run bundle docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.1 --index-image quay.io/rashmigottipati/memcached-catalog-index:0.0.1
./build/operator-sdk run bundle docker.io/vramaraj/nginx-operator-bundle:v0.0.1 --index-image quay.io/rashmigottipati/memcached-catalog-index:0.0.1
kind delete cluster

RESULTS:

kubectl get pods:

NAME                                                              READY   STATUS      RESTARTS      AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70becdbnxk   0/1     Completed   0             2m44s
d5c45fcd9b236bdd5984dc447d51e75bcea50fbcd17f0bb582621d99fbxjxwh   0/1     Completed   0             79s
docker-io-bpalmer-hello-kubernetes-operator-bundle-v0-0-1         1/1     Running     0             3m10s
docker-io-vramaraj-nginx-operator-bundle-v0-0-1                   1/1     Running     0             84s
hello-kubernetes-controller-manager-c5ccf8c5f-smkwx               2/2     Running     1 (20s ago)   2m27s
nginx-operator-controller-manager-769c985b56-vwv66                2/2     Running     0             63s

kubectl get csv:

NAME                      DISPLAY            VERSION   REPLACES   PHASE
hello-kubernetes.v0.0.1   hello-kubernetes   0.0.1                Succeeded
nginx-operator.v0.0.1     nginx              0.0.1                Succeeded

kubectl get sub:

NAME                          PACKAGE            SOURCE                     CHANNEL
hello-kubernetes-v0-0-1-sub   hello-kubernetes   hello-kubernetes-catalog   alpha
nginx-operator-v0-0-1-sub     nginx-operator     nginx-operator-catalog     alpha

`run bundle` with large FBC `--index-image`

kind create cluster
./build/operator-sdk olm install
./build/operator-sdk run bundle docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.1 --index-image quay.io/operatorhubio/catalog:latest
./build/operator-sdk run bundle docker.io/vramaraj/memcached-operator-bundle:v0.0.1 --index-image quay.io/operatorhubio/catalog:latest
./build/operator-sdk run bundle docker.io/vramaraj/nginx-operator-bundle:v0.0.1 --index-image quay.io/operatorhubio/catalog:latest
kind delete cluster

RESULTS:

kubectl get pods:

NAME                                                              READY   STATUS      RESTARTS      AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70becf55s8   0/1     Completed   0             3m46s
d5c45fcd9b236bdd5984dc447d51e75bcea50fbcd17f0bb582621d99fbrc858   0/1     Completed   0             74s
d9a67cc68f7ece68aed6e49e9bac8847d783b08992b6ae2147970116968lmfv   0/1     Completed   0             2m37s
docker-io-bpalmer-hello-kubernetes-operator-bundle-v0-0-1         1/1     Running     0             3m50s
docker-io-vramaraj-memcached-operator-bundle-v0-0-1               1/1     Running     0             2m42s
docker-io-vramaraj-nginx-operator-bundle-v0-0-1                   1/1     Running     0             79s
hello-kubernetes-controller-manager-c5ccf8c5f-stng5               2/2     Running     1 (87s ago)   3m33s
memcached-operator-controller-manager-54f5f89956-5hpqf            2/2     Running     0             2m21s
nginx-operator-controller-manager-769c985b56-ss2xr                2/2     Running     0             58s

kubectl get csv:

NAME                        DISPLAY            VERSION   REPLACES   PHASE
hello-kubernetes.v0.0.1     hello-kubernetes   0.0.1                Succeeded
memcached-operator.v0.0.1   memcached          0.0.1                Succeeded
nginx-operator.v0.0.1       nginx              0.0.1                Succeeded

kubectl get sub:

NAME                            PACKAGE              SOURCE                       CHANNEL
hello-kubernetes-v0-0-1-sub     hello-kubernetes     hello-kubernetes-catalog     alpha
memcached-operator-v0-0-1-sub   memcached-operator   memcached-operator-catalog   alpha
nginx-operator-v0-0-1-sub       nginx-operator       nginx-operator-catalog       alpha

NOTE: for run bundle-upgrade tests you may have to manually approve InstallPlans

`run bundle-upgrade` from `run bundle`

kind create cluster
./build/operator-sdk olm install
./build/operator-sdk run bundle docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.1
./build/operator-sdk run bundle-upgrade docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.2
./build/operator-sdk run bundle docker.io/vramaraj/memcached-operator-bundle:v0.0.1
./build/operator-sdk run bundle-upgrade docker.io/vramaraj/memcached-operator-bundle:v0.0.2
./build/operator-sdk run bundle docker.io/vramaraj/nginx-operator-bundle:v0.0.1
./build/operator-sdk run bundle-upgrade docker.io/vramaraj/nginx-operator-bundle:v0.0.2
kind delete cluster

RESULTS:

kubectl get pods:

NAME                                                              READY   STATUS      RESTARTS      AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70bec8b9jk   0/1     Completed   0             9m55s
5027f4ecd29d0c58e1edb80f0103b1f56e8e6d81cee44a89a506f074d5tb9pf   0/1     Completed   0             8m58s
9af806535d5d66fe7fe15963d1ee5fc93aef80cf1122150bacf51d4062t8wmh   0/1     Completed   0             77s
d5c45fcd9b236bdd5984dc447d51e75bcea50fbcd17f0bb582621d99fbmkttn   0/1     Completed   0             4m24s
d9a67cc68f7ece68aed6e49e9bac8847d783b08992b6ae214797011696c2nls   0/1     Completed   0             7m14s
docker-io-bpalmer-hello-kubernetes-operator-bundle-v0-0-2         1/1     Running     0             9m
docker-io-vramaraj-memcached-operator-bundle-v0-0-2               1/1     Running     0             6m29s
docker-io-vramaraj-nginx-operator-bundle-v0-0-2                   1/1     Running     0             81s
fa7ca32af5857066b6263b5619868c7d5a5418cb2254ab735bb6577c5c2k2t2   0/1     Completed   0             6m27s
hello-kubernetes-controller-manager-655f585c9b-qlnn8              2/2     Running     3 (86s ago)   8m44s
memcached-operator-controller-manager-66575c5bb6-hh2ll            2/2     Running     0             5m2s
nginx-operator-controller-manager-b548bc6b5-kc4nm                 2/2     Running     0             44s

kubectl get csv:

NAME                        DISPLAY            VERSION   REPLACES                    PHASE
hello-kubernetes.v0.0.2     hello-kubernetes   0.0.2     hello-kubernetes.v0.0.1     Succeeded
memcached-operator.v0.0.2   memcached          0.0.2     memcached-operator.v0.0.1   Succeeded
nginx-operator.v0.0.2       nginx              0.0.2     nginx-operator.v0.0.1       Succeeded

kubectl get sub:

NAME                            PACKAGE              SOURCE                       CHANNEL
hello-kubernetes-v0-0-1-sub     hello-kubernetes     hello-kubernetes-catalog     alpha
memcached-operator-v0-0-1-sub   memcached-operator   memcached-operator-catalog   alpha
nginx-operator-v0-0-1-sub       nginx-operator       nginx-operator-catalog       alpha

`run bundle-upgrade` from OLM traditional case with `replaces` field in the upgrade CSV

kind create cluster
./build/operator-sdk olm install
Create the traditional CatalogSource:

cat << EOF | kubectl apply -f -
apiVersion: operators.coreos.com/v1alpha1
kind: CatalogSource
metadata:
  name: hello-kubernetes-operator-catalog
  namespace: operators
spec:
  displayName: Hello Kubernetes Operator
  image: docker.io/bpalmer/hello-kubernetes-sqlite-index:latest
  sourceType: grpc
EOF

Create the traditional Subscription:

cat << EOF | kubectl apply -f -
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
  name: hello-kubernetes
  namespace: operators
spec:
  channel: "alpha"
  installPlanApproval: Automatic
  name: hello-kubernetes
  source: hello-kubernetes-operator-catalog
  sourceNamespace: operators
  startingCSV: hello-kubernetes.v0.0.1
EOF

Wait for CSV state to be Succeeded by using watch -n 5 kubectl -n operators get csv
./build/operator-sdk run bundle-upgrade docker.io/bpalmer/hello-kubernetes-operator-bundle-replace:v0.0.2 -n operators
kind delete cluster

RESULTS:

kubectl -n operators get pods:

NAME                                                              READY   STATUS      RESTARTS   AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70bec9kptx   0/1     Completed   0          6m33s
b5baced527f486f89f6136c35b3c9745abba3c7f4a3ccd6eb8eb09f7dbqmrrr   0/1     Completed   0          108s
cker-io-bpalmer-hello-kubernetes-operator-bundle-replace-v0-0-2   1/1     Running     0          117s
hello-kubernetes-controller-manager-8566477c4f-25thh              2/2     Running     0          87s
hello-kubernetes-operator-catalog-4vh8s                           1/1     Running     0          7m49s

kubectl -n operators get csv:

NAME                      DISPLAY            VERSION   REPLACES                  PHASE
hello-kubernetes.v0.0.2   hello-kubernetes   0.0.2     hello-kubernetes.v0.0.1   Succeeded

kubectl -n operators get sub:

NAME               PACKAGE            SOURCE                              CHANNEL
hello-kubernetes   hello-kubernetes   hello-kubernetes-operator-catalog   alpha

`run bundle-upgrade` from OLM traditional case with `--mode semver`

kind create cluster
./build/operator-sdk olm install
Create the traditional CatalogSource:

cat << EOF | kubectl apply -f -
apiVersion: operators.coreos.com/v1alpha1
kind: CatalogSource
metadata:
  name: hello-kubernetes-operator-catalog
  namespace: operators
spec:
  displayName: Hello Kubernetes Operator
  image: docker.io/bpalmer/hello-kubernetes-sqlite-index:latest
  sourceType: grpc
EOF

Create the traditional Subscription:

cat << EOF | kubectl apply -f -
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
  name: hello-kubernetes
  namespace: operators
spec:
  channel: "alpha"
  installPlanApproval: Automatic
  name: hello-kubernetes
  source: hello-kubernetes-operator-catalog
  sourceNamespace: operators
  startingCSV: hello-kubernetes.v0.0.1
EOF

Wait for CSV state to be Succeeded by using watch -n 5 kubectl -n operators get csv
./build/operator-sdk run bundle-upgrade docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.2 -n operators --mode semver
kind delete cluster

RESULTS:

kubectl -n operators get pods:

NAME                                                              READY   STATUS      RESTARTS   AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70bec5j2m7   0/1     Completed   0          4m55s
5027f4ecd29d0c58e1edb80f0103b1f56e8e6d81cee44a89a506f074d56v2bl   0/1     Completed   0          38s
docker-io-bpalmer-hello-kubernetes-operator-bundle-v0-0-2         1/1     Running     0          43s
hello-kubernetes-controller-manager-977d868c9-4r6jg               2/2     Running     0          24s
hello-kubernetes-operator-catalog-8zdm8                           1/1     Running     0          5m14s

kubectl -n operators get csv:

NAME                      DISPLAY            VERSION   REPLACES                  PHASE
hello-kubernetes.v0.0.2   hello-kubernetes   0.0.2     hello-kubernetes.v0.0.1   Succeeded

kubectl -n operators get sub:

NAME               PACKAGE            SOURCE                              CHANNEL
hello-kubernetes   hello-kubernetes   hello-kubernetes-operator-catalog   alpha

sequential `run bundle-upgrade` from `run bundle`

kind create cluster
./build/operator-sdk olm install
./build/operator-sdk run bundle docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.1
./build/operator-sdk run bundle-upgrade docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.2
./build/operator-sdk run bundle-upgrade docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.3
./build/operator-sdk run bundle docker.io/vramaraj/memcached-operator-bundle:v0.0.1
./build/operator-sdk run bundle-upgrade docker.io/vramaraj/memcached-operator-bundle:v0.0.2
./build/operator-sdk run bundle-upgrade docker.io/vramaraj/memcached-operator-bundle:v0.0.3
./build/operator-sdk run bundle docker.io/vramaraj/nginx-operator-bundle:v0.0.1
./build/operator-sdk run bundle-upgrade docker.io/vramaraj/nginx-operator-bundle:v0.0.2
./build/operator-sdk run bundle-upgrade docker.io/vramaraj/nginx-operator-bundle:v0.0.3
kind delete cluster

RESULTS:

kubectl get pods:

NAME                                                              READY   STATUS      RESTARTS       AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70becg785k   0/1     Completed   0              14m
461cd7fc535a0c48bff67543834ad617c9a3dd7a66093c4d00aac35a2bqmsl4   0/1     Completed   0              2m9s
5027f4ecd29d0c58e1edb80f0103b1f56e8e6d81cee44a89a506f074d5km8tt   0/1     Completed   0              13m
9af806535d5d66fe7fe15963d1ee5fc93aef80cf1122150bacf51d4062r9pjj   0/1     Completed   0              3m56s
a2b0380e5e2b579202936f3609c7557681273187c9ac0dcc69871cc695d4rkn   0/1     Completed   0              12m
d5c45fcd9b236bdd5984dc447d51e75bcea50fbcd17f0bb582621d99fbcw69p   0/1     Completed   0              4m47s
d9a67cc68f7ece68aed6e49e9bac8847d783b08992b6ae214797011696wgnd7   0/1     Completed   0              11m
docker-io-bpalmer-hello-kubernetes-operator-bundle-v0-0-3         1/1     Running     0              12m
docker-io-vramaraj-memcached-operator-bundle-v0-0-3               1/1     Running     0              5m46s
docker-io-vramaraj-nginx-operator-bundle-v0-0-3                   1/1     Running     0              2m11s
fa11d19190f3997c5393cb0ec623a0cae7a4e93d151f0611a958097e3a9wfcj   0/1     Completed   0              5m43s
fa7ca32af5857066b6263b5619868c7d5a5418cb2254ab735bb6577c5cf7czh   0/1     Completed   0              10m
hello-kubernetes-controller-manager-7cbb6b667b-tjv5f              2/2     Running     4 (113s ago)   11m
memcached-operator-controller-manager-69f48468d4-d9ktl            2/2     Running     0              5m28s
nginx-operator-controller-manager-8548564b7c-j544r                2/2     Running     0              102s

The 4 restarts on the hello-kubernetes-controller-manager-7cbb6b667b-tjv5f seem to be due to an error in the hello-kuberenetes operator itself and has nothing to do with
run bundle or run bundle-upgrade

kubectl get csv:

NAME                        DISPLAY            VERSION   REPLACES                    PHASE
hello-kubernetes.v0.0.3     hello-kubernetes   0.0.3     hello-kubernetes.v0.0.2     Installing
memcached-operator.v0.0.3   memcached          0.0.3     memcached-operator.v0.0.2   Succeeded
nginx-operator.v0.0.3       nginx              0.0.3     nginx-operator.v0.0.2       Succeeded

hello-kuberenetes.v0.0.3 is stuck in the Installing phase due to an error in the v0.0.3 version of the hello-kubernetes operator itself and has nothing to do with
run bundle or run bundle-upgrade

kubectl get sub:

NAME                            PACKAGE              SOURCE                       CHANNEL
hello-kubernetes-v0-0-1-sub     hello-kubernetes     hello-kubernetes-catalog     alpha
memcached-operator-v0-0-1-sub   memcached-operator   memcached-operator-catalog   alpha
nginx-operator-v0-0-1-sub       nginx-operator       nginx-operator-catalog       alpha

sequential `run bundle-upgrade` from OLM traditional case with `--mode semver`

kind create cluster
./build/operator-sdk olm install
Create the traditional CatalogSource:

cat << EOF | kubectl apply -f -
apiVersion: operators.coreos.com/v1alpha1
kind: CatalogSource
metadata:
  name: hello-kubernetes-operator-catalog
  namespace: operators
spec:
  displayName: Hello Kubernetes Operator
  image: docker.io/bpalmer/hello-kubernetes-sqlite-index:latest
  sourceType: grpc
EOF

Create the traditional Subscription:

cat << EOF | kubectl apply -f -
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
  name: hello-kubernetes
  namespace: operators
spec:
  channel: "alpha"
  installPlanApproval: Automatic
  name: hello-kubernetes
  source: hello-kubernetes-operator-catalog
  sourceNamespace: operators
  startingCSV: hello-kubernetes.v0.0.1
EOF

Wait for CSV state to be Succeeded by using watch -n 5 kubectl -n operators get csv
./build/operator-sdk run bundle-upgrade docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.2 -n operators --mode semver
./build/operator-sdk run bundle-upgrade docker.io/bpalmer/hello-kubernetes-operator-bundle:v0.0.3 -n operators --mode semver
kind delete cluster

RESULTS:

kubectl -n operators get pods:

NAME                                                              READY   STATUS      RESTARTS   AGE
0218697a0b6f89b58fdb43fce61d2f6332f17d4ff22fc18ec40be70bec7p2kn   0/1     Completed   0          5m7s
5027f4ecd29d0c58e1edb80f0103b1f56e8e6d81cee44a89a506f074d5jznzs   0/1     Completed   0          106s
a2b0380e5e2b579202936f3609c7557681273187c9ac0dcc69871cc695kb2fq   0/1     Completed   0          50s
docker-io-bpalmer-hello-kubernetes-operator-bundle-v0-0-3         1/1     Running     0          57s
hello-kubernetes-controller-manager-54c96fcb97-t24r7              2/2     Running     0          41s
hello-kubernetes-operator-catalog-j8bnr                           1/1     Running     0          5m54s

kubectl -n operators get csv:

NAME                      DISPLAY            VERSION   REPLACES                  PHASE
hello-kubernetes.v0.0.3   hello-kubernetes   0.0.3     hello-kubernetes.v0.0.2   Succeeded

kubectl -n operators get sub:

NAME               PACKAGE            SOURCE                              CHANNEL
hello-kubernetes   hello-kubernetes   hello-kubernetes-operator-catalog   alpha

Signed-off-by: Bryce Palmer <bpalmer@redhat.com>

rashmigottipati

/lgtm

Tested various scenarios extensively and these changes don't break any existing scenarios and also fixes a traditional upgrade use-case bug.

rashmigottipati · 2022-06-23T20:45:22Z

/hold cancel

everettraven · 2022-06-30T15:39:26Z

/cherry-pick v1.22.x

openshift-cherrypick-robot · 2022-06-30T15:40:04Z

@everettraven: #5894 failed to apply on top of branch "v1.22.x":

Applying: update registry pod index directory
Applying: copy channel from original and always add package
Using index info to reconstruct a base tree...
M	internal/olm/operator/registry/index_image.go
Falling back to patching base and 3-way merge...
Auto-merging internal/olm/operator/registry/index_image.go
CONFLICT (content): Merge conflict in internal/olm/operator/registry/index_image.go
error: Failed to merge in the changes.
hint: Use 'git am --show-current-patch=diff' to see the failed patch
Patch failed at 0002 copy channel from original and always add package
When you have resolved this problem, run "git am --continue".
If you prefer to skip this patch, run "git am --skip" instead.
To restore the original branch and stop patching, run "git am --abort".

Details

In response to this:

/cherry-pick v1.22.x

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

everettraven · 2022-06-30T21:36:48Z

/cherry-pick v1.22.x

openshift-cherrypick-robot · 2022-06-30T21:37:27Z

@everettraven: new pull request created: #5920

Details

In response to this:

/cherry-pick v1.22.x

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

update registry pod index directory

d4fc57c

The registry pod index directory now includes the CatalogSource name to prevent conflicts between indexes in the same directory Signed-off-by: Bryce Palmer <bpalmer@redhat.com>

everettraven requested a review from rashmigottipati June 22, 2022 17:15

openshift-ci bot requested review from asmacdo and varshaprasad96 June 22, 2022 17:15

openshift-ci bot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Jun 22, 2022

copy channel from original and always add package

80304d4

Signed-off-by: Bryce Palmer <bpalmer@redhat.com>

copy Other metadata from original DC

9a402f8

Signed-off-by: Bryce Palmer <bpalmer@redhat.com>

rashmigottipati approved these changes Jun 23, 2022

View reviewed changes

openshift-ci bot assigned rashmigottipati Jun 23, 2022

openshift-ci bot added the lgtm Indicates that a PR is ready to be merged. label Jun 23, 2022

rashmigottipati removed the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Jun 23, 2022

everettraven merged commit 121fac5 into operator-framework:master Jun 23, 2022

everettraven deleted the bugfix/run-bundle-upgrade-traditional branch June 23, 2022 20:56

openshift-cherrypick-robot mentioned this pull request Jun 30, 2022

[v1.22.x] (bugfix): run bundle and run bundle-upgrade - update registry pod index directory #5920

Merged

Conversation

everettraven commented Jun 22, 2022

Uh oh!

everettraven commented Jun 22, 2022

Uh oh!

rashmigottipati commented Jun 22, 2022

Uh oh!

everettraven commented Jun 23, 2022

run bundle

run bundle with --index-image

run bundle with large FBC --index-image

run bundle-upgrade from run bundle

run bundle-upgrade from OLM traditional case with replaces field in the upgrade CSV

run bundle-upgrade from OLM traditional case with --mode semver

sequential run bundle-upgrade from run bundle

sequential run bundle-upgrade from OLM traditional case with --mode semver

Uh oh!

rashmigottipati left a comment

Choose a reason for hiding this comment

Uh oh!

rashmigottipati commented Jun 23, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

everettraven commented Jun 30, 2022

Uh oh!

openshift-cherrypick-robot commented Jun 30, 2022

Uh oh!

everettraven commented Jun 30, 2022

Uh oh!

openshift-cherrypick-robot commented Jun 30, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

`run bundle`

`run bundle` with `--index-image`

`run bundle` with large FBC `--index-image`

`run bundle-upgrade` from `run bundle`

`run bundle-upgrade` from OLM traditional case with `replaces` field in the upgrade CSV

`run bundle-upgrade` from OLM traditional case with `--mode semver`

sequential `run bundle-upgrade` from `run bundle`

sequential `run bundle-upgrade` from OLM traditional case with `--mode semver`

rashmigottipati commented Jun 23, 2022 •

edited

Loading