You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi folks! I'm creating this issue thread to discuss the relevant topics wrt the process of building scanner images.
Creating this here because stackrox/scanner does not have an Issues or Discussions tab.
As part of a broader effort, we've been trying to enable multi-arch image building so that we can run ACS jobs on Power.
So far, we've successfully built the collector-builder and collector-slim image and we're currently working on the images under stackrox tree (main, central-db, webhookserver), the PR for which is on the way. Once merged, the scanner set of images (scanner-slim, scanner-db-slim) is what we'll be looking at, but here's the deal:
As it stands today, IIUC, the scanner images are built using OpenShift-CI. This is different from the other two repos i.e collector and stackrox, both of which employ GitHub Actions to build the upstream images and as I understand, it is not straightforward to build multi-arch images on OpenShift-CI. We are aware that there are plans to move the scanner CI workflows from OpenShift-CI to Actions, however we wanted to discuss more on this.
There is a Power cluster available today to run jobs using OpenShift-CI, however since the stackrox images aren't multi-arched yet - we can't run any stackrox jobs. My question here is, is it possible, at all, to utilize this cluster to build the scanner images once we have the prerequisite images multi-arched? If so, what kind of efforts would we be looking at?
Follow up question, if we do build the ppc64le images for scanner using OpenShift-CI, how would one go about creating the multi-arch manifest?
Or instead of going through all that, is it better to just wait for the move to GitHub Actions?
Looking for some insight here from the stackrox/scanner team. Thanks!
The text was updated successfully, but these errors were encountered:
+1 @anup-kodlekere , when you say linux/s390x, wouldn't it require PostgreSQL installation refactoring?
Although PostgreSQL build farm has references to almost every possible architecture, RPM packages don't exist for either linux/ppc64le, or linux/s390x on EL-9.
Hi folks! I'm creating this issue thread to discuss the relevant topics wrt the process of building scanner images.
Creating this here because stackrox/scanner does not have an
Issues
orDiscussions
tab.As part of a broader effort, we've been trying to enable multi-arch image building so that we can run ACS jobs on Power.
So far, we've successfully built the collector-builder and collector-slim image and we're currently working on the images under stackrox tree (
main
,central-db
,webhookserver
), the PR for which is on the way. Once merged, the scanner set of images (scanner-slim
,scanner-db-slim
) is what we'll be looking at, but here's the deal:As it stands today, IIUC, the
scanner
images are built using OpenShift-CI. This is different from the other two repos i.ecollector
andstackrox
, both of which employ GitHub Actions to build the upstream images and as I understand, it is not straightforward to build multi-arch images on OpenShift-CI. We are aware that there are plans to move the scanner CI workflows from OpenShift-CI to Actions, however we wanted to discuss more on this.There is a Power cluster available today to run jobs using OpenShift-CI, however since the stackrox images aren't multi-arched yet - we can't run any stackrox jobs. My question here is, is it possible, at all, to utilize this cluster to build the scanner images once we have the prerequisite images multi-arched? If so, what kind of efforts would we be looking at?
Follow up question, if we do build the ppc64le images for scanner using OpenShift-CI, how would one go about creating the multi-arch manifest?
Or instead of going through all that, is it better to just wait for the move to GitHub Actions?
Looking for some insight here from the stackrox/scanner team. Thanks!
The text was updated successfully, but these errors were encountered: