Description of problem: Many released kubernetes implementations don't yet support the 'Width' or 'Height' style stuff on the container terminal websocket. So lets filter that stuff out for now. The kubernetes-container-terminal project sends it, but we have a chance to intercept it. It breaks tests. Version-Release number of selected component (if applicable): 122-1 How reproducible: Always Additional info: Fixed upstream in 395be7953ab4fe6506e6d24e87fe7d18f9691e70
Upstream master commit https://github.com/cockpit-project/cockpit/commit/fb72d3c4ed13cb84900050ac35ad1341c56bfea1
Actually, I don't see this bug in Cockpit 122. The updated container-terminal code arrived in 123. Dominik is this an actual issue in 122 somehow?
Reassigning this to OCP, since it doesn't have to do with Cockpit as delivered in RHEL. It has to do with the Registry. However I don't think this affects Cockpit 122 at all. I believe it is and will be a bug in Cockpit 123.
(In reply to Stef Walter from comment #2) > Actually, I don't see this bug in Cockpit 122. The updated > container-terminal code arrived in 123. Dominik is this an actual issue in > 122 somehow? Several weak reasons combine into a good sum: * This patch fixes some obvious errors (e.g. using '=' instead of '===') * Cockpit 122 will be around for a while (products that use Cockpit from RHEL Extras), I interpreted this as better compatibility * Another fix (for https://bugzilla.redhat.com/show_bug.cgi?id=1394675) is based on changes in this patch. More splitting up of patches increases risk of things breaking.
After discussing upstream, we've decided to pick just the one line from here that affects https://bugzilla.redhat.com/show_bug.cgi?id=1394675. The remainder of the patch does not matter for Cockpit 122 or the registry container which is based on it.
Development Management has reviewed and declined this request. You may appeal this decision by reopening this request.