Date
1 - 8 of 8
Silicon MRI Version Bump happening now
|
On Wed, Oct 21, 2020 at 11:06 AM Robert Varga <nite@...> wrote: On 21/10/2020 09:44, Lori Jakab wrote: I removed the "WIP" from the commit message, and now the verify job is UNSTABLE, not sure why... should I still merge?
|
|
|
|
Robert Varga
On 21/10/2020 09:44, Lori Jakab wrote:
Robert,Awesome, and good to hear we are squashing bugs :) The verify job passes now, the distribution check fails. Is thatYes, it is, the patch is okay to be merged. Thanks, Robert |
|
|
|
Robert, I finally fixed all tests in lispflowmapping, they were catching real bugs, which could have resulted in run-time exceptions, so it's great! The verify job passes now, the distribution check fails. Is that expected at this point? -Lori On Mon, Oct 19, 2020 at 10:19 PM Robert Varga <nite@...> wrote: On 19/10/2020 00:21, Robert Varga wrote: |
|
|
|
Robert Varga
On 19/10/2020 00:21, Robert Varga wrote:
The third is tracked by MDSAL-600 and causes an UT failure inand this has bitten transportpce as well. netvirt/lispflowpluggin patches have been updated. all others are being updated right now. Regards, Robert |
|
|
|
Robert Varga
On 19/10/2020 09:23, Lori Jakab wrote:
Hi Robert,Ah, yes, it's this one: https://docs.opendaylight.org/en/latest/release-notes/upgrade-process.html#binding-dtos-enforce-non-null-keys (BTW, the title says Silicon, but then the text talks about Aluminium).Thanks for noticing, yeah, the document needs polishing :) I'll take a look at the tests.Thanks, Robert |
|
|
|
Hi Robert, Do you know if the lispflowmapping failures are related to one of the items on the MD-SAL impacts list on this document? https://docs.opendaylight.org/en/latest/release-notes/upgrade-process.html (BTW, the title says Silicon, but then the text talks about Aluminium). I'll take a look at the tests. -Lori On Mon, Oct 19, 2020 at 12:21 AM Robert Varga <nite@...> wrote: On 16/10/2020 08:59, Robert Varga wrote: |
|
|
|
Robert Varga
On 16/10/2020 08:59, Robert Varga wrote:
Hello everyone,So we have real progress (thanks, Richard) and are down to two projects: lispflowmapping and netvirt, both of which need attention from regular committers. Lispflowmapping is the easier, as it has just some issues around keys not being filled up. Netvirt has three problems: - unfilled key properties causing NPEs - modernizer violations - a failure with MDSAL :) The first two need attention from people knowledgeable of the code to properly fix (first) or fix-or-ignore (second) issue. The third is tracked by MDSAL-600 and causes an UT failure in bgpmanager-impl with a nasty stack trace. Fixed-up mdsal-7.0.1 will be available and make its way down the projects tomorrow. Given that these are leaf projects holding up autorelease/distribution, please indicate whether you will be able to lend some resources quickly, or whether the 'drop-from-autorelease' option is more appropriate in the bid of spend minimal amount resources. Thanks, Robert |
|
|
|
Robert Varga
Hello everyone,
this is a heads up that the "Version Bump" step of Silicon release is happening right now. The patches to managed projects are staged here: https://git.opendaylight.org/gerrit/q/topic:silicon-mri The migration guide is here: https://docs.opendaylight.org/en/latest/release-notes/upgrade-process.html Not all the patches are ready, but that is okay as per the rules governing SimRel: we have the next two weeks to stabilize autorelease build (by 10/29/2020, Version Bump Checkpoint). This also means that distribution-check jobs are expected to fail and we are flying without CSIT. I therefore encourage all projects to look at their respective patches and invest some time into sorting them out. Regards, Robert |
|
|