[DISCUSS] JanusGraph 0.3.2 release


Oleksandr Porunov <alexand...@...>
 

Hi everyone,

I am going to make "0.3.2" release PR after conflicts in #1609 is resolved and the PR is merged.
After that I am going to finalize artifacts, docs and start a voting process. 
I have moved all opened issues from "0.3.2" milestone into "v0.3.x" milestone.

I have removed these two issues from milestones: #1547 and #1588 because they look more like features and I am not sure that they will be resolved soon (Of course I will be happy to review them when they are fixed). As I understand we re-targeted them from master into "0.3" because they are partially connected to the bugfix but on the other hand they are introducing a new parameter, so I don't know which branch they should be targeted. I am OK with both "0.3" and "master" branches.

Please, tell if we need to include something else into "0.3.2" release.
Estimated time for starting a release process is 1-2 days.

Best regards,
Oleksandr


Chris Hupman <chris...@...>
 

As an FYI. I found some changes in 0.3 that will make my release notes not entirely accurate for the 0.3.2 and 0.4.0 releases. My plan is to open a PR updating RELEASING.md with my notes. I'll do my best to get it out this week, but it might be early next week.

I think you'll want to get #1601 in as well. Right now it shows test failures, but those are for extended tests from travis.yml.cassandra so those failures will need to get addressed in the release. It'll make it easier for you to test out all the Cassandra versions that we claim to support. Rather than doing a separate test in your local branch with the alternate travis config you could just add [full build] to your commit message.

I agree on not including #1547 and #1588. There's some testing going on into whether or not they're still necessary after #1606. The only other possible PR to go in seems like #1406?



On Thursday, May 30, 2019 at 7:49:50 AM UTC-7, Oleksandr Porunov wrote:
Hi everyone,

I am going to make "0.3.2" release PR after conflicts in #1609 is resolved and the PR is merged.
After that I am going to finalize artifacts, docs and start a voting process. 
I have moved all opened issues from "0.3.2" milestone into "v0.3.x" milestone.

I have removed these two issues from milestones: #1547 and #1588 because they look more like features and I am not sure that they will be resolved soon (Of course I will be happy to review them when they are fixed). As I understand we re-targeted them from master into "0.3" because they are partially connected to the bugfix but on the other hand they are introducing a new parameter, so I don't know which branch they should be targeted. I am OK with both "0.3" and "master" branches.

Please, tell if we need to include something else into "0.3.2" release.
Estimated time for starting a release process is 1-2 days.

Best regards,
Oleksandr


Oleksandr Porunov <alexand...@...>
 

I have reviewed and added #1601 into milestone. We should then figure our why those tests are falling.
I am not sure if #1406 will be fixed soon. If it is then I would like to get it into the release.

Right now the plan is:
- Merge #1601
- Fix tests (or bugs if they are found) which are shown in #1601
- Wait for RELEASING.md update from Chris
- Prepare a release commit


On Thursday, May 30, 2019 at 10:20:34 PM UTC+3, Chris Hupman wrote:
As an FYI. I found some changes in 0.3 that will make my release notes not entirely accurate for the 0.3.2 and 0.4.0 releases. My plan is to open a PR updating RELEASING.md with my notes. I'll do my best to get it out this week, but it might be early next week.

I think you'll want to get #1601 in as well. Right now it shows test failures, but those are for extended tests from travis.yml.cassandra so those failures will need to get addressed in the release. It'll make it easier for you to test out all the Cassandra versions that we claim to support. Rather than doing a separate test in your local branch with the alternate travis config you could just add [full build] to your commit message.

I agree on not including #1547 and #1588. There's some testing going on into whether or not they're still necessary after #1606. The only other possible PR to go in seems like #1406?



On Thursday, May 30, 2019 at 7:49:50 AM UTC-7, Oleksandr Porunov wrote:
Hi everyone,

I am going to make "0.3.2" release PR after conflicts in #1609 is resolved and the PR is merged.
After that I am going to finalize artifacts, docs and start a voting process. 
I have moved all opened issues from "0.3.2" milestone into "v0.3.x" milestone.

I have removed these two issues from milestones: #1547 and #1588 because they look more like features and I am not sure that they will be resolved soon (Of course I will be happy to review them when they are fixed). As I understand we re-targeted them from master into "0.3" because they are partially connected to the bugfix but on the other hand they are introducing a new parameter, so I don't know which branch they should be targeted. I am OK with both "0.3" and "master" branches.

Please, tell if we need to include something else into "0.3.2" release.
Estimated time for starting a release process is 1-2 days.

Best regards,
Oleksandr


Chris Hupman <chris...@...>
 

I'm out tomorrow, but I was able to get a PR out for the RELEASING.md updates. I plan to look at it again with fresh eyes next week, but I feel like it's ready for feedback.

Also, I think we should wait to merge #1601 until we figure out what's causing the failures in case it can be resolved by updating the test declaration in travis. 

Regards,

Chris


On Thursday, May 30, 2019 at 4:01:54 PM UTC-7, Oleksandr Porunov wrote:
I have reviewed and added #1601 into milestone. We should then figure our why those tests are falling.
I am not sure if #1406 will be fixed soon. If it is then I would like to get it into the release.

Right now the plan is:
- Merge #1601
- Fix tests (or bugs if they are found) which are shown in #1601
- Wait for RELEASING.md update from Chris
- Prepare a release commit

On Thursday, May 30, 2019 at 10:20:34 PM UTC+3, Chris Hupman wrote:
As an FYI. I found some changes in 0.3 that will make my release notes not entirely accurate for the 0.3.2 and 0.4.0 releases. My plan is to open a PR updating RELEASING.md with my notes. I'll do my best to get it out this week, but it might be early next week.

I think you'll want to get #1601 in as well. Right now it shows test failures, but those are for extended tests from travis.yml.cassandra so those failures will need to get addressed in the release. It'll make it easier for you to test out all the Cassandra versions that we claim to support. Rather than doing a separate test in your local branch with the alternate travis config you could just add [full build] to your commit message.

I agree on not including #1547 and #1588. There's some testing going on into whether or not they're still necessary after #1606. The only other possible PR to go in seems like #1406?



On Thursday, May 30, 2019 at 7:49:50 AM UTC-7, Oleksandr Porunov wrote:
Hi everyone,

I am going to make "0.3.2" release PR after conflicts in #1609 is resolved and the PR is merged.
After that I am going to finalize artifacts, docs and start a voting process. 
I have moved all opened issues from "0.3.2" milestone into "v0.3.x" milestone.

I have removed these two issues from milestones: #1547 and #1588 because they look more like features and I am not sure that they will be resolved soon (Of course I will be happy to review them when they are fixed). As I understand we re-targeted them from master into "0.3" because they are partially connected to the bugfix but on the other hand they are introducing a new parameter, so I don't know which branch they should be targeted. I am OK with both "0.3" and "master" branches.

Please, tell if we need to include something else into "0.3.2" release.
Estimated time for starting a release process is 1-2 days.

Best regards,
Oleksandr


Chris Hupman <chris...@...>
 

Hey All,

Some of the extended Cassandra tests are failing on the 0.3 branch. Details can be found in PR 1601. I think the tests have been broken since 0.3.0 was released. Based off the milestone this is currently blocking the release. How do we want to proceed? 

I can remove the 2 failing tests from that PR and open an issue to fix the tests in the next release cycle. I could also move them into an allowed failure group for the time being (my vote). Or we could hold up until the tests are fixed.

I spent a few hours this week trying to resolve the failures, but was unsuccessful I don't have the cycles right now to keep digging. I'm hoping we can get the releases moving again. 

Regards,

Chris


On Thursday, May 30, 2019 at 4:18:22 PM UTC-7, Chris Hupman wrote:
I'm out tomorrow, but I was able to get a PR out for the RELEASING.md updates. I plan to look at it again with fresh eyes next week, but I feel like it's ready for feedback.

Also, I think we should wait to merge #1601 until we figure out what's causing the failures in case it can be resolved by updating the test declaration in travis. 

Regards,

Chris

On Thursday, May 30, 2019 at 4:01:54 PM UTC-7, Oleksandr Porunov wrote:
I have reviewed and added #1601 into milestone. We should then figure our why those tests are falling.
I am not sure if #1406 will be fixed soon. If it is then I would like to get it into the release.

Right now the plan is:
- Merge #1601
- Fix tests (or bugs if they are found) which are shown in #1601
- Wait for RELEASING.md update from Chris
- Prepare a release commit

On Thursday, May 30, 2019 at 10:20:34 PM UTC+3, Chris Hupman wrote:
As an FYI. I found some changes in 0.3 that will make my release notes not entirely accurate for the 0.3.2 and 0.4.0 releases. My plan is to open a PR updating RELEASING.md with my notes. I'll do my best to get it out this week, but it might be early next week.

I think you'll want to get #1601 in as well. Right now it shows test failures, but those are for extended tests from travis.yml.cassandra so those failures will need to get addressed in the release. It'll make it easier for you to test out all the Cassandra versions that we claim to support. Rather than doing a separate test in your local branch with the alternate travis config you could just add [full build] to your commit message.

I agree on not including #1547 and #1588. There's some testing going on into whether or not they're still necessary after #1606. The only other possible PR to go in seems like #1406?



On Thursday, May 30, 2019 at 7:49:50 AM UTC-7, Oleksandr Porunov wrote:
Hi everyone,

I am going to make "0.3.2" release PR after conflicts in #1609 is resolved and the PR is merged.
After that I am going to finalize artifacts, docs and start a voting process. 
I have moved all opened issues from "0.3.2" milestone into "v0.3.x" milestone.

I have removed these two issues from milestones: #1547 and #1588 because they look more like features and I am not sure that they will be resolved soon (Of course I will be happy to review them when they are fixed). As I understand we re-targeted them from master into "0.3" because they are partially connected to the bugfix but on the other hand they are introducing a new parameter, so I don't know which branch they should be targeted. I am OK with both "0.3" and "master" branches.

Please, tell if we need to include something else into "0.3.2" release.
Estimated time for starting a release process is 1-2 days.

Best regards,
Oleksandr


Oleksandr Porunov <alexand...@...>
 

Hello Chris,

Yes, this is the only PR right now for 0.3.2 release. After that I can start a release preparation.
My vote is also to move them into an allowed failure group and making necessary issues for fixing those tests.
In other case we can just move #1601 into the next release milestone.

On Friday, June 14, 2019 at 2:04:51 AM UTC+3, Chris Hupman wrote:
Hey All,

Some of the extended Cassandra tests are failing on the 0.3 branch. Details can be found in PR 1601. I think the tests have been broken since 0.3.0 was released. Based off the milestone this is currently blocking the release. How do we want to proceed? 

I can remove the 2 failing tests from that PR and open an issue to fix the tests in the next release cycle. I could also move them into an allowed failure group for the time being (my vote). Or we could hold up until the tests are fixed.

I spent a few hours this week trying to resolve the failures, but was unsuccessful I don't have the cycles right now to keep digging. I'm hoping we can get the releases moving again. 

Regards,

Chris

On Thursday, May 30, 2019 at 4:18:22 PM UTC-7, Chris Hupman wrote:
I'm out tomorrow, but I was able to get a PR out for the RELEASING.md updates. I plan to look at it again with fresh eyes next week, but I feel like it's ready for feedback.

Also, I think we should wait to merge #1601 until we figure out what's causing the failures in case it can be resolved by updating the test declaration in travis. 

Regards,

Chris

On Thursday, May 30, 2019 at 4:01:54 PM UTC-7, Oleksandr Porunov wrote:
I have reviewed and added #1601 into milestone. We should then figure our why those tests are falling.
I am not sure if #1406 will be fixed soon. If it is then I would like to get it into the release.

Right now the plan is:
- Merge #1601
- Fix tests (or bugs if they are found) which are shown in #1601
- Wait for RELEASING.md update from Chris
- Prepare a release commit

On Thursday, May 30, 2019 at 10:20:34 PM UTC+3, Chris Hupman wrote:
As an FYI. I found some changes in 0.3 that will make my release notes not entirely accurate for the 0.3.2 and 0.4.0 releases. My plan is to open a PR updating RELEASING.md with my notes. I'll do my best to get it out this week, but it might be early next week.

I think you'll want to get #1601 in as well. Right now it shows test failures, but those are for extended tests from travis.yml.cassandra so those failures will need to get addressed in the release. It'll make it easier for you to test out all the Cassandra versions that we claim to support. Rather than doing a separate test in your local branch with the alternate travis config you could just add [full build] to your commit message.

I agree on not including #1547 and #1588. There's some testing going on into whether or not they're still necessary after #1606. The only other possible PR to go in seems like #1406?



On Thursday, May 30, 2019 at 7:49:50 AM UTC-7, Oleksandr Porunov wrote:
Hi everyone,

I am going to make "0.3.2" release PR after conflicts in #1609 is resolved and the PR is merged.
After that I am going to finalize artifacts, docs and start a voting process. 
I have moved all opened issues from "0.3.2" milestone into "v0.3.x" milestone.

I have removed these two issues from milestones: #1547 and #1588 because they look more like features and I am not sure that they will be resolved soon (Of course I will be happy to review them when they are fixed). As I understand we re-targeted them from master into "0.3" because they are partially connected to the bugfix but on the other hand they are introducing a new parameter, so I don't know which branch they should be targeted. I am OK with both "0.3" and "master" branches.

Please, tell if we need to include something else into "0.3.2" release.
Estimated time for starting a release process is 1-2 days.

Best regards,
Oleksandr


Florian Hockmann <f...@...>
 

I could also move them into an allowed failure group for the time being (my vote).

Agreed, that sounds like the best approach. We don't have to fix everything at once if it was already broken in past releases.

Am Freitag, 14. Juni 2019 01:34:15 UTC+2 schrieb Oleksandr Porunov:
Hello Chris,

Yes, this is the only PR right now for 0.3.2 release. After that I can start a release preparation.
My vote is also to move them into an allowed failure group and making necessary issues for fixing those tests.
In other case we can just move #1601 into the next release milestone.

On Friday, June 14, 2019 at 2:04:51 AM UTC+3, Chris Hupman wrote:
Hey All,

Some of the extended Cassandra tests are failing on the 0.3 branch. Details can be found in PR 1601. I think the tests have been broken since 0.3.0 was released. Based off the milestone this is currently blocking the release. How do we want to proceed? 

I can remove the 2 failing tests from that PR and open an issue to fix the tests in the next release cycle. I could also move them into an allowed failure group for the time being (my vote). Or we could hold up until the tests are fixed.

I spent a few hours this week trying to resolve the failures, but was unsuccessful I don't have the cycles right now to keep digging. I'm hoping we can get the releases moving again. 

Regards,

Chris

On Thursday, May 30, 2019 at 4:18:22 PM UTC-7, Chris Hupman wrote:
I'm out tomorrow, but I was able to get a PR out for the RELEASING.md updates. I plan to look at it again with fresh eyes next week, but I feel like it's ready for feedback.

Also, I think we should wait to merge #1601 until we figure out what's causing the failures in case it can be resolved by updating the test declaration in travis. 

Regards,

Chris

On Thursday, May 30, 2019 at 4:01:54 PM UTC-7, Oleksandr Porunov wrote:
I have reviewed and added #1601 into milestone. We should then figure our why those tests are falling.
I am not sure if #1406 will be fixed soon. If it is then I would like to get it into the release.

Right now the plan is:
- Merge #1601
- Fix tests (or bugs if they are found) which are shown in #1601
- Wait for RELEASING.md update from Chris
- Prepare a release commit

On Thursday, May 30, 2019 at 10:20:34 PM UTC+3, Chris Hupman wrote:
As an FYI. I found some changes in 0.3 that will make my release notes not entirely accurate for the 0.3.2 and 0.4.0 releases. My plan is to open a PR updating RELEASING.md with my notes. I'll do my best to get it out this week, but it might be early next week.

I think you'll want to get #1601 in as well. Right now it shows test failures, but those are for extended tests from travis.yml.cassandra so those failures will need to get addressed in the release. It'll make it easier for you to test out all the Cassandra versions that we claim to support. Rather than doing a separate test in your local branch with the alternate travis config you could just add [full build] to your commit message.

I agree on not including #1547 and #1588. There's some testing going on into whether or not they're still necessary after #1606. The only other possible PR to go in seems like #1406?



On Thursday, May 30, 2019 at 7:49:50 AM UTC-7, Oleksandr Porunov wrote:
Hi everyone,

I am going to make "0.3.2" release PR after conflicts in #1609 is resolved and the PR is merged.
After that I am going to finalize artifacts, docs and start a voting process. 
I have moved all opened issues from "0.3.2" milestone into "v0.3.x" milestone.

I have removed these two issues from milestones: #1547 and #1588 because they look more like features and I am not sure that they will be resolved soon (Of course I will be happy to review them when they are fixed). As I understand we re-targeted them from master into "0.3" because they are partially connected to the bugfix but on the other hand they are introducing a new parameter, so I don't know which branch they should be targeted. I am OK with both "0.3" and "master" branches.

Please, tell if we need to include something else into "0.3.2" release.
Estimated time for starting a release process is 1-2 days.

Best regards,
Oleksandr