Apache Ignite 2.8.1 RELEASE [Time, Scope, Manager]

[Maxim Muzafarov]

Igniters,


Let's start a discussion about the next Apache Ignite 2.8.1 bugfix
release. The following things need to be discussed:
 - time/deadlines
 - scope of the release
 - a release manager candidate


My personal vision that the 2.8.1 release is a bugfix release, so
`improvement` issues are not allowed. Anyway, it's up to the release
manager which issues can be included but I think the scope must be
discussed with the whole community prior to cherry-picking PRs.

I've prepared the wiki page [1] with the list of current bugs which
needs to be resolved and a release dates. The ignite-2.8.1 [2] branch
has been also created.


I propose the following dates for 2.8.1 release:
Scope Freeze: April 15, 2020
Code Freeze: April 22, 2020
Voting Date: April 27, 2020
Release Date: May 1, 2020


Please, share your thoughts.


[1] https://cwiki.apache.org/confluence/display/IGNITE/Apache+Ignite+2.8.1
[2] https://github.com/apache/ignite/tree/ignite-2.8.1

[Ilya Kasnacheev]

Hello!

I have marked https://issues.apache.org/jira/browse/IGNITE-12672 to 2.8.1
since it is an important Spring Data issue with trivial fix.

"Query annotations are not working if statement keywords are in lower case"

We should make sure that all fixes in 2.8.1 also end up in master - needs
to be re-checked after 2.8.1 freeze, perhaps.

Regards,
--
Ilya Kasnacheev


чт, 12 мар. 2020 г. в 17:05, Maxim Muzafarov <[hidden email]>:

> Igniters,
>
>
> Let's start a discussion about the next Apache Ignite 2.8.1 bugfix
> release. The following things need to be discussed:
>  - time/deadlines
>  - scope of the release
>  - a release manager candidate
>
>
> My personal vision that the 2.8.1 release is a bugfix release, so
> `improvement` issues are not allowed. Anyway, it's up to the release
> manager which issues can be included but I think the scope must be
> discussed with the whole community prior to cherry-picking PRs.
>
> I've prepared the wiki page [1] with the list of current bugs which
> needs to be resolved and a release dates. The ignite-2.8.1 [2] branch
> has been also created.
>
>
> I propose the following dates for 2.8.1 release:
> Scope Freeze: April 15, 2020
> Code Freeze: April 22, 2020
> Voting Date: April 27, 2020
> Release Date: May 1, 2020
>
>
> Please, share your thoughts.
>
>
> [1] https://cwiki.apache.org/confluence/display/IGNITE/Apache+Ignite+2.8.1
> [2] https://github.com/apache/ignite/tree/ignite-2.8.1
>

[schernolyas]

Hi!

I don't see  https://issues.apache.org/jira/browse/IGNITE-12582 in future
release plan. Append it too, please

On Thu, 12 Mar 2020 at 17:12, Ilya Kasnacheev <[hidden email]>
wrote:

> Hello!
>
> I have marked https://issues.apache.org/jira/browse/IGNITE-12672 to 2.8.1
> since it is an important Spring Data issue with trivial fix.
>
> "Query annotations are not working if statement keywords are in lower case"
>
> We should make sure that all fixes in 2.8.1 also end up in master - needs
> to be re-checked after 2.8.1 freeze, perhaps.
>
> Regards,
> --
> Ilya Kasnacheev
>
>
> чт, 12 мар. 2020 г. в 17:05, Maxim Muzafarov <[hidden email]>:
>
> > Igniters,
> >
> >
> > Let's start a discussion about the next Apache Ignite 2.8.1 bugfix
> > release. The following things need to be discussed:
> >  - time/deadlines
> >  - scope of the release
> >  - a release manager candidate
> >
> >
> > My personal vision that the 2.8.1 release is a bugfix release, so
> > `improvement` issues are not allowed. Anyway, it's up to the release
> > manager which issues can be included but I think the scope must be
> > discussed with the whole community prior to cherry-picking PRs.
> >
> > I've prepared the wiki page [1] with the list of current bugs which
> > needs to be resolved and a release dates. The ignite-2.8.1 [2] branch
> > has been also created.
> >
> >
> > I propose the following dates for 2.8.1 release:
> > Scope Freeze: April 15, 2020
> > Code Freeze: April 22, 2020
> > Voting Date: April 27, 2020
> > Release Date: May 1, 2020
> >
> >
> > Please, share your thoughts.
> >
> >
> > [1]
> https://cwiki.apache.org/confluence/display/IGNITE/Apache+Ignite+2.8.1
> > [2] https://github.com/apache/ignite/tree/ignite-2.8.1
> >
>

--
---------------------

With best regards, Sergey Chernolyas

[Nikolay Izhikov-2]

Hello, Igniters.

I propose to add these tickets to the 2.8.1 scope

[1] Disallow silent deactivation in CLI and REST - https://issues.apache.org/jira/browse/IGNITE-12701
[2] Compute job system view - https://issues.apache.org/jira/browse/IGNITE-12745
[3] Transaction operations metrics - https://issues.apache.org/jira/browse/IGNITE-12450

> 12 марта 2020 г., в 17:12, Ilya Kasnacheev <[hidden email]> написал(а):
>
> Hello!
>
> I have marked https://issues.apache.org/jira/browse/IGNITE-12672 to 2.8.1
> since it is an important Spring Data issue with trivial fix.
>
> "Query annotations are not working if statement keywords are in lower case"
>
> We should make sure that all fixes in 2.8.1 also end up in master - needs
> to be re-checked after 2.8.1 freeze, perhaps.
>
> Regards,
> --
> Ilya Kasnacheev
>
>
> чт, 12 мар. 2020 г. в 17:05, Maxim Muzafarov <[hidden email]>:
>
>> Igniters,
>>
>>
>> Let's start a discussion about the next Apache Ignite 2.8.1 bugfix
>> release. The following things need to be discussed:
>> - time/deadlines
>> - scope of the release
>> - a release manager candidate
>>
>>
>> My personal vision that the 2.8.1 release is a bugfix release, so
>> `improvement` issues are not allowed. Anyway, it's up to the release
>> manager which issues can be included but I think the scope must be
>> discussed with the whole community prior to cherry-picking PRs.
>>
>> I've prepared the wiki page [1] with the list of current bugs which
>> needs to be resolved and a release dates. The ignite-2.8.1 [2] branch
>> has been also created.
>>
>>
>> I propose the following dates for 2.8.1 release:
>> Scope Freeze: April 15, 2020
>> Code Freeze: April 22, 2020
>> Voting Date: April 27, 2020
>> Release Date: May 1, 2020
>>
>>
>> Please, share your thoughts.
>>
>>
>> [1] https://cwiki.apache.org/confluence/display/IGNITE/Apache+Ignite+2.8.1
>> [2] https://github.com/apache/ignite/tree/ignite-2.8.1
>>

[Aleksey Chetaev]

Hi, Igniters!
I propose to discuss immediately the timing of 2.9.0 and interim releases!
My propose:
~ June 01 - 2.8.1
~ September 01 - 2.8.2
~ November 01 - 2.9.0

Sberbank team, for its part, is ready to take on the work of management
interim releases.

Regards,
Aleksey Chetaev.





--
Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/

[vveider]

I thought the release cycle should be community driven, not single donator.
Should not 2.8.1 be bug fixing / stabilising release with release date as soon as all issues are resolved?

Also — what's the haste with next minor version? Cannot we deliver next fix versions with per-feature release cycle until something big is on horizon, and if it is already — why do we need 2.8.2 after 2.8.1 and not 2.9.0 along?

> On 12 Mar 2020, at 17:38, Aleksey Chetaev <[hidden email]> wrote:
>
> Hi, Igniters!
> I propose to discuss immediately the timing of 2.9.0 and interim releases!
> My propose:
> ~ June 01 - 2.8.1
> ~ September 01 - 2.8.2
> ~ November 01 - 2.9.0
>
> Sberbank team, for its part, is ready to take on the work of management
> interim releases.
>
> Regards,
> Aleksey Chetaev.
>
>
>
>
>
> --
> Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/

[Alexey Zinoviev]

Great dates for released, I support mr. Chetaev

чт, 12 мар. 2020 г., 17:38 Aleksey Chetaev <[hidden email]>:

> Hi, Igniters!
> I propose to discuss immediately the timing of 2.9.0 and interim releases!
> My propose:
> ~ June 01 - 2.8.1
> ~ September 01 - 2.8.2
> ~ November 01 - 2.9.0
>
> Sberbank team, for its part, is ready to take on the work of management
> interim releases.
>
> Regards,
> Aleksey Chetaev.
>
>
>
>
>
> --
> Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/
>

[Nikolay Izhikov-2]

Hello, Petr.

> I thought the release cycle should be community driven, not single donator.

We are the part of the global Ignite community, itn’t it?

If you know some reason to postpone bug fix releases, please, write it down.

> what's the haste with next minor version?

Users of the Ignite actively try 2.8 release and sent the bug reports.
You can find it on the user-list.
I think we all want to provide fixes as fast as we can.

So frequent releases is a good thing. Do you agree?

> 12 марта 2020 г., в 17:48, Petr Ivanov <[hidden email]> написал(а):
>
> I thought the release cycle should be community driven, not single donator.
> Should not 2.8.1 be bug fixing / stabilising release with release date as soon as all issues are resolved?
>
> Also — what's the haste with next minor version? Cannot we deliver next fix versions with per-feature release cycle until something big is on horizon, and if it is already — why do we need 2.8.2 after 2.8.1 and not 2.9.0 along?
>
>> On 12 Mar 2020, at 17:38, Aleksey Chetaev <[hidden email]> wrote:
>>
>> Hi, Igniters!
>> I propose to discuss immediately the timing of 2.9.0 and interim releases!
>> My propose:
>> ~ June 01 - 2.8.1
>> ~ September 01 - 2.8.2
>> ~ November 01 - 2.9.0
>>
>> Sberbank team, for its part, is ready to take on the work of management
>> interim releases.
>>
>> Regards,
>> Aleksey Chetaev.
>>
>>
>>
>>
>>
>> --
>> Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/
>

[Aleksey Chetaev]

I don’t see the donator desire here. This desire is to give AI consumers more
frequent releases with bug fix and understanding when we, as a community,
will release new, announced features to them.

Now it’s not clear why consumers should fix bugs themselves, if they don’t
have the opportunity to get the expected delivery time.



--
Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/

[Maxim Muzafarov]

Alexey,

> Now it’s not clear why consumers should fix bugs themselves if they don’t have the opportunity to get the expected delivery time.

Fully agree.
Users should see the release plan for the current year.

On Thu, 12 Mar 2020 at 18:16, Aleksey Chetaev <[hidden email]> wrote:

>
> I don’t see the donator desire here. This desire is to give AI consumers more
> frequent releases with bug fix and understanding when we, as a community,
> will release new, announced features to them.
>
> Now it’s not clear why consumers should fix bugs themselves, if they don’t
> have the opportunity to get the expected delivery time.
>
>
>
> --
> Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/

[Nikita Amelchev]

Hello, Igniters.

I propose to cherry-pick this bug-fix ticket to the 2.8.1:

IGNITE-12728 The cache#putAllAsync method does not collect statistics [1]

[1] https://issues.apache.org/jira/browse/IGNITE-12728

чт, 12 мар. 2020 г. в 18:27, Maxim Muzafarov <[hidden email]>:

>
> Alexey,
>
> > Now it’s not clear why consumers should fix bugs themselves if they don’t have the opportunity to get the expected delivery time.
>
> Fully agree.
> Users should see the release plan for the current year.
>
> On Thu, 12 Mar 2020 at 18:16, Aleksey Chetaev <[hidden email]> wrote:
> >
> > I don’t see the donator desire here. This desire is to give AI consumers more
> > frequent releases with bug fix and understanding when we, as a community,
> > will release new, announced features to them.
> >
> > Now it’s not clear why consumers should fix bugs themselves, if they don’t
> > have the opportunity to get the expected delivery time.
> >
> >
> >
> > --
> > Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/

--
Best wishes,
Amelchev Nikita

[dmagda]

Folks,

Not sure that we will be moving in the right direction discussing dates
with no clarity of what is to be released in a specific version. Instead,
let's talk out our plans/tasks first and then put them on a timeline.  That
certainly applies to big releases such as 2.9 that deserve a separate
discussion (we can initiate Ignite 2020 directions/roadmap thread and come
up with releases schedule once the community is an agreement).

As for the maintenance releases, such as 2.8.1, I would encourage us to
keep scheduling them based on necessity. For instance, we already see that
2.8 goes with several regressions and significant issues, which makes it
reasonable to plan through 2.8.1 for a ~month from now.


-
Denis


On Thu, Mar 12, 2020 at 7:38 AM Aleksey Chetaev <[hidden email]>
wrote:

> Hi, Igniters!
> I propose to discuss immediately the timing of 2.9.0 and interim releases!
> My propose:
> ~ June 01 - 2.8.1
> ~ September 01 - 2.8.2
> ~ November 01 - 2.9.0
>
> Sberbank team, for its part, is ready to take on the work of management
> interim releases.
>
> Regards,
> Aleksey Chetaev.
>
>
>
>
>
> --
> Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/
>

[agura]

Aleksey,

> My propose:
> ~ June 01 - 2.8.1
> ~ September 01 - 2.8.2
> ~ November 01 - 2.9.0

What is your motivation for such release schedule? I believe release
should be features/fixes driven not by time/deadline driven.

We can release 2.8.x versions every week yet (as maintenance
releases). But I don't see any reason for 2.9 release at this moment.

> Sberbank team, for its part, is ready to take on the work of management interim releases.

I am glad for your enthusiasm but I believe that release engineer role
should be rotated in order to share experience and expertise in the
release process between community members.

On Thu, Mar 12, 2020 at 6:43 PM Denis Magda <[hidden email]> wrote:

>
> Folks,
>
> Not sure that we will be moving in the right direction discussing dates
> with no clarity of what is to be released in a specific version. Instead,
> let's talk out our plans/tasks first and then put them on a timeline.  That
> certainly applies to big releases such as 2.9 that deserve a separate
> discussion (we can initiate Ignite 2020 directions/roadmap thread and come
> up with releases schedule once the community is an agreement).
>
> As for the maintenance releases, such as 2.8.1, I would encourage us to
> keep scheduling them based on necessity. For instance, we already see that
> 2.8 goes with several regressions and significant issues, which makes it
> reasonable to plan through 2.8.1 for a ~month from now.
>
>
> -
> Denis
>
>
> On Thu, Mar 12, 2020 at 7:38 AM Aleksey Chetaev <[hidden email]>
> wrote:
>
> > Hi, Igniters!
> > I propose to discuss immediately the timing of 2.9.0 and interim releases!
> > My propose:
> > ~ June 01 - 2.8.1
> > ~ September 01 - 2.8.2
> > ~ November 01 - 2.9.0
> >
> > Sberbank team, for its part, is ready to take on the work of management
> > interim releases.
> >
> > Regards,
> > Aleksey Chetaev.
> >
> >
> >
> >
> >
> > --
> > Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/
> >

[kpushenko]

Hi, igniters!

I would like to propose a discussion of version 2.9 and scope releases for
2020 to another discussion thread.

I understand that there is no objection to release 2.8.1.
I support the date from Maxim and dmagda ~ 1 month to May 1.

I offer my help with the release of Ignite 2.8.1. I can take on
administrative functions that do not require PMC: a scope release, a
reminder about ticket solutions, jobs without PMC, writing readme and
documentation. I am a new person in the community. But I have been testing
and supporting ignition-based products for the past three years. I want to
benefit the community.



--
Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/

[Nikolay Izhikov-2]

Hello, guys.

I propose myself as a release manager for 2.8.1.

> 16 марта 2020 г., в 14:19, kpushenko <[hidden email]> написал(а):
>
> Hi, igniters!
>
> I would like to propose a discussion of version 2.9 and scope releases for
> 2020 to another discussion thread.
>
> I understand that there is no objection to release 2.8.1.
> I support the date from Maxim and dmagda ~ 1 month to May 1.
>
> I offer my help with the release of Ignite 2.8.1. I can take on
> administrative functions that do not require PMC: a scope release, a
> reminder about ticket solutions, jobs without PMC, writing readme and
> documentation. I am a new person in the community. But I have been testing
> and supporting ignition-based products for the past three years. I want to
> benefit the community.
>
>
>
> --
> Sent from: http://apache-ignite-developers.2346864.n4.nabble.com/

[Alexey Goncharuk]

Folks,

I've walked through all the commits to master since 2.8 branch was cut and
filtered some tickets that in my opinion are worth including to 2.8.1
release below (note that they are ready end the effort of including them to
the release should be low as long as there are no implicit dependencies
between tickets). Please share your opinion on whether we should include
them to the 2.8.1.

IGNITE-12717 SQL: index creation refactoring
IGNITE-12590 MERGE INTO query is failing on Ignite client node
IGNITE-12671 Update of partition's states can stuck when rebalance
completed during exchange
IGNITE-11798 Memory leak on unstable topology caused by partition
reservation
IGNITE-12665 SQL: Potential race on MapResult close.
IGNITE-12605 Historical (WAL) rebalance can start on a cleared partition if
some baseline node leaves the cluster and then joins back.
IGNITE-12654 Some of rentingFutures in GridDhtPartitionTopologyImpl may
accumulate a huge number of eviction callbacks
IGNITE-12631 Incorrect rewriting wal record type in marshalled mode during
iteration
IGNITE-12621 Node leave may cause NullPointerException during IO message
processing if security is enabled
IGNITE-12636 Full rebalance instead of a historical one
IGNITE-12618 Affinity cache for version of last server event can be wiped
from history
IGNITE-12013 NullPointerException is thrown by ExchangeLatchManager during
cache creation
IGNITE-11797 Fix consistency issues for atomic and mixed tx-atomic cache
groups.
IGNITE-12557 Destroy of big cache which is not only cache in cache group
causes IgniteOOME
IGNITE-12567 H2Tree goes into illegal state when non-indexed columns are
dropped
IGNITE-12569 Can't set serialized enum to a BinaryObject's field
IGNITE-12460 Cluster fails to find the node by consistent ID
IGNITE-12459 Searching checkpoint record in WAL doesn't work with segment
compaction
IGNITE-12548 Possible tx desync during recovery on near node left.
IGNITE-12546 Prevent partitions owned by other nodes switch their state to
MOVING due to counter difference on node join.
IGNITE-12551 Partition desync if a partition is evicted then owned again
and historically rebalanced
IGNITE-12536 Inconsistency between cache data and indexes when cache
operation is interrupted
IGNITE-12403 Throttle page difference output in PageMemoryTracker
IGNITE-12523 Continuously generated thread dumps in failure processor slow
down the whole system
IGNITE-12489 Error during purges by expiration: Unknown page type

[Maxim Muzafarov]

Folks,

Can we add ignite-2.8.1 [2] branch under TC.Bot protection [1]?


[1] https://mtcga.gridgain.com/guard.html
[2] https://github.com/apache/ignite/tree/ignite-2.8.1

On Mon, 16 Mar 2020 at 16:32, Alexey Goncharuk
<[hidden email]> wrote:

>
> Folks,
>
> I've walked through all the commits to master since 2.8 branch was cut and
> filtered some tickets that in my opinion are worth including to 2.8.1
> release below (note that they are ready end the effort of including them to
> the release should be low as long as there are no implicit dependencies
> between tickets). Please share your opinion on whether we should include
> them to the 2.8.1.
>
> IGNITE-12717 SQL: index creation refactoring
> IGNITE-12590 MERGE INTO query is failing on Ignite client node
> IGNITE-12671 Update of partition's states can stuck when rebalance
> completed during exchange
> IGNITE-11798 Memory leak on unstable topology caused by partition
> reservation
> IGNITE-12665 SQL: Potential race on MapResult close.
> IGNITE-12605 Historical (WAL) rebalance can start on a cleared partition if
> some baseline node leaves the cluster and then joins back.
> IGNITE-12654 Some of rentingFutures in GridDhtPartitionTopologyImpl may
> accumulate a huge number of eviction callbacks
> IGNITE-12631 Incorrect rewriting wal record type in marshalled mode during
> iteration
> IGNITE-12621 Node leave may cause NullPointerException during IO message
> processing if security is enabled
> IGNITE-12636 Full rebalance instead of a historical one
> IGNITE-12618 Affinity cache for version of last server event can be wiped
> from history
> IGNITE-12013 NullPointerException is thrown by ExchangeLatchManager during
> cache creation
> IGNITE-11797 Fix consistency issues for atomic and mixed tx-atomic cache
> groups.
> IGNITE-12557 Destroy of big cache which is not only cache in cache group
> causes IgniteOOME
> IGNITE-12567 H2Tree goes into illegal state when non-indexed columns are
> dropped
> IGNITE-12569 Can't set serialized enum to a BinaryObject's field
> IGNITE-12460 Cluster fails to find the node by consistent ID
> IGNITE-12459 Searching checkpoint record in WAL doesn't work with segment
> compaction
> IGNITE-12548 Possible tx desync during recovery on near node left.
> IGNITE-12546 Prevent partitions owned by other nodes switch their state to
> MOVING due to counter difference on node join.
> IGNITE-12551 Partition desync if a partition is evicted then owned again
> and historically rebalanced
> IGNITE-12536 Inconsistency between cache data and indexes when cache
> operation is interrupted
> IGNITE-12403 Throttle page difference output in PageMemoryTracker
> IGNITE-12523 Continuously generated thread dumps in failure processor slow
> down the whole system
> IGNITE-12489 Error during purges by expiration: Unknown page type

[antonovsergey93]

Folks,

I'd like to add ticket IGNITE-12774 Transaction hangs after too many open
files NIO exception [1] to ignite-2.8.1 scope.

[1] https://issues.apache.org/jira/browse/IGNITE-12774

ср, 18 мар. 2020 г. в 16:53, Maxim Muzafarov <[hidden email]>:

> Folks,
>
> Can we add ignite-2.8.1 [2] branch under TC.Bot protection [1]?
>
>
> [1] https://mtcga.gridgain.com/guard.html
> [2] https://github.com/apache/ignite/tree/ignite-2.8.1
>
> On Mon, 16 Mar 2020 at 16:32, Alexey Goncharuk
> <[hidden email]> wrote:
> >
> > Folks,
> >
> > I've walked through all the commits to master since 2.8 branch was cut
> and
> > filtered some tickets that in my opinion are worth including to 2.8.1
> > release below (note that they are ready end the effort of including them
> to
> > the release should be low as long as there are no implicit dependencies
> > between tickets). Please share your opinion on whether we should include
> > them to the 2.8.1.
> >
> > IGNITE-12717 SQL: index creation refactoring
> > IGNITE-12590 MERGE INTO query is failing on Ignite client node
> > IGNITE-12671 Update of partition's states can stuck when rebalance
> > completed during exchange
> > IGNITE-11798 Memory leak on unstable topology caused by partition
> > reservation
> > IGNITE-12665 SQL: Potential race on MapResult close.
> > IGNITE-12605 Historical (WAL) rebalance can start on a cleared partition
> if
> > some baseline node leaves the cluster and then joins back.
> > IGNITE-12654 Some of rentingFutures in GridDhtPartitionTopologyImpl may
> > accumulate a huge number of eviction callbacks
> > IGNITE-12631 Incorrect rewriting wal record type in marshalled mode
> during
> > iteration
> > IGNITE-12621 Node leave may cause NullPointerException during IO message
> > processing if security is enabled
> > IGNITE-12636 Full rebalance instead of a historical one
> > IGNITE-12618 Affinity cache for version of last server event can be wiped
> > from history
> > IGNITE-12013 NullPointerException is thrown by ExchangeLatchManager
> during
> > cache creation
> > IGNITE-11797 Fix consistency issues for atomic and mixed tx-atomic cache
> > groups.
> > IGNITE-12557 Destroy of big cache which is not only cache in cache group
> > causes IgniteOOME
> > IGNITE-12567 H2Tree goes into illegal state when non-indexed columns are
> > dropped
> > IGNITE-12569 Can't set serialized enum to a BinaryObject's field
> > IGNITE-12460 Cluster fails to find the node by consistent ID
> > IGNITE-12459 Searching checkpoint record in WAL doesn't work with segment
> > compaction
> > IGNITE-12548 Possible tx desync during recovery on near node left.
> > IGNITE-12546 Prevent partitions owned by other nodes switch their state
> to
> > MOVING due to counter difference on node join.
> > IGNITE-12551 Partition desync if a partition is evicted then owned again
> > and historically rebalanced
> > IGNITE-12536 Inconsistency between cache data and indexes when cache
> > operation is interrupted
> > IGNITE-12403 Throttle page difference output in PageMemoryTracker
> > IGNITE-12523 Continuously generated thread dumps in failure processor
> slow
> > down the whole system
> > IGNITE-12489 Error during purges by expiration: Unknown page type
>

--
BR, Sergey Antonov

[dmagda]

Igniters,

As long as 2.8.1 is inevitable and we already keep adding critical issues
to the working queue, let's settle on the release time frames and decide
who will be a release manager. This is the time proposed by Maxim and,
personally, I concur with such a schedule:

   - Scope Freeze: April 15, 2020
   - Code Freeze: April 22, 2020
   - Voting Date: April 27, 2020
   - Release Date: May 1, 2020

Do we agree on this time? Is there anybody who ready to drive the release
as a release manager?

-
Denis


On Thu, Mar 19, 2020 at 5:50 AM Sergey Antonov <[hidden email]>
wrote:

> Folks,
>
> I'd like to add ticket IGNITE-12774 Transaction hangs after too many open
> files NIO exception [1] to ignite-2.8.1 scope.
>
> [1] https://issues.apache.org/jira/browse/IGNITE-12774
>
> ср, 18 мар. 2020 г. в 16:53, Maxim Muzafarov <[hidden email]>:
>
> > Folks,
> >
> > Can we add ignite-2.8.1 [2] branch under TC.Bot protection [1]?
> >
> >
> > [1] https://mtcga.gridgain.com/guard.html
> > [2] https://github.com/apache/ignite/tree/ignite-2.8.1
> >
> > On Mon, 16 Mar 2020 at 16:32, Alexey Goncharuk
> > <[hidden email]> wrote:
> > >
> > > Folks,
> > >
> > > I've walked through all the commits to master since 2.8 branch was cut
> > and
> > > filtered some tickets that in my opinion are worth including to 2.8.1
> > > release below (note that they are ready end the effort of including
> them
> > to
> > > the release should be low as long as there are no implicit dependencies
> > > between tickets). Please share your opinion on whether we should
> include
> > > them to the 2.8.1.
> > >
> > > IGNITE-12717 SQL: index creation refactoring
> > > IGNITE-12590 MERGE INTO query is failing on Ignite client node
> > > IGNITE-12671 Update of partition's states can stuck when rebalance
> > > completed during exchange
> > > IGNITE-11798 Memory leak on unstable topology caused by partition
> > > reservation
> > > IGNITE-12665 SQL: Potential race on MapResult close.
> > > IGNITE-12605 Historical (WAL) rebalance can start on a cleared
> partition
> > if
> > > some baseline node leaves the cluster and then joins back.
> > > IGNITE-12654 Some of rentingFutures in GridDhtPartitionTopologyImpl may
> > > accumulate a huge number of eviction callbacks
> > > IGNITE-12631 Incorrect rewriting wal record type in marshalled mode
> > during
> > > iteration
> > > IGNITE-12621 Node leave may cause NullPointerException during IO
> message
> > > processing if security is enabled
> > > IGNITE-12636 Full rebalance instead of a historical one
> > > IGNITE-12618 Affinity cache for version of last server event can be
> wiped
> > > from history
> > > IGNITE-12013 NullPointerException is thrown by ExchangeLatchManager
> > during
> > > cache creation
> > > IGNITE-11797 Fix consistency issues for atomic and mixed tx-atomic
> cache
> > > groups.
> > > IGNITE-12557 Destroy of big cache which is not only cache in cache
> group
> > > causes IgniteOOME
> > > IGNITE-12567 H2Tree goes into illegal state when non-indexed columns
> are
> > > dropped
> > > IGNITE-12569 Can't set serialized enum to a BinaryObject's field
> > > IGNITE-12460 Cluster fails to find the node by consistent ID
> > > IGNITE-12459 Searching checkpoint record in WAL doesn't work with
> segment
> > > compaction
> > > IGNITE-12548 Possible tx desync during recovery on near node left.
> > > IGNITE-12546 Prevent partitions owned by other nodes switch their state
> > to
> > > MOVING due to counter difference on node join.
> > > IGNITE-12551 Partition desync if a partition is evicted then owned
> again
> > > and historically rebalanced
> > > IGNITE-12536 Inconsistency between cache data and indexes when cache
> > > operation is interrupted
> > > IGNITE-12403 Throttle page difference output in PageMemoryTracker
> > > IGNITE-12523 Continuously generated thread dumps in failure processor
> > slow
> > > down the whole system
> > > IGNITE-12489 Error during purges by expiration: Unknown page type
> >
>
>
> --
> BR, Sergey Antonov
>

[Nikolay Izhikov-2]

I want to be a release manager.

> 19 марта 2020 г., в 17:52, Denis Magda <[hidden email]> написал(а):
>
> Igniters,
>
> As long as 2.8.1 is inevitable and we already keep adding critical issues
> to the working queue, let's settle on the release time frames and decide
> who will be a release manager. This is the time proposed by Maxim and,
> personally, I concur with such a schedule:
>
>   - Scope Freeze: April 15, 2020
>   - Code Freeze: April 22, 2020
>   - Voting Date: April 27, 2020
>   - Release Date: May 1, 2020
>
> Do we agree on this time? Is there anybody who ready to drive the release
> as a release manager?
>
> -
> Denis
>
>
> On Thu, Mar 19, 2020 at 5:50 AM Sergey Antonov <[hidden email]>
> wrote:
>
>> Folks,
>>
>> I'd like to add ticket IGNITE-12774 Transaction hangs after too many open
>> files NIO exception [1] to ignite-2.8.1 scope.
>>
>> [1] https://issues.apache.org/jira/browse/IGNITE-12774
>>
>> ср, 18 мар. 2020 г. в 16:53, Maxim Muzafarov <[hidden email]>:
>>
>>> Folks,
>>>
>>> Can we add ignite-2.8.1 [2] branch under TC.Bot protection [1]?
>>>
>>>
>>> [1] https://mtcga.gridgain.com/guard.html
>>> [2] https://github.com/apache/ignite/tree/ignite-2.8.1
>>>
>>> On Mon, 16 Mar 2020 at 16:32, Alexey Goncharuk
>>> <[hidden email]> wrote:
>>>>
>>>> Folks,
>>>>
>>>> I've walked through all the commits to master since 2.8 branch was cut
>>> and
>>>> filtered some tickets that in my opinion are worth including to 2.8.1
>>>> release below (note that they are ready end the effort of including
>> them
>>> to
>>>> the release should be low as long as there are no implicit dependencies
>>>> between tickets). Please share your opinion on whether we should
>> include
>>>> them to the 2.8.1.
>>>>
>>>> IGNITE-12717 SQL: index creation refactoring
>>>> IGNITE-12590 MERGE INTO query is failing on Ignite client node
>>>> IGNITE-12671 Update of partition's states can stuck when rebalance
>>>> completed during exchange
>>>> IGNITE-11798 Memory leak on unstable topology caused by partition
>>>> reservation
>>>> IGNITE-12665 SQL: Potential race on MapResult close.
>>>> IGNITE-12605 Historical (WAL) rebalance can start on a cleared
>> partition
>>> if
>>>> some baseline node leaves the cluster and then joins back.
>>>> IGNITE-12654 Some of rentingFutures in GridDhtPartitionTopologyImpl may
>>>> accumulate a huge number of eviction callbacks
>>>> IGNITE-12631 Incorrect rewriting wal record type in marshalled mode
>>> during
>>>> iteration
>>>> IGNITE-12621 Node leave may cause NullPointerException during IO
>> message
>>>> processing if security is enabled
>>>> IGNITE-12636 Full rebalance instead of a historical one
>>>> IGNITE-12618 Affinity cache for version of last server event can be
>> wiped
>>>> from history
>>>> IGNITE-12013 NullPointerException is thrown by ExchangeLatchManager
>>> during
>>>> cache creation
>>>> IGNITE-11797 Fix consistency issues for atomic and mixed tx-atomic
>> cache
>>>> groups.
>>>> IGNITE-12557 Destroy of big cache which is not only cache in cache
>> group
>>>> causes IgniteOOME
>>>> IGNITE-12567 H2Tree goes into illegal state when non-indexed columns
>> are
>>>> dropped
>>>> IGNITE-12569 Can't set serialized enum to a BinaryObject's field
>>>> IGNITE-12460 Cluster fails to find the node by consistent ID
>>>> IGNITE-12459 Searching checkpoint record in WAL doesn't work with
>> segment
>>>> compaction
>>>> IGNITE-12548 Possible tx desync during recovery on near node left.
>>>> IGNITE-12546 Prevent partitions owned by other nodes switch their state
>>> to
>>>> MOVING due to counter difference on node join.
>>>> IGNITE-12551 Partition desync if a partition is evicted then owned
>> again
>>>> and historically rebalanced
>>>> IGNITE-12536 Inconsistency between cache data and indexes when cache
>>>> operation is interrupted
>>>> IGNITE-12403 Throttle page difference output in PageMemoryTracker
>>>> IGNITE-12523 Continuously generated thread dumps in failure processor
>>> slow
>>>> down the whole system
>>>> IGNITE-12489 Error during purges by expiration: Unknown page type
>>>
>>
>>
>> --
>> BR, Sergey Antonov
>>