Re: [DISCUSS] Harry in-tree

[Aleksey Yeshchenko]

+1

> On 29 Nov 2023, at 07:23, Marcus Eriksson  wrote:
> 
> +1!
> 
> /Marcus
> 
> On Fri, Nov 24, 2023 at 04:43:36PM +0100, Alex Petrov wrote:
>> Hi everyone,
>> 
>> With TCM landed, there will be way more Harry tests in-tree: we are using it 
>> for many coordination tests, and there's now a simulator test that uses 
>> Harry. During development, Harry has allowed us to uncover and resolve 
>> numerous elusive edge cases.
>> 
>> I had conversations with several folks, and wanted to propose to move 
>> harry-core to Cassandra test tree. This will substantially 
>> simplify/streamline co-development of Cassandra and Harry. With a new 
>> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
>> will also be much more approachable.
>> 
>> Besides making it easier for everyone to develop new fuzz tests, it will 
>> also substantially lower the barrier to entry. Currently, debugging an issue 
>> found by Harry involves a cumbersome process of rebuilding and transferring 
>> jars between Cassandra and Harry, depending on which side you modify. This 
>> not only hampers efficiency but also deters broader adoption. By merging 
>> harry-core into the Cassandra test tree, we eliminate this barrier.
>> 
>> Thank you,
>> --Alex
>> 
>> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
>> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
>> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Marcus Eriksson]

+1!

/Marcus

On Fri, Nov 24, 2023 at 04:43:36PM +0100, Alex Petrov wrote:
> Hi everyone,
> 
> With TCM landed, there will be way more Harry tests in-tree: we are using it 
> for many coordination tests, and there's now a simulator test that uses 
> Harry. During development, Harry has allowed us to uncover and resolve 
> numerous elusive edge cases.
> 
> I had conversations with several folks, and wanted to propose to move 
> harry-core to Cassandra test tree. This will substantially 
> simplify/streamline co-development of Cassandra and Harry. With a new 
> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
> will also be much more approachable.
> 
> Besides making it easier for everyone to develop new fuzz tests, it will also 
> substantially lower the barrier to entry. Currently, debugging an issue found 
> by Harry involves a cumbersome process of rebuilding and transferring jars 
> between Cassandra and Harry, depending on which side you modify. This not 
> only hampers efficiency but also deters broader adoption. By merging 
> harry-core into the Cassandra test tree, we eliminate this barrier.
> 
> Thank you,
> --Alex
> 
> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Abe Ratnofsky]

Another strong +1 to have Harry in-tree, and another +1 to building shaded 
dtest JARs on release. There are a number of projects that would benefit from 
having these JARs available in a central repository, like Sidecar, Driver, etc. 
I didn't see a ticket so created one: 
https://issues.apache.org/jira/browse/CASSANDRA-19113

> On Nov 28, 2023, at 3:12 AM, Alex Petrov  wrote:
> 
> Sure, that should be possible. I will check and will get back.
> 
> On Mon, Nov 27, 2023, at 10:10 PM, Ekaterina Dimitrova wrote:
>> +1, also, Alex, just an idea - maybe you want to make a virtual talk, as 
>> part of the contributors meetings? 
>> 
>> 
>> На понеделник, 27 ноември 2023 г. Yifan Cai > <mailto:yc25c...@gmail.com>> написа:
>> +1
>> 
>> 
>> 发件人: Sam Tunnicliffe mailto:s...@beobal.com>>
>> 发送时间: Tuesday, November 28, 2023 2:43:51 AM
>> 收件人: dev mailto:dev@cassandra.apache.org>>
>> 主题: Re: [DISCUSS] Harry in-tree
>>  
>> Definite +1 to bringing harry-core in tree.
>> 
>>> On 24 Nov 2023, at 15:43, Alex Petrov >> <mailto:al...@coffeenco.de>> wrote:
>>> 
>>> Hi everyone,
>>> 
>>> With TCM landed, there will be way more Harry tests in-tree: we are using 
>>> it for many coordination tests, and there's now a simulator test that uses 
>>> Harry. During development, Harry has allowed us to uncover and resolve 
>>> numerous elusive edge cases.
>>> 
>>> I had conversations with several folks, and wanted to propose to move 
>>> harry-core to Cassandra test tree. This will substantially 
>>> simplify/streamline co-development of Cassandra and Harry. With a new 
>>> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
>>> will also be much more approachable.
>>> 
>>> Besides making it easier for everyone to develop new fuzz tests, it will 
>>> also substantially lower the barrier to entry. Currently, debugging an 
>>> issue found by Harry involves a cumbersome process of rebuilding and 
>>> transferring jars between Cassandra and Harry, depending on which side you 
>>> modify. This not only hampers efficiency but also deters broader adoption. 
>>> By merging harry-core into the Cassandra test tree, we eliminate this 
>>> barrier.
>>> 
>>> Thank you,
>>> --Alex
>>> 
>>> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
>>> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
>>> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Alex Petrov]

Sure, that should be possible. I will check and will get back.

On Mon, Nov 27, 2023, at 10:10 PM, Ekaterina Dimitrova wrote:
> +1, also, Alex, just an idea - maybe you want to make a virtual talk, as part 
> of the contributors meetings? 
> 
> 
> На понеделник, 27 ноември 2023 г. Yifan Cai  написа:
>> +1
>> 
>> 
>> *发件人:* Sam Tunnicliffe 
>> *发送时间:* Tuesday, November 28, 2023 2:43:51 AM
>> *收件人:* dev 
>> *主题:* Re: [DISCUSS] Harry in-tree
>>  
>> Definite +1 to bringing harry-core in tree.
>> 
>>> On 24 Nov 2023, at 15:43, Alex Petrov  wrote:
>>> 
>>> Hi everyone,
>>> 
>>> With TCM landed, there will be way more Harry tests in-tree: we are using 
>>> it for many coordination tests, and there's now a simulator test that uses 
>>> Harry. During development, Harry has allowed us to uncover and resolve 
>>> numerous elusive edge cases.
>>> 
>>> I had conversations with several folks, and wanted to propose to move 
>>> harry-core to Cassandra test tree. This will substantially 
>>> simplify/streamline co-development of Cassandra and Harry. With a new 
>>> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
>>> will also be much more approachable.
>>> 
>>> Besides making it easier for everyone to develop new fuzz tests, it will 
>>> also substantially lower the barrier to entry. Currently, debugging an 
>>> issue found by Harry involves a cumbersome process of rebuilding and 
>>> transferring jars between Cassandra and Harry, depending on which side you 
>>> modify. This not only hampers efficiency but also deters broader adoption. 
>>> By merging harry-core into the Cassandra test tree, we eliminate this 
>>> barrier.
>>> 
>>> Thank you,
>>> --Alex
>>> 
>>> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
>>> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
>>> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Jacek Lewandowski]

+1

- - -- --- -  -
Jacek Lewandowski


pon., 27 lis 2023 o 22:11 Ekaterina Dimitrova 
napisał(a):

> +1, also, Alex, just an idea - maybe you want to make a virtual talk, as
> part of the contributors meetings?
>
>
> На понеделник, 27 ноември 2023 г. Yifan Cai  написа:
>
>> +1
>> --
>> *发件人:* Sam Tunnicliffe 
>> *发送时间:* Tuesday, November 28, 2023 2:43:51 AM
>> *收件人:* dev 
>> *主题:* Re: [DISCUSS] Harry in-tree
>>
>> Definite +1 to bringing harry-core in tree.
>>
>> On 24 Nov 2023, at 15:43, Alex Petrov  wrote:
>>
>> Hi everyone,
>>
>> With TCM landed, there will be way more Harry tests in-tree: we are using
>> it for many coordination tests, and there's now a simulator test that uses
>> Harry. During development, Harry has allowed us to uncover and resolve
>> numerous elusive edge cases.
>>
>> I had conversations with several folks, and wanted to propose to move
>> harry-core to Cassandra test tree. This will substantially
>> simplify/streamline co-development of Cassandra and Harry. With a new
>> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it
>> will also be much more approachable.
>>
>> Besides making it easier for everyone to develop new fuzz tests, it will
>> also substantially lower the barrier to entry. Currently, debugging an
>> issue found by Harry involves a cumbersome process of rebuilding and
>> transferring jars between Cassandra and Harry, depending on which side you
>> modify. This not only hampers efficiency but also deters broader adoption.
>> By merging harry-core into the Cassandra test tree, we eliminate this
>> barrier.
>>
>> Thank you,
>> --Alex
>>
>> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
>> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
>> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932
>>
>>
>>

Re: [DISCUSS] Harry in-tree

[Ekaterina Dimitrova]

+1, also, Alex, just an idea - maybe you want to make a virtual talk, as
part of the contributors meetings?


На понеделник, 27 ноември 2023 г. Yifan Cai  написа:

> +1
> --
> *发件人:* Sam Tunnicliffe 
> *发送时间:* Tuesday, November 28, 2023 2:43:51 AM
> *收件人:* dev 
> *主题:* Re: [DISCUSS] Harry in-tree
>
> Definite +1 to bringing harry-core in tree.
>
> On 24 Nov 2023, at 15:43, Alex Petrov  wrote:
>
> Hi everyone,
>
> With TCM landed, there will be way more Harry tests in-tree: we are using
> it for many coordination tests, and there's now a simulator test that uses
> Harry. During development, Harry has allowed us to uncover and resolve
> numerous elusive edge cases.
>
> I had conversations with several folks, and wanted to propose to move
> harry-core to Cassandra test tree. This will substantially
> simplify/streamline co-development of Cassandra and Harry. With a new
> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it
> will also be much more approachable.
>
> Besides making it easier for everyone to develop new fuzz tests, it will
> also substantially lower the barrier to entry. Currently, debugging an
> issue found by Harry involves a cumbersome process of rebuilding and
> transferring jars between Cassandra and Harry, depending on which side you
> modify. This not only hampers efficiency but also deters broader adoption.
> By merging harry-core into the Cassandra test tree, we eliminate this
> barrier.
>
> Thank you,
> --Alex
>
> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932
>
>
>

Re: [DISCUSS] Harry in-tree

[Yifan Cai]

+1

发件人: Sam Tunnicliffe 
发送时间: Tuesday, November 28, 2023 2:43:51 AM
收件人: dev 
主题: Re: [DISCUSS] Harry in-tree

Definite +1 to bringing harry-core in tree.

On 24 Nov 2023, at 15:43, Alex Petrov  wrote:

Hi everyone,

With TCM landed, there will be way more Harry tests in-tree: we are using it 
for many coordination tests, and there's now a simulator test that uses Harry. 
During development, Harry has allowed us to uncover and resolve numerous 
elusive edge cases.

I had conversations with several folks, and wanted to propose to move 
harry-core to Cassandra test tree. This will substantially simplify/streamline 
co-development of Cassandra and Harry. With a new HistoryBuilder API that has 
helped to find and trigger [1] [2] and [3], it will also be much more 
approachable.

Besides making it easier for everyone to develop new fuzz tests, it will also 
substantially lower the barrier to entry. Currently, debugging an issue found 
by Harry involves a cumbersome process of rebuilding and transferring jars 
between Cassandra and Harry, depending on which side you modify. This not only 
hampers efficiency but also deters broader adoption. By merging harry-core into 
the Cassandra test tree, we eliminate this barrier.

Thank you,
--Alex

[1] https://issues.apache.org/jira/browse/CASSANDRA-19011
[2] https://issues.apache.org/jira/browse/CASSANDRA-18993
[3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Sam Tunnicliffe]

Definite +1 to bringing harry-core in tree.

> On 24 Nov 2023, at 15:43, Alex Petrov  wrote:
> 
> Hi everyone,
> 
> With TCM landed, there will be way more Harry tests in-tree: we are using it 
> for many coordination tests, and there's now a simulator test that uses 
> Harry. During development, Harry has allowed us to uncover and resolve 
> numerous elusive edge cases.
> 
> I had conversations with several folks, and wanted to propose to move 
> harry-core to Cassandra test tree. This will substantially 
> simplify/streamline co-development of Cassandra and Harry. With a new 
> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
> will also be much more approachable.
> 
> Besides making it easier for everyone to develop new fuzz tests, it will also 
> substantially lower the barrier to entry. Currently, debugging an issue found 
> by Harry involves a cumbersome process of rebuilding and transferring jars 
> between Cassandra and Harry, depending on which side you modify. This not 
> only hampers efficiency but also deters broader adoption. By merging 
> harry-core into the Cassandra test tree, we eliminate this barrier.
> 
> Thank you,
> --Alex
> 
> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Francisco Guerrero]

+1 (nb)

On 2023/11/24 16:32:33 Alex Petrov wrote:

> > (We can publish jar files at C* release time if there's a call for this, 
> > doesn't really matter if they don't contain changes.). 
> 
> I would say we should publish dtest jar files when releasing, this would be 
> very helpful! 

Yes, projects in the Cassandra ecosystem would benefit a lot if we started 
publishing dtest jars. I am a big +1 on this as well
 
> 
> 
> 
> On Fri, Nov 24, 2023, at 5:23 PM, Mick Semb Wever wrote:
> > 
> >  
> >  
> >> I had conversations with several folks, and wanted to propose to move 
> >> harry-core to Cassandra test tree. This will substantially 
> >> simplify/streamline co-development of Cassandra and Harry. With a new 
> >> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
> >> will also be much more approachable.
> > 
> > 
> > Yes.  Any reason to have releases for users outside of Cassandra ? (We can 
> > publish jar files at C* release time if there's a call for this, doesn't 
> > really matter if they don't contain changes.).  And is there any 
> > cross-branch/version interaction, like there is with the jvm-dtest-api ?
> > 
> > 
> >  
> 

Re: [DISCUSS] Harry in-tree

[David Capwell]

+1 to in-tree

> On Nov 27, 2023, at 9:17 AM, Benjamin Lerer  wrote:
> 
> +1
> 
> Le lun. 27 nov. 2023 à 18:01, Brandon Williams  > a écrit :
>> I am +1 on including Harry in-tree.
>> 
>> Kind Regards,
>> Brandon
>> 
>> On Fri, Nov 24, 2023 at 9:44 AM Alex Petrov > > wrote:
>> >
>> > Hi everyone,
>> >
>> > With TCM landed, there will be way more Harry tests in-tree: we are using 
>> > it for many coordination tests, and there's now a simulator test that uses 
>> > Harry. During development, Harry has allowed us to uncover and resolve 
>> > numerous elusive edge cases.
>> >
>> > I had conversations with several folks, and wanted to propose to move 
>> > harry-core to Cassandra test tree. This will substantially 
>> > simplify/streamline co-development of Cassandra and Harry. With a new 
>> > HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
>> > will also be much more approachable.
>> >
>> > Besides making it easier for everyone to develop new fuzz tests, it will 
>> > also substantially lower the barrier to entry. Currently, debugging an 
>> > issue found by Harry involves a cumbersome process of rebuilding and 
>> > transferring jars between Cassandra and Harry, depending on which side you 
>> > modify. This not only hampers efficiency but also deters broader adoption. 
>> > By merging harry-core into the Cassandra test tree, we eliminate this 
>> > barrier.
>> >
>> > Thank you,
>> > --Alex
>> >
>> > [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
>> > [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
>> > [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Benjamin Lerer]

+1

Le lun. 27 nov. 2023 à 18:01, Brandon Williams  a écrit :

> I am +1 on including Harry in-tree.
>
> Kind Regards,
> Brandon
>
> On Fri, Nov 24, 2023 at 9:44 AM Alex Petrov  wrote:
> >
> > Hi everyone,
> >
> > With TCM landed, there will be way more Harry tests in-tree: we are
> using it for many coordination tests, and there's now a simulator test that
> uses Harry. During development, Harry has allowed us to uncover and resolve
> numerous elusive edge cases.
> >
> > I had conversations with several folks, and wanted to propose to move
> harry-core to Cassandra test tree. This will substantially
> simplify/streamline co-development of Cassandra and Harry. With a new
> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it
> will also be much more approachable.
> >
> > Besides making it easier for everyone to develop new fuzz tests, it will
> also substantially lower the barrier to entry. Currently, debugging an
> issue found by Harry involves a cumbersome process of rebuilding and
> transferring jars between Cassandra and Harry, depending on which side you
> modify. This not only hampers efficiency but also deters broader adoption.
> By merging harry-core into the Cassandra test tree, we eliminate this
> barrier.
> >
> > Thank you,
> > --Alex
> >
> > [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
> > [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
> > [3] https://issues.apache.org/jira/browse/CASSANDRA-18932
>

Re: [DISCUSS] Harry in-tree

[Brandon Williams]

I am +1 on including Harry in-tree.

Kind Regards,
Brandon

On Fri, Nov 24, 2023 at 9:44 AM Alex Petrov  wrote:
>
> Hi everyone,
>
> With TCM landed, there will be way more Harry tests in-tree: we are using it 
> for many coordination tests, and there's now a simulator test that uses 
> Harry. During development, Harry has allowed us to uncover and resolve 
> numerous elusive edge cases.
>
> I had conversations with several folks, and wanted to propose to move 
> harry-core to Cassandra test tree. This will substantially 
> simplify/streamline co-development of Cassandra and Harry. With a new 
> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
> will also be much more approachable.
>
> Besides making it easier for everyone to develop new fuzz tests, it will also 
> substantially lower the barrier to entry. Currently, debugging an issue found 
> by Harry involves a cumbersome process of rebuilding and transferring jars 
> between Cassandra and Harry, depending on which side you modify. This not 
> only hampers efficiency but also deters broader adoption. By merging 
> harry-core into the Cassandra test tree, we eliminate this barrier.
>
> Thank you,
> --Alex
>
> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Josh McKenzie]

Strong +1 to including harry in-tree and further, integrating a harry stress 
soak into our pre-commit and post-commit CI.

On Fri, Nov 24, 2023, at 5:10 PM, Alex Petrov wrote:
> Unfortunately my Harry talk got declined. Of course I’ll be happy to talk 
> about Harry and how it can be useful for contributors and about people’s 
> expectations. My talk is going to be about TCM again this time.
> 
> I will make sure examples are in place and are expressive by the summit.
> 
> On Fri, Nov 24, 2023, at 6:18 PM, Jeremy Hanna wrote:
>> I'm excited for Harry to come in-tree to improve the project stability and 
>> quality.  I know you're doing a talk at the Cassandra Summit about Harry to 
>> go over it.  If there's anything that can be done as part of this process to 
>> improve onboarding for Harry too, that would be great.  I'm just thinking 
>> about examples and things like that so people new to Harry can more easily 
>> write and run tests, test new features, and have a standard process for 
>> reporting findings.
>> 
>> Thanks Alex and all involved!
>> 
>> Jeremy
>> 
>>> On Nov 24, 2023, at 9:43 AM, Alex Petrov  wrote:
>>> 
>>> Hi everyone,
>>> 
>>> With TCM landed, there will be way more Harry tests in-tree: we are using 
>>> it for many coordination tests, and there's now a simulator test that uses 
>>> Harry. During development, Harry has allowed us to uncover and resolve 
>>> numerous elusive edge cases.
>>> 
>>> I had conversations with several folks, and wanted to propose to move 
>>> harry-core to Cassandra test tree. This will substantially 
>>> simplify/streamline co-development of Cassandra and Harry. With a new 
>>> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
>>> will also be much more approachable.
>>> 
>>> Besides making it easier for everyone to develop new fuzz tests, it will 
>>> also substantially lower the barrier to entry. Currently, debugging an 
>>> issue found by Harry involves a cumbersome process of rebuilding and 
>>> transferring jars between Cassandra and Harry, depending on which side you 
>>> modify. This not only hampers efficiency but also deters broader adoption. 
>>> By merging harry-core into the Cassandra test tree, we eliminate this 
>>> barrier.
>>> 
>>> Thank you,
>>> --Alex
>>> 
>>> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
>>> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
>>> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932
> 

Re: [DISCUSS] Harry in-tree

[Alex Petrov]

Unfortunately my Harry talk got declined. Of course I’ll be happy to talk about 
Harry and how it can be useful for contributors and about people’s 
expectations. My talk is going to be about TCM again this time.

I will make sure examples are in place and are expressive by the summit.

On Fri, Nov 24, 2023, at 6:18 PM, Jeremy Hanna wrote:
> I'm excited for Harry to come in-tree to improve the project stability and 
> quality.  I know you're doing a talk at the Cassandra Summit about Harry to 
> go over it.  If there's anything that can be done as part of this process to 
> improve onboarding for Harry too, that would be great.  I'm just thinking 
> about examples and things like that so people new to Harry can more easily 
> write and run tests, test new features, and have a standard process for 
> reporting findings.
> 
> Thanks Alex and all involved!
> 
> Jeremy
> 
>> On Nov 24, 2023, at 9:43 AM, Alex Petrov  wrote:
>> 
>> Hi everyone,
>> 
>> With TCM landed, there will be way more Harry tests in-tree: we are using it 
>> for many coordination tests, and there's now a simulator test that uses 
>> Harry. During development, Harry has allowed us to uncover and resolve 
>> numerous elusive edge cases.
>> 
>> I had conversations with several folks, and wanted to propose to move 
>> harry-core to Cassandra test tree. This will substantially 
>> simplify/streamline co-development of Cassandra and Harry. With a new 
>> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
>> will also be much more approachable.
>> 
>> Besides making it easier for everyone to develop new fuzz tests, it will 
>> also substantially lower the barrier to entry. Currently, debugging an issue 
>> found by Harry involves a cumbersome process of rebuilding and transferring 
>> jars between Cassandra and Harry, depending on which side you modify. This 
>> not only hampers efficiency but also deters broader adoption. By merging 
>> harry-core into the Cassandra test tree, we eliminate this barrier.
>> 
>> Thank you,
>> --Alex
>> 
>> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
>> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
>> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Jeremy Hanna]

I'm excited for Harry to come in-tree to improve the project stability and 
quality.  I know you're doing a talk at the Cassandra Summit about Harry to go 
over it.  If there's anything that can be done as part of this process to 
improve onboarding for Harry too, that would be great.  I'm just thinking about 
examples and things like that so people new to Harry can more easily write and 
run tests, test new features, and have a standard process for reporting 
findings.

Thanks Alex and all involved!

Jeremy

> On Nov 24, 2023, at 9:43 AM, Alex Petrov  wrote:
> 
> Hi everyone,
> 
> With TCM landed, there will be way more Harry tests in-tree: we are using it 
> for many coordination tests, and there's now a simulator test that uses 
> Harry. During development, Harry has allowed us to uncover and resolve 
> numerous elusive edge cases.
> 
> I had conversations with several folks, and wanted to propose to move 
> harry-core to Cassandra test tree. This will substantially 
> simplify/streamline co-development of Cassandra and Harry. With a new 
> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
> will also be much more approachable.
> 
> Besides making it easier for everyone to develop new fuzz tests, it will also 
> substantially lower the barrier to entry. Currently, debugging an issue found 
> by Harry involves a cumbersome process of rebuilding and transferring jars 
> between Cassandra and Harry, depending on which side you modify. This not 
> only hampers efficiency but also deters broader adoption. By merging 
> harry-core into the Cassandra test tree, we eliminate this barrier.
> 
> Thank you,
> --Alex
> 
> [1] https://issues.apache.org/jira/browse/CASSANDRA-19011
> [2] https://issues.apache.org/jira/browse/CASSANDRA-18993
> [3] https://issues.apache.org/jira/browse/CASSANDRA-18932

Re: [DISCUSS] Harry in-tree

[Alex Petrov]

> Yes.  Any reason to have releases for users outside of Cassandra ?  And is 
> there any cross-branch/version interaction, like there is with the 
> jvm-dtest-api ?

As of today, Harry tries not to rely on branch versions. Harry-core is 
completely free of Cassandra dependency (for which I had to copy some 
concurrency primitives, but they will go away if we bring it in-tree). 

You will still be able to write upgrade tests more or less how you write 
upgrade tests today: with API that is limited to cross-version jvm-dtest-api.

> (We can publish jar files at C* release time if there's a call for this, 
> doesn't really matter if they don't contain changes.). 

I would say we should publish dtest jar files when releasing, this would be 
very helpful! 




On Fri, Nov 24, 2023, at 5:23 PM, Mick Semb Wever wrote:
> 
>  
>  
>> I had conversations with several folks, and wanted to propose to move 
>> harry-core to Cassandra test tree. This will substantially 
>> simplify/streamline co-development of Cassandra and Harry. With a new 
>> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it 
>> will also be much more approachable.
> 
> 
> Yes.  Any reason to have releases for users outside of Cassandra ? (We can 
> publish jar files at C* release time if there's a call for this, doesn't 
> really matter if they don't contain changes.).  And is there any 
> cross-branch/version interaction, like there is with the jvm-dtest-api ?
> 
> 
>  

Re: [DISCUSS] Harry in-tree

[Mick Semb Wever]

> I had conversations with several folks, and wanted to propose to move
> harry-core to Cassandra test tree. This will substantially
> simplify/streamline co-development of Cassandra and Harry. With a new
> HistoryBuilder API that has helped to find and trigger [1] [2] and [3], it
> will also be much more approachable.
>


Yes.  Any reason to have releases for users outside of Cassandra ? (We can
publish jar files at C* release time if there's a call for this, doesn't
really matter if they don't contain changes.).  And is there any
cross-branch/version interaction, like there is with the jvm-dtest-api ?