Generator-jhipster: Remove Infinispan option

[jdubois]

Infinispan distributed cache support has been added by @srinivasa-vasu and I propose to remove this option:

• Last month it was only used by 0,58% of applications
• It has quite a high maintenance cost as it's complex to test
• The vast majority of users use Hazelcast, which basically does the same

@srinivasa-vasu I know you coded this while working at Red Hat, but as you left the company I'm guessing you don't have much incentive to continue supporting it?

Would anybody be interesting in maintaining Infinispan support?

[srinivasa-vasu]

Julien,

I see the usage is minimal. But If it makes sense to make it available for
our users, I don't have any issues maintaining it.

Thanks,
-Srini

On Mon, Dec 10, 2018, 12:41 Julien Dubois <[email protected] wrote:

Infinispan distributed cache support has been added by @srinivasa-vasu
https://github.com/srinivasa-vasu and I propose to remove this option:

• Last month it was only used by 0,58% of applications
• It has quite a high maintenance cost as it's complex to test
• The vast majority of users use Hazelcast, which basically does the
  same

@srinivasa-vasu https://github.com/srinivasa-vasu I know you coded this
while working at Red Hat, but as you left the company I'm guessing you
don't have much incentive to continue supporting it?

Would anybody be interesting in maintaining Infinispan support?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937, or mute the
thread
https://github.com/notifications/unsubscribe-auth/ATlnTyvGYBu7OpDHlZquCGzPYkVRbOKeks5u3gk6gaJpZM4ZKxxk
.

[jdubois]

Thanks @srinivasa-vasu : let's keep it as long as it doesn't cause any big issues, then we'll see if it's worth the effort or not.

[DanielFran]

@srinivasa-vasu if you can take a look in spring-boot 2.1 branch, the update to latest version needs some adjustments... Thanks

[srinivasa-vasu]

Hi,

I am on vacation. Will check that once I am back.

Thanks,
-Srini

On Sat, Dec 15, 2018, 18:34 Daniel Franco <[email protected] wrote:

@srinivasa-vasu https://github.com/srinivasa-vasu if you can take a
look in spring-boot 2.1 branch, the update to latest version needs some
adjustments... Thanks

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937#issuecomment-447565404,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ATlnT3d1GL_sPpxvgZimqW4H7yL6pb6eks5u5OxdgaJpZM4ZKxxk
.

[kdejaeger]

@srinivasa-vasu I check out the infinispan implementation yesterday. Can you verify if the hibernate cache L2 inits/overrides are working properly. I had the impression that the defined configuration was not taken into account by the infinispan framework and just the configs defined in the default infinispan-configs.xml were activated.

[DanielFran]

@srinivasa-vasu @kdejaeger Do not forget to look at the correct branch:

• https://github.com/jhipster/jhipster/commits/spring-boot_2.1.0
• https://github.com/jhipster/generator-jhipster/commits/spring-boot_2.1.0

[srinivasa-vasu]

@DanielFran, I tested for a sample monolith - don't see any issues. Will
test for a micro as well.

On Wed, Dec 19, 2018 at 4:47 PM Daniel Franco [e-mail]
wrote:

@srinivasa-vasu https://github.com/srinivasa-vasu @kdejaeger
https://github.com/kdejaeger Do not forget to look at the correct
branch:

• https://github.com/jhipster/jhipster/commits/spring-boot_2.1.0
  -
  https://github.com/jhipster/generator-jhipster/commits/spring-boot_2.1.0

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937#issuecomment-448559353,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ATlnT0E5u8Fh4OaBoCWcMk4XVu5zpxRYks5u6iA-gaJpZM4ZKxxk
.

[srinivasa-vasu]

@kdejaeger - I don't see any issues mate. Defined custom config is getting
reflected. Do you want to be more specific?

On Wed, Dec 19, 2018 at 2:17 PM kdejaeger [e-mail] wrote:

@srinivasa-vasu https://github.com/srinivasa-vasu I check out the
infinispan implementation yesterday. Can you verify if the hibernate cache
L2 inits/overrides are working properly. I had the impression that the
defined configuration was not taken into account by the infinispan
framework and just the configs defined in the default
infinispan-configs.xml were activated.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937#issuecomment-448515288,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ATlnT60amr24DFnF1DJeGbUCNKR3Q37hks5u6f00gaJpZM4ZKxxk
.

[kdejaeger]

My experience with this is much lower, so I trust you fully. For instance I didn't already understand why I believe 'replicated-query' is defined in cacheconfig when the infinispan xml file seemed to contain only 'query' or something. I was as if the key's to override did not match the config file. I thought it had to mach those key's, just as the onces defined in the DataType.class enum. I just had put some breakpoints in the infinispan code where it seemed as if the framework was still preferring the infinispan xml config. Based on some changes I made in the time-to-live java configuration. I'll check it out again when the spring boot 2.1.x generator is released. Thanks for looking into this.

[srinivasa-vasu]

@kdejaeger - Sure, let me know if you face any issues. To answer your
question, if you explicitly mention the infinispan xml config, then it will
override your configuration. If not, defined config will be taken into
consideration. For L2, 'entity' config is applicable for all the domain
objects. Few things are hibernate specific.

On Mon, Dec 24, 2018 at 2:23 PM kdejaeger [e-mail] wrote:

My experience with this is much lower, so I trust you fully. For instance
I didn't already understand why I believe 'replicated-query' is defined in
cacheconfig when the infinispan xml file seemed to contain only 'query' or
something. I was as if the key's to override did not match the config file.
I thought it had to mach those key's, just as the onces defined in the
DataType.class enum. I just had put some breakpoints in the infinispan code
where it seemed as if the framework was still preferring the infinispan xml
config. Based on some changes I made in the time-to-live java
configuration. I'll check it out again when the spring boot 2.1.x generator
is released. Thanks for looking into this.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937#issuecomment-449704859,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ATlnT8tUuwy8WAU5NCrk0tJE5VNVenSVks5u8JYbgaJpZM4ZKxxk
.

[DanielFran]

@srinivasa-vasu example in Travis: https://travis-ci.org/jhipster/generator-jhipster/jobs/470923637

[srinivasa-vasu]

@DanielFran - Is it intermittent? I ran the tests locally from the same
branch, don't see any problems. Do you want to restart the travis build
again? or have you tried already?

On Mon, Dec 24, 2018 at 4:14 PM Daniel Franco [e-mail]
wrote:

@srinivasa-vasu https://github.com/srinivasa-vasu example in Travis:
https://travis-ci.org/jhipster/generator-jhipster/jobs/470923637

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937#issuecomment-449717658,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ATlnT-2Z3A2Y83tB3VR-64Qm1w23T1Z-ks5u8LANgaJpZM4ZKxxk
.

[DanielFran]

@srinivasa-vasu it happens always in Travis... You can check the historic.

[srinivasa-vasu]

@DanielFran - I do see a problem with microservices app. I am getting a
different exception, but not the one in the travis build. I've pushed a
changed to the 2.1 branch. Let's see if this fixes the issue. I'll check
other cases as well.

On Mon, Dec 24, 2018 at 7:50 PM Daniel Franco [e-mail]
wrote:

@srinivasa-vasu https://github.com/srinivasa-vasu it happens always in
Travis... You can check the historic.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937#issuecomment-449739599,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ATlnT5UALWb9J85OqECTV8YvdPYtWBEsks5u8OKlgaJpZM4ZKxxk
.

[srinivasa-vasu]

@DanielFran - infinispan-spring-boot-start - 2.1.1.Final has some issues
with micrometer and JCache based cache initialisation. travis build null
pointer exception is because of this. I've submitted a pull request to
infinispan to address that particular issue. Let's not use 2.1.1, instead
will use 2.0.0.Final till the request is being acted upon.
2.0.0.Final

On Tue, Dec 25, 2018 at 11:43 PM Srinivasan [e-mail]
wrote:

@DanielFran - I do see a problem with microservices app. I am getting a
different exception, but not the one in the travis build. I've pushed a
changed to the 2.1 branch. Let's see if this fixes the issue. I'll check
other cases as well.

On Mon, Dec 24, 2018 at 7:50 PM Daniel Franco [e-mail]
wrote:

@srinivasa-vasu https://github.com/srinivasa-vasu it happens always in
Travis... You can check the historic.

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/jhipster/generator-jhipster/issues/8937#issuecomment-449739599,
or mute the thread
https://github.com/notifications/unsubscribe-auth/ATlnT5UALWb9J85OqECTV8YvdPYtWBEsks5u8OKlgaJpZM4ZKxxk
.

[kdejaeger]

Why are there no expiration times defined in the hibernate l2 caches? --> I think I get it. The lifespan is overriden globally through the yml config files. (hibernate -> infinispan) section.