[DISCUSS] Revisiting Java 11's experimental status
Support for Java 11 was added a long time ago, and it's been about 2 years since it was released (Sept 2018). Had we released Cassandra 4 close to that date, I'd be fine with keeping the status as experimental, but at this point I'm wondering if releasing a new major version of C* that's primarily targeting Java 8 as the only "official" supported version is a good idea. To those of you that are planning on rolling out C* 4.0, are you planning on using Java 8 still, or moving to 11? Speaking for myself, I can say I don't think I'd want to use 8 anymore. If most folks are testing with 11 at this point, I think we should consider making 11 the recommended version and really only encouraging Java 8 for legacy purposes - teams who have a restriction that prevents them from upgrading. To those of you planning on moving to 4.0 soon after it's release, are you planning on deploying to JDK 11 or 8? [1] https://www.oracle.com/java/technologies/java-se-support-roadmap.html
Re: [DISCUSS] Revisiting Java 11's experimental status
> To those of you planning on moving to 4.0 soon after it's release, are you > planning on deploying to JDK 11 or 8? Curious, what's the harm in leaving it in experimental until 4.1/4.0.x ? I thought we are emphasising putting in place a new precedence of Quality First. The age, or stability, of JDK11 itself isn't the all here :shrug: I suspect most will upgrade to 4.0 on java 8 …wait a bit… and then upgrade to java 11. (And again wait a bit before trying any of the new collectors.) I would certainly recommend that approach. And won't driving momentum onto JDK 11 be just as easily/much done with blog posts, benchmarks and articles?
Re: [DISCUSS] Revisiting Java 11's experimental status
Thanks for bringing this up Jon! My current thinking is we should officially support both 8 and 11. That increases the surface area we need to test but I think its hard to predict what different users will run given the current transition in the Java landscape. Jordan On Mon, Jul 13, 2020 at 11:42 AM Jon Haddad wrote: > Support for Java 11 was added a long time ago, and it's been about 2 years > since it was released (Sept 2018). Had we released Cassandra 4 close to > that date, I'd be fine with keeping the status as experimental, but at this > point I'm wondering if releasing a new major version of C* that's primarily > targeting Java 8 as the only "official" supported version is a good idea. > > To those of you that are planning on rolling out C* 4.0, are you planning > on using Java 8 still, or moving to 11? Speaking for myself, I can say I > don't think I'd want to use 8 anymore. If most folks are testing with 11 > at this point, I think we should consider making 11 the recommended version > and really only encouraging Java 8 for legacy purposes - teams who have a > restriction that prevents them from upgrading. > > To those of you planning on moving to 4.0 soon after it's release, are you > planning on deploying to JDK 11 or 8? > > [1] https://www.oracle.com/java/technologies/java-se-support-roadmap.html >
Re: [DISCUSS] Revisiting Java 11's experimental status
Jon, But Java 11 hasn't been tested in production. I would need to submit a patch for documentation if Java 11 is made recommended version. Based on a recent survey the majority are still using Java 8, probably because it involves code review and update to migrate to a latter version. "At 58%, the majority of respondents reported using Java 8 as the programming language of choice in their main application. Java 11 was the next highest at 23% of respondents. " https://www.jrebel.com/blog/2020-java-technology-report#:~:text=At%2058%25%2C%20the%20majority%20of,using%20Java%2012%20or%20newer. Deepak On Monday, July 13, 2020, 11:42:18 a.m. PDT, Jon Haddad wrote: Support for Java 11 was added a long time ago, and it's been about 2 years since it was released (Sept 2018). Had we released Cassandra 4 close to that date, I'd be fine with keeping the status as experimental, but at this point I'm wondering if releasing a new major version of C* that's primarily targeting Java 8 as the only "official" supported version is a good idea. To those of you that are planning on rolling out C* 4.0, are you planning on using Java 8 still, or moving to 11? Speaking for myself, I can say I don't think I'd want to use 8 anymore. If most folks are testing with 11 at this point, I think we should consider making 11 the recommended version and really only encouraging Java 8 for legacy purposes - teams who have a restriction that prevents them from upgrading. To those of you planning on moving to 4.0 soon after it's release, are you planning on deploying to JDK 11 or 8? [1] https://www.oracle.com/java/technologies/java-se-support-roadmap.html
Re: [DISCUSS] Revisiting Java 11's experimental status
Personally, I'd planned to upgrade to 4.0 on JDK8 but only wait a few weeks before starting to update to JDK11 afterwards. Everything else we run's been updated to JDK11, so the Cassandra clusters are the odd one out at this point. On Mon, Jul 13, 2020 at 12:19 PM Jordan West wrote: > Thanks for bringing this up Jon! My current thinking is we should > officially support both 8 and 11. That increases the surface area we need > to test but I think its hard to predict what different users will run given > the current transition in the Java landscape. > > Jordan > > On Mon, Jul 13, 2020 at 11:42 AM Jon Haddad wrote: > > > Support for Java 11 was added a long time ago, and it's been about 2 > years > > since it was released (Sept 2018). Had we released Cassandra 4 close to > > that date, I'd be fine with keeping the status as experimental, but at > this > > point I'm wondering if releasing a new major version of C* that's > primarily > > targeting Java 8 as the only "official" supported version is a good idea. > > > > To those of you that are planning on rolling out C* 4.0, are you planning > > on using Java 8 still, or moving to 11? Speaking for myself, I can say I > > don't think I'd want to use 8 anymore. If most folks are testing with 11 > > at this point, I think we should consider making 11 the recommended > version > > and really only encouraging Java 8 for legacy purposes - teams who have a > > restriction that prevents them from upgrading. > > > > To those of you planning on moving to 4.0 soon after it's release, are > you > > planning on deploying to JDK 11 or 8? > > > > [1] > https://www.oracle.com/java/technologies/java-se-support-roadmap.html > > >
Re: [DISCUSS] Revisiting Java 11's experimental status
All the major linux vendors are supporting JDK8 LTS (like 2024+ or something) if I'm not mistaken, so I don't think there's a burning *need* to push for JDK11 support specifically in 4.0. To Mick's point, no reason why project folks that want JDK11 officially supported can't get started working on validating it now and we give it our blessing in whatever C* patch rel we've validated it on right? If most folks are testing with 11 at this point, How do we get this data? I think most of the DS folks are testing on JDK8 since that's the one we as a project have been signalling as officially supported for 4.0. Anyone else have another PoV and been doing testing on JDK11 instead? Are there features JDK11 brings to the table (improvements in GC, etc) that are compelling that increase the urgency of pushing for this? Isn't our official CI all JDK8? https://builds.apache.org/view/A-D/view/Cassandra/ On Mon, Jul 13, 2020 at 3:26 PM Elliott Sims wrote: > Personally, I'd planned to upgrade to 4.0 on JDK8 but only wait a few weeks > before starting to update to JDK11 afterwards. Everything else we run's > been updated to JDK11, so the Cassandra clusters are the odd one out at > this point. > > On Mon, Jul 13, 2020 at 12:19 PM Jordan West wrote: > > > Thanks for bringing this up Jon! My current thinking is we should > > officially support both 8 and 11. That increases the surface area we need > > to test but I think its hard to predict what different users will run > given > > the current transition in the Java landscape. > > > > Jordan > > > > On Mon, Jul 13, 2020 at 11:42 AM Jon Haddad wrote: > > > > > Support for Java 11 was added a long time ago, and it's been about 2 > > years > > > since it was released (Sept 2018). Had we released Cassandra 4 close > to > > > that date, I'd be fine with keeping the status as experimental, but at > > this > > > point I'm wondering if releasing a new major version of C* that's > > primarily > > > targeting Java 8 as the only "official" supported version is a good > idea. > > > > > > To those of you that are planning on rolling out C* 4.0, are you > planning > > > on using Java 8 still, or moving to 11? Speaking for myself, I can > say I > > > don't think I'd want to use 8 anymore. If most folks are testing with > 11 > > > at this point, I think we should consider making 11 the recommended > > version > > > and really only encouraging Java 8 for legacy purposes - teams who > have a > > > restriction that prevents them from upgrading. > > > > > > To those of you planning on moving to 4.0 soon after it's release, are > > you > > > planning on deploying to JDK 11 or 8? > > > > > > [1] > > https://www.oracle.com/java/technologies/java-se-support-roadmap.html > > > > > >
[DISCUSS] A point of view on Testing Cassandra
Link: https://docs.google.com/document/d/1ktuBWpD2NLurB9PUvmbwGgrXsgnyU58koOseZAfaFBQ/edit# Myself and a few other contributors are working with this point of view as our frame of where we're going to work on improving testing on the project. I figured it might be useful to foster collaboration more broadly in the community as well as provide people with the opportunity to discuss work they're doing they may not yet have had a chance to bring up or open source. While fallout is already open-sourced, expect the schema anonymizer and some of the cassandra-diff + nosqlbench framework effort to be open-sourced / openly worked on soon. Anyone that's interested in collaborating, that would be highly welcome. Doc is view only; figured we could keep this to the ML. Thanks. ~Josh
Re: [DISCUSS] Revisiting Java 11's experimental status
JDK8 seems like the safe devil we know, but in the interest of trying to gather a bit of data, I just posted a twitter poll. https://twitter.com/patrickmcfadin/status/1282791302065557504?s=21 On Mon, Jul 13, 2020 at 12:26 PM Elliott Sims wrote: > Personally, I'd planned to upgrade to 4.0 on JDK8 but only wait a few weeks > before starting to update to JDK11 afterwards. Everything else we run's > been updated to JDK11, so the Cassandra clusters are the odd one out at > this point. > > On Mon, Jul 13, 2020 at 12:19 PM Jordan West wrote: > > > Thanks for bringing this up Jon! My current thinking is we should > > officially support both 8 and 11. That increases the surface area we need > > to test but I think its hard to predict what different users will run > given > > the current transition in the Java landscape. > > > > Jordan > > > > On Mon, Jul 13, 2020 at 11:42 AM Jon Haddad wrote: > > > > > Support for Java 11 was added a long time ago, and it's been about 2 > > years > > > since it was released (Sept 2018). Had we released Cassandra 4 close > to > > > that date, I'd be fine with keeping the status as experimental, but at > > this > > > point I'm wondering if releasing a new major version of C* that's > > primarily > > > targeting Java 8 as the only "official" supported version is a good > idea. > > > > > > To those of you that are planning on rolling out C* 4.0, are you > planning > > > on using Java 8 still, or moving to 11? Speaking for myself, I can > say I > > > don't think I'd want to use 8 anymore. If most folks are testing with > 11 > > > at this point, I think we should consider making 11 the recommended > > version > > > and really only encouraging Java 8 for legacy purposes - teams who > have a > > > restriction that prevents them from upgrading. > > > > > > To those of you planning on moving to 4.0 soon after it's release, are > > you > > > planning on deploying to JDK 11 or 8? > > > > > > [1] > > https://www.oracle.com/java/technologies/java-se-support-roadmap.html > > > > > >
Re: [DISCUSS] A point of view on Testing Cassandra
Can you please allow comments on the doc so we can leave feedback. On Mon, Jul 13, 2020 at 2:16 PM Joshua McKenzie wrote: > Link: > > https://docs.google.com/document/d/1ktuBWpD2NLurB9PUvmbwGgrXsgnyU58koOseZAfaFBQ/edit# > > > Myself and a few other contributors are working with this point of view as > our frame of where we're going to work on improving testing on the project. > I figured it might be useful to foster collaboration more broadly in the > community as well as provide people with the opportunity to discuss work > they're doing they may not yet have had a chance to bring up or open > source. While fallout is already open-sourced, expect the schema anonymizer > and some of the cassandra-diff + nosqlbench framework effort to be > open-sourced / openly worked on soon. Anyone that's interested in > collaborating, that would be highly welcome. > > Doc is view only; figured we could keep this to the ML. > > Thanks. > > ~Josh >
Re: [DISCUSS] A point of view on Testing Cassandra
> > Can you please allow comments on the doc so we can leave feedback. > > Doc is view only; figured we could keep this to the ML. > That's a feature, not a bug. Happy to chat here or on slack w/anyone. This is a complex topic so long-form or high bandwidth communication is a better fit than gdoc comments. They rapidly become unwieldy. On Mon, Jul 13, 2020 at 6:17 PM sankalp kohli wrote: > Can you please allow comments on the doc so we can leave feedback. > > On Mon, Jul 13, 2020 at 2:16 PM Joshua McKenzie > wrote: > > > Link: > > > > > https://docs.google.com/document/d/1ktuBWpD2NLurB9PUvmbwGgrXsgnyU58koOseZAfaFBQ/edit# > > > > > > Myself and a few other contributors are working with this point of view > as > > our frame of where we're going to work on improving testing on the > project. > > I figured it might be useful to foster collaboration more broadly in the > > community as well as provide people with the opportunity to discuss work > > they're doing they may not yet have had a chance to bring up or open > > source. While fallout is already open-sourced, expect the schema > anonymizer > > and some of the cassandra-diff + nosqlbench framework effort to be > > open-sourced / openly worked on soon. Anyone that's interested in > > collaborating, that would be highly welcome. > > > > Doc is view only; figured we could keep this to the ML. > > > > Thanks. > > > > ~Josh > > >
Re: [DISCUSS] Revisiting Java 11's experimental status
We at Netflix have been testing 4.0 on Java 8, and we do not plan to use Java 11 yet for C*, since we are, and for the considerable future will be running Java 8 only in production. On Mon, Jul 13, 2020 at 2:41 PM Patrick McFadin wrote: > JDK8 seems like the safe devil we know, but in the interest of trying to > gather a bit of data, I just posted a twitter poll. > > https://twitter.com/patrickmcfadin/status/1282791302065557504?s=21 > > > On Mon, Jul 13, 2020 at 12:26 PM Elliott Sims > wrote: > > > Personally, I'd planned to upgrade to 4.0 on JDK8 but only wait a few > weeks > > before starting to update to JDK11 afterwards. Everything else we run's > > been updated to JDK11, so the Cassandra clusters are the odd one out at > > this point. > > > > On Mon, Jul 13, 2020 at 12:19 PM Jordan West wrote: > > > > > Thanks for bringing this up Jon! My current thinking is we should > > > officially support both 8 and 11. That increases the surface area we > need > > > to test but I think its hard to predict what different users will run > > given > > > the current transition in the Java landscape. > > > > > > Jordan > > > > > > On Mon, Jul 13, 2020 at 11:42 AM Jon Haddad wrote: > > > > > > > Support for Java 11 was added a long time ago, and it's been about 2 > > > years > > > > since it was released (Sept 2018). Had we released Cassandra 4 close > > to > > > > that date, I'd be fine with keeping the status as experimental, but > at > > > this > > > > point I'm wondering if releasing a new major version of C* that's > > > primarily > > > > targeting Java 8 as the only "official" supported version is a good > > idea. > > > > > > > > To those of you that are planning on rolling out C* 4.0, are you > > planning > > > > on using Java 8 still, or moving to 11? Speaking for myself, I can > > say I > > > > don't think I'd want to use 8 anymore. If most folks are testing > with > > 11 > > > > at this point, I think we should consider making 11 the recommended > > > version > > > > and really only encouraging Java 8 for legacy purposes - teams who > > have a > > > > restriction that prevents them from upgrading. > > > > > > > > To those of you planning on moving to 4.0 soon after it's release, > are > > > you > > > > planning on deploying to JDK 11 or 8? > > > > > > > > [1] > > > https://www.oracle.com/java/technologies/java-se-support-roadmap.html > > > > > > > > > >
Re: [DISCUSS] Revisiting Java 11's experimental status
The majority of the testing that I'm doing and likely the people I work with day to day tests 4.0 on Java 11, so we should gain some experience over the next few months. On Mon, Jul 13, 2020 at 5:27 PM Sumanth Pasupuleti wrote: > > We at Netflix have been testing 4.0 on Java 8, and we do not plan to > use Java 11 yet for C*, since we are, and for the considerable future will > be running Java 8 only in production. > > On Mon, Jul 13, 2020 at 2:41 PM Patrick McFadin wrote: > > > JDK8 seems like the safe devil we know, but in the interest of trying to > > gather a bit of data, I just posted a twitter poll. > > > > https://twitter.com/patrickmcfadin/status/1282791302065557504?s=21 > > > > > > On Mon, Jul 13, 2020 at 12:26 PM Elliott Sims > > wrote: > > > > > Personally, I'd planned to upgrade to 4.0 on JDK8 but only wait a few > > weeks > > > before starting to update to JDK11 afterwards. Everything else we run's > > > been updated to JDK11, so the Cassandra clusters are the odd one out at > > > this point. > > > > > > On Mon, Jul 13, 2020 at 12:19 PM Jordan West wrote: > > > > > > > Thanks for bringing this up Jon! My current thinking is we should > > > > officially support both 8 and 11. That increases the surface area we > > need > > > > to test but I think its hard to predict what different users will run > > > given > > > > the current transition in the Java landscape. > > > > > > > > Jordan > > > > > > > > On Mon, Jul 13, 2020 at 11:42 AM Jon Haddad wrote: > > > > > > > > > Support for Java 11 was added a long time ago, and it's been about 2 > > > > years > > > > > since it was released (Sept 2018). Had we released Cassandra 4 close > > > to > > > > > that date, I'd be fine with keeping the status as experimental, but > > at > > > > this > > > > > point I'm wondering if releasing a new major version of C* that's > > > > primarily > > > > > targeting Java 8 as the only "official" supported version is a good > > > idea. > > > > > > > > > > To those of you that are planning on rolling out C* 4.0, are you > > > planning > > > > > on using Java 8 still, or moving to 11? Speaking for myself, I can > > > say I > > > > > don't think I'd want to use 8 anymore. If most folks are testing > > with > > > 11 > > > > > at this point, I think we should consider making 11 the recommended > > > > version > > > > > and really only encouraging Java 8 for legacy purposes - teams who > > > have a > > > > > restriction that prevents them from upgrading. > > > > > > > > > > To those of you planning on moving to 4.0 soon after it's release, > > are > > > > you > > > > > planning on deploying to JDK 11 or 8? > > > > > > > > > > [1] > > > > https://www.oracle.com/java/technologies/java-se-support-roadmap.html > > > > > > > > > > > > > > - To unsubscribe, e-mail: dev-unsubscr...@cassandra.apache.org For additional commands, e-mail: dev-h...@cassandra.apache.org
Re: [DISCUSS] Revisiting Java 11's experimental status
> On Jul 13, 2020, at 11:42 AM, Jon Haddad wrote: > > Support for Java 11 was added a long time ago, and it's been about 2 years > since it was released (Sept 2018). Had we released Cassandra 4 close to > that date, I'd be fine with keeping the status as experimental, but at this > point I'm wondering if releasing a new major version of C* that's primarily > targeting Java 8 as the only "official" supported version is a good idea. LTS is a thing > > To those of you that are planning on rolling out C* 4.0, are you planning > on using Java 8 still, or moving to 11? Speaking for myself, I can say I > don't think I'd want to use 8 anymore. If most folks are testing with 11 > at this point, I think we should consider making 11 the recommended version > and really only encouraging Java 8 for legacy purposes - teams who have a > restriction that prevents them from upgrading. > > To those of you planning on moving to 4.0 soon after it's release, are you > planning on deploying to JDK 11 or 8? Perhaps the most notable parts of jdk11 (for cassandra) aren’t even prod ready in jdk11 , so what’s the motivation and what does the project gain from revisiting the experimental designation on jdk11? - To unsubscribe, e-mail: dev-unsubscr...@cassandra.apache.org For additional commands, e-mail: dev-h...@cassandra.apache.org
