jmeter-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From sebb <>
Subject Re: Java8 for 3.1 ?
Date Sun, 11 Sep 2016 12:40:23 GMT
On 10 September 2016 at 20:58, Felix Schumacher
<> wrote:
> Am 10.09.2016 um 21:53 schrieb Vladimir Sitnikov:
>> Felix>My main concern is, that we exclude all those people, that are
>> Felix>still using java 7 (for whatever reason).
>> Are there people who want to use JMeter 3.1 and they cannot use Java 8?
>> If one is stuck with Java 7, then JMeter 3.0 will work just fine.
> There are a few minor nags in 3.0 which I would really like to get solved in
> a 3.1 or 3.0.1.
> After that I am fine with updating to java 8.
>> Is there anybody who contributes to JMeter and who cannot use Java 8?
> As contribution is everything from writing patches to telling your friends
> how awesome jmeter is, I have to say: I don't know ;)
>> "we exclude..still using java 7" is not practical as there might be always
>> be someone
>> who "still uses java 7".
> Right and that is why we will update to java 8 after the (hopefully near)
> next release.
> If everyone else thinks we should update to java 8 right now, I might change
> my mind, though.

It's all about the cost-benfit analysis.

JMeter is generally used stand-alone and can be run in its own JVM or
host if necessary, so Java upgrades shouldn't involve much work.

It's still important to ensure that the Java version is widely
available (and stable) on different OSes so that upgrading is possible
for as many end-users as possible.
So Java 9 is out of scope at present.

Also there will still be costs if users have to upgrade, so there must
be sufficient end-user benefits to justify the costs.

In principle, I am not against requiring Java 8 now, except that there
are some bugs that really deserve to be fixed first.

I think moving to 8 now will take effort which is better spent on
fixing the bugs and regressions that have been reported.

> Felix
>> Vladimir

View raw message