Topic: SMF 2.0.13 Released

[d3vcho();]

There's one last bug we must fix before releasing 2.0.14.

No further updates

[Ben_S]

It is disappointing as I’ve previously only really posted in the private boards but the length of time to release such an important update to enable running HTTPS without a whole bunch of errors really does make it hard to justify continuing running SMF.

Hopefully it arrives before October when Chrome will declare all HTTP sites as insecure, login form or not.

[Shambles]

But this topic is about the release of 2.0.13

[Arantor]

But this topic is about the release of 2.0.13

And the time until 2.0.14...

[MobileCS]

I think the lack of communication is what's frustrating everyone. Nothing is being offering voluntarily. The only time we hear anything is when someone specifically asks about it.

A "Hey, we are still working on it, hang tight - we're real close now" from someone working on the project would go a long way.

[d3vcho();]

Thank you for your thoughts. In the end we are a team, and everything we are going to do is previously discussed with everyone. We have some informative topics already done, but we have decided to wait for the best moment to publish them.

We beg you for patience. Please remember we're all volunteers. Probably 2.0.14 would have been published a long ago with all developers working altogether, but unfortunately is not like that because of real life matters.

And please stop asking about 2.0.14. This topic is about 2.0.13 and its release.

[Arantor]

And please stop asking about 2.0.14. This topic is about 2.0.13 and its release.

Yes, yes it is, and people find the release lacking because the things that external forces deem necessary are finally pushing SMF to fix things.

If it helps rationalise it, every person who is pushing for 2.0.14 is expressing the fact that 2.0.13 has some major gaps in it.

The fact that people with the skills and time have indicated a willingness to help fix the problem only to be given a highly cryptic and unhelpful response is part of the problem.

And unfortunately the 'we're all volunteers' card doesn't carry enough weight when peoples' sites are losing members because of continuously deferring a problem until it's someone else's problem - this situation with 2.0.14 was a known issue in 2014 (hah). Every release from 2.0.8 onwards was an opportunity to lay the groundwork for dealing with this issue. Including 2.0.13.

You want to talk about 2.0.13 rather than 2.0.14? Why didn't SMF tackle this problem in 2014 or even throughout the bulk of 2015 before it would have been a problem? Why is it now, 18 months after 7.0 released, that it's finally being dealt with? Because no-one in that time could possibly have said 'maybe we should fix this'?

[Kindred]

Arantor... except the problem - in this case - is not something that needs to be addressed by those people...

The biggest delay in getting this out is the coordination necessary for 2.0.x releases at this time, given the systems involved - as has been evident with the problems (not with code, but with the release process itself) for the past 4 releases.

[Arantor]

Funny, the last update in this thread is that there was still an issue outstanding to be looked at, except when I asked what it was, I got only a cryptic link that didn't actually seem to be a relevant issue.

It's also interesting to note that as far as I know the release process hasn't changed in forever... why is that an issue now?

[Kindred]

It's an issue because the release process HAS changed... as evidenced by the issues with the release process in the last 4 releases.

To be clear, the remaining issue (solved) was regarding a specific situation with proxy avatars which we didn't note initially but encountered, identified and fixed after some additional testing.  However, identifying and fixing that issue meant that we had to rebuild the release, retest and confirm that the release process would not "suck in" the previous candidate that had been replaced (which was what happened with more than one of the previous 4 releases)

[kat★]

Oh dear sounds like a LOT of extra work... Hope things got solved and won't cause more issues in the future! Am becoming really anxious of when new releases come out! We're counting on you, devs!!

[GravuTrad]

Why 2.0.14 here? Thanks.

[Linkjay]

Why 2.0.14 here? Thanks.

Testing.

[MobileCS]

It's been 5 days now -  it surely can't take that long to test to make sure proxy.php is returning 404 correctly.

I'm assuming you've run in to more issues with 2.0.14?

[Kindred]

Yes

[Arantor]

So engage the community to help. There are people who can help. Who are offering to help with this.

[MobileCS]

I'm just frustrated that it's costing me an extra $160 a month (for an additional server) so I can run PHP 7.

[Arantor]

I'm just frustrated that this project apparently has the same issues it had three years ago when I found it impossible to get anything done.

Though, $160 a month for a test server is rather extortionate. I pay $50/month for a VPS with 4GB RAM and a lot of hard disk and that includes regular backups - with Linode.

[SleePy]

The proxy issue was fixed and we even silently updated the test patch we have here to keep testing.  However a bug was discovered regarding search results and the proxy for https.  Devs are busy with real life and haven't had a chance to dig into what caused that issue (and neither have I).

Again, this update changes a lot of stuff and we are wanting to carefully test it to not find any bugs, like said bug.  The bug from my quick testing is only showing in 2.0, and hasn't affected 2.1.

[MobileCS]

Though, $160 a month for a test server is rather extortionate. I pay $50/month for a VPS with 4GB RAM and a lot of hard disk and that includes regular backups - with Linode.

Oh, it's not a test server - it's in full use. I wouldn't spend that kind of money just for testing. As soon as 2.0.14 is released, I can move everything back to my main server where it belongs.