Discussion:
Regarding latest errata
Add Reply
Patrick Harper
2018-04-15 18:58:38 UTC
Reply
Permalink
Raw Message
Hi All,

Unless I am mistaken, the errata posted on the 14th April is the first that has been applied to more than two releases, implying that 6.1-stable is still supported. Does this signify a change to the lifecycle process?

Regards,
--
Patrick Harper
***@fastmail.com
Theo de Raadt
2018-04-15 19:02:27 UTC
Reply
Permalink
Raw Message
Post by Patrick Harper
Unless I am mistaken, the errata posted on the 14th April is the first
that has been applied to more than two releases, implying that
6.1-stable is still supported. Does this signify a change to the
lifecycle process?
No it does not indicate that.

Official release date of 6.3 is April 15. Yes, the release went out
the door early, but the *official* date is April 15.

Therefore we made it for 6.1 also, since 6.1 people may still be
running on the day before the *official* release day.

We only support 2 active releases. Pulling this trick out of our hat
was extra effort, and hopefully won't be repeated again. Thanks to
robert and tb.
Solene Rapenne
2018-04-15 21:05:43 UTC
Reply
Permalink
Raw Message
Post by Theo de Raadt
Official release date of 6.3 is April 15. Yes, the release went out
the door early, but the *official* date is April 15.
The release date is wrong in index.html, following patch fix the date to
April 15 th.

Index: index.html
===================================================================
RCS file: /cvs/www/index.html,v
retrieving revision 1.724
diff -r1.724 index.html
107c107
< The current release is <a href="63.html">OpenBSD 6.3</a>, released Apr 2, 2018.
---
Post by Theo de Raadt
The current release is <a href="63.html">OpenBSD 6.3</a>, released Apr 15, 2018.
Patrick Harper
2018-04-17 15:15:13 UTC
Reply
Permalink
Raw Message
The best solution I can think of is planning, announcing and implementing oldstable EOLs in advance, but I'm not sure this would kill enough time in building patches to be worth a process change, and users would have to trade patches for contingency. Make of this whatever you will, I don't know what is more important.
--
Patrick Harper
Post by Theo de Raadt
Post by Patrick Harper
Unless I am mistaken, the errata posted on the 14th April is the first
that has been applied to more than two releases, implying that
6.1-stable is still supported. Does this signify a change to the
lifecycle process?
No it does not indicate that.
Official release date of 6.3 is April 15. Yes, the release went out
the door early, but the *official* date is April 15.
Therefore we made it for 6.1 also, since 6.1 people may still be
running on the day before the *official* release day.
We only support 2 active releases. Pulling this trick out of our hat
was extra effort, and hopefully won't be repeated again. Thanks to
robert and tb.
Theo de Raadt
2018-04-17 15:19:34 UTC
Reply
Permalink
Raw Message
Huh? We've told everyone 2 releases maintained with errata/syspatches,
6 months apart, only. Nothing changed here. We don't need to
change a single word about EOL. It is exactly the same as before.
Post by Patrick Harper
The best solution I can think of is planning, announcing and
implementing oldstable EOLs in advance, but I'm not sure this would
kill enough time in building patches to be worth a process change, and
users would have to trade patches for contingency. Make of this
whatever you will, I don't know what is more important.
--
Patrick Harper
Post by Theo de Raadt
Post by Patrick Harper
Unless I am mistaken, the errata posted on the 14th April is the first
that has been applied to more than two releases, implying that
6.1-stable is still supported. Does this signify a change to the
lifecycle process?
No it does not indicate that.
Official release date of 6.3 is April 15. Yes, the release went out
the door early, but the *official* date is April 15.
Therefore we made it for 6.1 also, since 6.1 people may still be
running on the day before the *official* release day.
We only support 2 active releases. Pulling this trick out of our hat
was extra effort, and hopefully won't be repeated again. Thanks to
robert and tb.
Patrick Harper
2018-04-17 15:54:08 UTC
Reply
Permalink
Raw Message
What changed was that there was a period after 6.3 was pushed out the door (2-15 April) in which there were effectively three active releases and the project felt obliged to support 6.1 until 6.3's projected release date. My previous post attempted to review a possible workaround, though I suspect this sort of anomaly might not be practically avoidable.

(Theo received this twice, sorry)
--
Patrick Harper
Post by Theo de Raadt
Huh? We've told everyone 2 releases maintained with errata/syspatches,
6 months apart, only. Nothing changed here. We don't need to
change a single word about EOL. It is exactly the same as before.
Post by Patrick Harper
The best solution I can think of is planning, announcing and
implementing oldstable EOLs in advance, but I'm not sure this would
kill enough time in building patches to be worth a process change, and
users would have to trade patches for contingency. Make of this
whatever you will, I don't know what is more important.
--
Patrick Harper
Post by Theo de Raadt
Post by Patrick Harper
Unless I am mistaken, the errata posted on the 14th April is the first
that has been applied to more than two releases, implying that
6.1-stable is still supported. Does this signify a change to the
lifecycle process?
No it does not indicate that.
Official release date of 6.3 is April 15. Yes, the release went out
the door early, but the *official* date is April 15.
Therefore we made it for 6.1 also, since 6.1 people may still be
running on the day before the *official* release day.
We only support 2 active releases. Pulling this trick out of our hat
was extra effort, and hopefully won't be repeated again. Thanks to
robert and tb.
Theo de Raadt
2018-04-17 15:57:20 UTC
Reply
Permalink
Raw Message
Post by Patrick Harper
What changed was that there was a period after 6.3 was pushed out the
door (2-15 April) in which there were effectively three active
releases and the project felt obliged to support 6.1 until 6.3's
projected release date. My previous post attempted to review a
possible workaround, though I suspect this sort of anomaly might not
be practically avoidable.
You are making stuff up.
Patrick Harper
2018-04-17 17:25:31 UTC
Reply
Permalink
Raw Message
Assuming that is the case, was it 6.3 or 6.1 that was not 'active' from the 2nd to the 15th? Conveniently the original 6.3 release dates are now censored on the website, but if it had been built for the projected date then it would not have needed the 14th patches.
--
Patrick Harper
Post by Theo de Raadt
Post by Patrick Harper
What changed was that there was a period after 6.3 was pushed out the
door (2-15 April) in which there were effectively three active
releases and the project felt obliged to support 6.1 until 6.3's
projected release date. My previous post attempted to review a
possible workaround, though I suspect this sort of anomaly might not
be practically avoidable.
You are making stuff up.
Theo de Raadt
2018-04-17 17:28:10 UTC
Reply
Permalink
Raw Message
What a futile and pointless discussion.
Post by Patrick Harper
Assuming that is the case, was it 6.3 or 6.1 that was not 'active'
from the 2nd to the 15th? Conveniently the original 6.3 release dates
are now censored on the website, but if it had been built for the
projected date then it would not have needed the 14th patches.
--
Patrick Harper
Post by Theo de Raadt
Post by Patrick Harper
What changed was that there was a period after 6.3 was pushed out the
door (2-15 April) in which there were effectively three active
releases and the project felt obliged to support 6.1 until 6.3's
projected release date. My previous post attempted to review a
possible workaround, though I suspect this sort of anomaly might not
be practically avoidable.
You are making stuff up.
Patrick Harper
2018-04-17 19:04:49 UTC
Reply
Permalink
Raw Message
I agree. Your initial response was all I needed, I thought I needed more because I'm an absolutist.
--
Patrick Harper
Post by Theo de Raadt
What a futile and pointless discussion.
Post by Patrick Harper
Assuming that is the case, was it 6.3 or 6.1 that was not 'active'
from the 2nd to the 15th? Conveniently the original 6.3 release dates
are now censored on the website, but if it had been built for the
projected date then it would not have needed the 14th patches.
--
Patrick Harper
Post by Theo de Raadt
Post by Patrick Harper
What changed was that there was a period after 6.3 was pushed out the
door (2-15 April) in which there were effectively three active
releases and the project felt obliged to support 6.1 until 6.3's
projected release date. My previous post attempted to review a
possible workaround, though I suspect this sort of anomaly might not
be practically avoidable.
You are making stuff up.
Loading...