Hi Daniel,

We decided in the past that we would wait at least 1 year before releasing a freezed branch. We did not exclude to release more often... This could be discussed again with pro and cons...

Jacques

Le 11/02/2021 à 10:37, Daniel Watford a écrit :
Hi Jacques,

+1 to all the points raised by Michael.

It can be a bit disheartening knowing that changes we make to trunk today
probably won't be used by the stable-branch user community for a few years.
Anything that can be done to shorten the major release cycle will be a net
positive in my view, but I appreciate that a too rapid release cycle might
result in some users being reluctant to upgrade and remain on very old
branches.

Thanks,

Dan.

On Thu, 11 Feb 2021 at 09:15, Jacques Le Roux <jacques.le.r...@les7arts.com>
wrote:

Hi,

We started a discussion in the PMC private ML and I think we should
continue here.

Notably because of the Jcenter issue, Jacopo and Michael proposed to
create a last R17 release: 17.12.06, nobody opposed.

Michael also proposed:

  1. solve the jCenter problem (should also be adressed for r17 as it will
most likely be the last release!).
  2. release 17.12.06 asap
  3. stabilize/bugfix r18.12 asap and release (March/April?)
  4. migrate trunk to jdk11
  5. create r21 branch (around June?)
  6. after September 21: migrate trunk to jdk17
  7. release r21 in the 1st half of 2022 create r22 after that

Feedback from my side:

  1. I just tested, applying PR 270+53.I got some issues[1]/ @MIchael, may
I miss something (note: I'm on Windows 7)?
  2. +1
  3. +1, the sooner the better
  4. +1
  5. +1, I think we can create the R21 just after releasing R18 where all
bug fix should be, no reasons to wait, again the sooner the better
  6. +1
  7. +1, again the sooner the better

I use a lot of "the sooner the better". What do you think about doing the
1st release of R18 2 years after its creation? in our (PMC) defense, we had
to treat a lot of vulnerabilities between. Good side of the coin, OFBiz
has never been as secured (logical you'd say ;))

[1]
The integration tests (cleanAll eclipse loadAll testIntegration) failed:
java.util.concurrent.ExecutionException: java.lang.NoClassDefFoundError:
Could not initialize class
org.apache.commons.dbcp2.managed.PoolableManagedConnectionFactory
Got these errors running OFBiz just after (clean state):

2021-02-11 10:05:23,386 |delegator-startup-1
|ExecutionPool                 |E| null
java.util.concurrent.ExecutionException: java.lang.NoClassDefFoundError:
org/apache/logging/log4j/util/ReflectionUtil

of course
2021-02-11 10:05:23,392 |delegator-startup-1
|ExecutionPool                 |E| null
java.util.concurrent.ExecutionException: java.lang.NoClassDefFoundError:
Could not initialize class
org.apache.commons.dbcp2.managed.PoolableManagedConnectionFactory

Feb 11, 2021 10:05:24 AM org.apache.catalina.core.StandardContext
filterStart
SEVERE: Exception starting filter [SameSiteFilter]
java.lang.NoClassDefFoundError: Could not initialize class
org.apache.ofbiz.webapp.control.SameSiteFilter
[...]
SEVERE: Context [/ebay] startup failed due to previous errors
Feb 11, 2021 10:05:24 AM org.apache.catalina.loader.WebappClassLoaderBase
clearReferencesThreads
WARNING: The web application [ebay] appears to have started a thread named
[Log4j2-TF-10-Scheduled-3] but has failed to stop it. This is very likely
to create a memory leak. Stack trace of thread:
   sun.misc.Unsafe.park(Native Method)
[...]
SEVERE: One or more Filters failed to start. Full details will be found in
the appropriate container log file
[...]
SEVERE: Context [/marketing] startup failed due to previous errors
[...]
SEVERE: Context [/exampleext] startup failed due to previous errors
[...]
SEVERE: Context [/rest] startup failed due to previous errors

Etc.

Jacques


Reply via email to