On 2021-09-07 11:47, Achim Gratz wrote:
did you actuallky changed anything on these releases vs. the previous
test releases?  IN principle it should have sufficed to just untest
these.

I can try that next time with dash (see below). From the setup extracts below you can see that does not change the previous two releases kept. I thought in recent discussion, stable releases had to be bumped to replace test releases, so I did that with grep, where a test of a recent release replaced an earlier test release, and readline:

        grep                    3.7-2
                version: 3.7-2
                [test]
                version: 3.6-1
                [prev]
                version: 3.0-1
                [prev]
                version: 3.0-2
                [test]
                version: 3.7-1

        libreadline-devel       8.1-2
        libreadline7            8.1-2
        readline-debuginfo      8.1-2

                version: 8.1-2
                [prev]
                version: 7.0.1-2
                [prev]
                version: 7.0.3-3
                [test]
                version: 8.1-1

but gzip had a new recent release, mostly gnulib, autotools, NetBSD, IBM Z390, and checked clean, so I released that as stable:

                version: 1.11-1
                [prev]
                version: 1.7-2
                [prev]
                version: 1.8-1
                [test]
                version: 1.10-1

and a new dash was released, but as that's so critical to installs I'm keeping that release also in test for another couple of weeks:

                version: 0.5.9.1-1
                [test]
                version: 0.5.11.4-1
                [prev]
                version: 0.5.8-2
                [prev]
                version: 0.5.8-3
                [test]
                version: 0.5.11.5-1

and I use that as my production /bin/sh, which ensures it gets worked as hard as bash by X launches, hourly and daily cron jobs, and scheduled tasks.

--
Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada

This email may be disturbing to some readers as it contains
too much technical detail. Reader discretion is advised.
[Data in binary units and prefixes, physical quantities in SI.]

Reply via email to