[arch-releng] November release
Hi all, I have reviewed and tested the latest patches and all looks fine so far. There are no big changes which I don't dislike. :-) As discussed before, I would like to have a few days between tagging and the actual iso release. So how about we tag the current state today, test and package it. I was thinking about building the new iso on Thursday and announce it on Friday (if we still want to have announcements). I guess it then would be best to not commit new changes or put them into another branch; we have an unused testing branch atm. Or we introduce a new maint branch for releases. Greetings, Pierre -- Pierre Schmitz, https://pierre-schmitz.com
On 10/28/2012 05:22 AM, Pierre Schmitz wrote:
Hi all,
I have reviewed and tested the latest patches and all looks fine so far. There are no big changes which I don't dislike. :-) As discussed before, I would like to have a few days between tagging and the actual iso release. So how about we tag the current state today, test and package it. I was thinking about building the new iso on Thursday and announce it on Friday (if we still want to have announcements).
I guess it then would be best to not commit new changes or put them into another branch; we have an unused testing branch atm. Or we introduce a new maint branch for releases.
Greetings,
Pierre
Hello Please wait for syslinux 4.06 to hit testing! PS: anyway pending patches are documentation only [#1] Thanks. [#1] https://github.com/djgera/archiso/compare/master...djgera -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1
Am 28.10.2012 17:18, schrieb Gerardo Exequiel Pozzi:
On 10/28/2012 05:22 AM, Pierre Schmitz wrote:
Hi all,
I have reviewed and tested the latest patches and all looks fine so far. There are no big changes which I don't dislike. :-) As discussed before, I would like to have a few days between tagging and the actual iso release. So how about we tag the current state today, test and package it. I was thinking about building the new iso on Thursday and announce it on Friday (if we still want to have announcements).
I guess it then would be best to not commit new changes or put them into another branch; we have an unused testing branch atm. Or we introduce a new maint branch for releases.
Greetings,
Pierre
Hello
Please wait for syslinux 4.06 to hit testing!
It is now moved to core.
PS: anyway pending patches are documentation only [#1]
Thanks.
What about enabling serial-getty@ttyS0.service by default? Currently, it is only enabled when you pass console=ttyS0 via command line.
On 10/28/2012 03:27 PM, Thomas Bächler wrote:
Am 28.10.2012 17:18, schrieb Gerardo Exequiel Pozzi:
On 10/28/2012 05:22 AM, Pierre Schmitz wrote:
Hi all,
I have reviewed and tested the latest patches and all looks fine so far. There are no big changes which I don't dislike. :-) As discussed before, I would like to have a few days between tagging and the actual iso release. So how about we tag the current state today, test and package it. I was thinking about building the new iso on Thursday and announce it on Friday (if we still want to have announcements).
I guess it then would be best to not commit new changes or put them into another branch; we have an unused testing branch atm. Or we introduce a new maint branch for releases.
Greetings,
Pierre
Hello
Please wait for syslinux 4.06 to hit testing! It is now moved to core.
Thanks you.
PS: anyway pending patches are documentation only [#1]
Thanks. What about enabling serial-getty@ttyS0.service by default? Currently, it is only enabled when you pass console=ttyS0 via command line.
Good question... But I think this can be problematic on systems where you have connected some serial device. The same can be applied to ttyUSB0. -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1
On 10/28/2012 06:07 PM, Gerardo Exequiel Pozzi wrote:
On 10/28/2012 03:27 PM, Thomas Bächler wrote:
Am 28.10.2012 17:18, schrieb Gerardo Exequiel Pozzi:
On 10/28/2012 05:22 AM, Pierre Schmitz wrote:
Hi all,
I have reviewed and tested the latest patches and all looks fine so far. There are no big changes which I don't dislike. :-) As discussed before, I would like to have a few days between tagging and the actual iso release. So how about we tag the current state today, test and package it. I was thinking about building the new iso on Thursday and announce it on Friday (if we still want to have announcements).
I guess it then would be best to not commit new changes or put them into another branch; we have an unused testing branch atm. Or we introduce a new maint branch for releases.
Greetings,
Pierre
Hello
Please wait for syslinux 4.06 to hit testing! It is now moved to core.
Thanks you.
PS: anyway pending patches are documentation only [#1]
Thanks. What about enabling serial-getty@ttyS0.service by default? Currently, it is only enabled when you pass console=ttyS0 via command line.
Good question... But I think this can be problematic on systems where you have connected some serial device. The same can be applied to ttyUSB0.
Ignore me, nobody complains about problems with syslinux that enabled seria 0l by default. In that case, I have no problems in enabling by default. -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1
On 10/28/2012 06:09 PM, Gerardo Exequiel Pozzi wrote:
On 10/28/2012 06:07 PM, Gerardo Exequiel Pozzi wrote:
On 10/28/2012 03:27 PM, Thomas Bächler wrote:
Am 28.10.2012 17:18, schrieb Gerardo Exequiel Pozzi:
On 10/28/2012 05:22 AM, Pierre Schmitz wrote:
Hi all,
I have reviewed and tested the latest patches and all looks fine so far. There are no big changes which I don't dislike. :-) As discussed before, I would like to have a few days between tagging and the actual iso release. So how about we tag the current state today, test and package it. I was thinking about building the new iso on Thursday and announce it on Friday (if we still want to have announcements).
I guess it then would be best to not commit new changes or put them into another branch; we have an unused testing branch atm. Or we introduce a new maint branch for releases.
Greetings,
Pierre
Hello
Please wait for syslinux 4.06 to hit testing! It is now moved to core.
Thanks you.
PS: anyway pending patches are documentation only [#1]
Thanks. What about enabling serial-getty@ttyS0.service by default? Currently, it is only enabled when you pass console=ttyS0 via command line.
Good question... But I think this can be problematic on systems where you have connected some serial device. The same can be applied to ttyUSB0.
Ignore me, nobody complains about problems with syslinux that enabled seria 0l by default.
In that case, I have no problems in enabling by default.
mmm no!, on systems without serial port, serial-getty will block startup until some timeout is reached. Try this starting it without serial ports, (ie on kvm -serial no) -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1
Am 28.10.2012 22:29, schrieb Gerardo Exequiel Pozzi:
Good question... But I think this can be problematic on systems where you have connected some serial device. The same can be applied to ttyUSB0.
Ignore me, nobody complains about problems with syslinux that enabled seria 0l by default.
In that case, I have no problems in enabling by default.
mmm no!, on systems without serial port, serial-getty will block startup until some timeout is reached.
Try this starting it without serial ports, (ie on kvm -serial no)
Awww :( There are practically no systems without a serial port, but this is a showstopper. This will still work with console=ttyS0 (or console=hvc0).
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 10/28/2012 06:39 PM, Thomas Bächler wrote:
Am 28.10.2012 22:29, schrieb Gerardo Exequiel Pozzi:
Good question... But I think this can be problematic on systems where you have connected some serial device. The same can be applied to ttyUSB0.
Ignore me, nobody complains about problems with syslinux that enabled seria 0l by default.
In that case, I have no problems in enabling by default.
mmm no!, on systems without serial port, serial-getty will block startup until some timeout is reached.
Try this starting it without serial ports, (ie on kvm -serial no)
Awww :(
There are practically no systems without a serial port, but this is a showstopper. This will still work with console=ttyS0 (or console=hvc0).
But can be disabled by default (or ondemand) on firmware configuration if motherboard has a pin-socket for it (is rare to find today a motherboard with the "fichita" DB-9 on the panel) Notebook, netbooks...no serial. - -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1 -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.19 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://www.enigmail.net/ iQEcBAEBAgAGBQJQjaipAAoJEF7VFKRb1ck4B4YH/2LdX3wLxSfJVHkQgtVp2FX6 Dlde//8NC9FEpKrH/KY6bAy6Dw2K0SJF5FDu4mHhdAR1GxE3cejQ61piH4zUSQQw fy5v6UvzT071dR+IrghPtD6x24Gtrudva4MEwjR4LZ56I/T6opX3rWKdZngDD/Ql R3a+ui0jti5zCzUQtCMcoMQvST0mT1eG5Hh8aldgu4mPlyR4Bobp59X0bBnmsI7k 0267JYxuZf2dl+YCbdpFRdRh/AE6aR/AXhK98tzCAz/Yf0PWeaQzBZyWvCpbbxDc WDC+m42VAavkEVl3Hg+eTeP/hnsH7aX8mcTmxqehZbDFZxQafmfdEhr06mMH1RY= =Wzd0 -----END PGP SIGNATURE-----
Am 28.10.2012 22:50, schrieb Gerardo Exequiel Pozzi:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 10/28/2012 06:39 PM, Thomas Bächler wrote:
Am 28.10.2012 22:29, schrieb Gerardo Exequiel Pozzi:
Good question... But I think this can be problematic on systems where you have connected some serial device. The same can be applied to ttyUSB0.
Ignore me, nobody complains about problems with syslinux that enabled seria 0l by default.
In that case, I have no problems in enabling by default.
mmm no!, on systems without serial port, serial-getty will block startup until some timeout is reached.
Try this starting it without serial ports, (ie on kvm -serial no)
Awww :(
There are practically no systems without a serial port, but this is a showstopper. This will still work with console=ttyS0 (or console=hvc0).
But can be disabled by default (or ondemand) on firmware configuration if motherboard has a pin-socket for it (is rare to find today a motherboard with the "fichita" DB-9 on the panel)
Notebook, netbooks...no serial.
I suggest to postpone any change to this to next month. f there are no objections I'll tag a new archiso release tonight. Greetings, Pierre -- Pierre Schmitz, https://pierre-schmitz.com
Am 30.10.2012 05:36, schrieb Pierre Schmitz:
But can be disabled by default (or ondemand) on firmware configuration if motherboard has a pin-socket for it (is rare to find today a motherboard with the "fichita" DB-9 on the panel)
Notebook, netbooks...no serial.
I suggest to postpone any change to this to next month. f there are no objections I'll tag a new archiso release tonight.
Great. We should move the tracker to alberich before then. Ionut, can you change the 'tracker.archlinux.org' CNAME to point to alberich.archlinux.org? The new tracker is set up there. See http://alberich.archlinux.org:6969/stat
On 10/30/2012 10:01 AM, Thomas Bächler wrote:
Am 30.10.2012 05:36, schrieb Pierre Schmitz:
But can be disabled by default (or ondemand) on firmware configuration if motherboard has a pin-socket for it (is rare to find today a motherboard with the "fichita" DB-9 on the panel)
Notebook, netbooks...no serial.
I suggest to postpone any change to this to next month. f there are no objections I'll tag a new archiso release tonight.
Great.
We should move the tracker to alberich before then. Ionut, can you change the 'tracker.archlinux.org' CNAME to point to alberich.archlinux.org? The new tracker is set up there.
done. -- Ionuț
On 10/30/2012 01:36 AM, Pierre Schmitz wrote:
Am 28.10.2012 22:50, schrieb Gerardo Exequiel Pozzi:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 10/28/2012 06:39 PM, Thomas Bächler wrote:
Am 28.10.2012 22:29, schrieb Gerardo Exequiel Pozzi:
Good question... But I think this can be problematic on systems where you have connected some serial device. The same can be applied to ttyUSB0.
Ignore me, nobody complains about problems with syslinux that enabled seria 0l by default.
In that case, I have no problems in enabling by default.
mmm no!, on systems without serial port, serial-getty will block startup until some timeout is reached.
Try this starting it without serial ports, (ie on kvm -serial no)
Awww :(
There are practically no systems without a serial port, but this is a showstopper. This will still work with console=ttyS0 (or console=hvc0).
But can be disabled by default (or ondemand) on firmware configuration if motherboard has a pin-socket for it (is rare to find today a motherboard with the "fichita" DB-9 on the panel)
Notebook, netbooks...no serial.
I suggest to postpone any change to this to next month. f there are no objections I'll tag a new archiso release tonight.
Greetings,
Pierre
Sure go ahead. Thanks :) -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1
On 10/28/2012 05:22 AM, Pierre Schmitz wrote:
Hi all,
I have reviewed and tested the latest patches and all looks fine so far. There are no big changes which I don't dislike. :-) As discussed before, I would like to have a few days between tagging and the actual iso release. So how about we tag the current state today, test and package it. I was thinking about building the new iso on Thursday and announce it on Friday (if we still want to have announcements).
I guess it then would be best to not commit new changes or put them into another branch; we have an unused testing branch atm. Or we introduce a new maint branch for releases.
Greetings,
Pierre
User visible changes from archiso POV: * Introduce cowfile_size= bootparam mainly for persistent COW on VFAT. * Indirectly with syslinux 4.06 finally fixed a long-standing bug on VFAT and relative directories. So workaround for syslinux.cfg when transfering ISO to VFAT is not needed anymore. * Allow (but not supposed to work propertly) copytoram=n for PXE + NFS/NBD, needed for lowmem systems (< 512MB). -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1
Am 30.10.2012 10:47, schrieb Gerardo Exequiel Pozzi:
* Indirectly with syslinux 4.06 finally fixed a long-standing bug on VFAT and relative directories. So workaround for syslinux.cfg when transfering ISO to VFAT is not needed anymore.
This affects the syslinux installer that is used when generating the USB - it is not on the ISO itself.
participants (4)
-
Gerardo Exequiel Pozzi
-
Ionut Biru
-
Pierre Schmitz
-
Thomas Bächler