Am Montag, 16. Dezember 2019, 21:43:21 CET schrieb Takashi Iwai:
On Mon, 16 Dec 2019 21:35:05 +0100,
Takashi Iwai wrote:
On Mon, 16 Dec 2019 20:40:48 +0100,
Felix Miata wrote:
Hans-Peter Jansen composed on 2019-12-16 16:59 (UTC+0100):
While transitioning a 15.0 server (my '*' box) to 15.2 (the sed/zypper way), I stumbled across this:
The following 2 packages are going to be downgraded: kernel-firmware
20191023-lp150.2.25.1 -> 20190618-lp152.1.16
ucode-intel
20191115-lp150.2.36.1 -> 20191112a-lp152.1.1
I *guess*, that not to be expected, as it points to packages, where the update process is somewhat muddled up. Since this is an delicate area as well, it might be reasonable to investigate a bit further...
I suspect 15.2's firmware currency is being consciously delayed pending complete resolution of https://bugzilla.opensuse.org/show_bug.cgi?id=1143959 for TW before bringing that significant change into 15.2 alpha.
Well, it seems to be rather an issue in SLE15-SP1 kernel-firmware package. By some reason, it was wrongly forked and doesn't follow SLE15-Update any longer. The very same problem is seen on Leap 15.1.
Now I reported the problem for SLE side. Hopefully it'll be addressed soon.
FYI, Leap 15.2 will keep SLE15-Update kernel-firmware, hence it'll be the traditional all-in-one plain package, not the split/compressed firmware packages on the recent TW.
Thanks Takashi San for pushing that to the people in charge. @Carlos and @Felix no need to worry. I do have safety nets in place, calculated the risk beforehand, and have been prepared for any emergency surgery (as usual). The major reason for this move was the updated kernel version. The transition went absolutely smooth. Now, let's see, if this system still suffers from https://community.asterisk.org/t/asterisk-16-5-1-on-sip-trunk-stops-operatin... that usually takes days to trigger. BTW, comparing the changelog of nscd from 15.0 and 15.2 reveals a similar, but not that obvious situation: Distribution: openSUSE Leap 15.0 * Thu Jun 13 2019 Andreas Schwab <schwab@suse.de> - regex-parse-reg-exp.patch: ERE '0|()0|\1|0' causes regexec undefined behavior (CVE-2009-5155, bsc#1127223, BZ #18986) - regex-read-overrun.patch: regex: fix read overrun (CVE-2019-9169, bsc#1127308, BZ #24114) * Thu Jun 06 2019 Andreas Schwab <schwab@suse.de> - crt-nocompress-debug-sections.patch: Don't compress debug sections in crt*.o files (bsc#1123710) * Wed Jun 05 2019 Andreas Schwab <schwab@suse.de> - ldconfig-concurrency.patch: Avoid concurrency problem in ldconfig (bsc#1117993, BZ #23973) * Tue Jun 04 2019 Andreas Schwab <schwab@suse.de> - force-elision-race.patch: Fix race in pthread_mutex_lock while promoting to PTHREAD_MUTEX_ELISION_NP (bsc#1131330, BZ #23275) * Wed Apr 03 2019 Andreas Schwab <schwab@suse.de> - japanese-era-name-may-2019.patch: ja_JP locale: Add entry for the new Japanese era (bsc#1100396, BZ #22964) Distribution: openSUSE Leap 15.2 * Wed Apr 03 2019 Andreas Schwab <schwab@suse.de> - japanese-era-name-may-2019.patch: ja_JP locale: Add entry for the new Japanese era (bsc#1100396, BZ #22964) I would expect nscd of 15.2 to carry the latest fixes in the Leap series. Cheers, Pete -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org