[opensuse-factory] New snapshot delayed - but next one wil lbe GCC7 based
Dear Tumbleweed Users and Hackers, There has not been a new snapshot in a couple days. This is a combination of several factors: * Many contributors have been at the awesome openSUSE conference * On May 27, Tumbleweed has switched to GCC7 as its default compiler. After being in Staging for a long time, we felt confident to have ironed out the blocking issues * We were wrong :( Thanks to our brave Robby Engelmann (who updates a VM to 'Factory' at the time openQA is not yet done testing - so before we call it Tumbleweed), an issue could be identified just in time before the snapshot was released. The issue in short: mariadb did not like gcc7 that much; even though it built just fine, it actually crashed in situations. Even if you don't know you run a mariadb - if you're a KDE User and run akonadi, then there is a mariadb instance on your system and you would have suffered). The great Fabian Vogt took all the time needed to hunt down the issue and patched the package together to get this issue solved. The change has been checked in to Factory by now and a new snapshot (titled 0529) is currently building and will be reaching openQA in a couple hours. It is to be expected that this snapshot will then be ready for prime time and you should be receiving the update in the next day or so. A major update like GCC always take a lot of preparation time and, as it shows in this case, even then some things can slip through. Luckily, no damage was done to any system in the wild yet. I'd like to thank all the involved parties that helped chasing this issue today and that helped to get openSUSE Tumbleweed back into a rolling state. Future updates are supposed to roll in at a higher pace again once this gcc7 - based snapshot is out. Cheers, Dominique
Thanks for the honors :-) Actually, it was not a VM but my working notebook I regularly use for testing the to-be-released snapshots in real life. So yes, I am crazy. However, after checking that crypt-lvm and kde tests are fine in openqa, I very very rarely saw such an issues in the past. I even cannot remember the last time I halted a snapshot ;-) So, to all, lets keep it rolling. Cheers, Robby. -- On Dienstag, 30. Mai 2017 16:43:44 CEST Dominique Leuenberger / DimStar wrote:
Dear Tumbleweed Users and Hackers,
There has not been a new snapshot in a couple days. This is a combination of several factors:
* Many contributors have been at the awesome openSUSE conference * On May 27, Tumbleweed has switched to GCC7 as its default compiler. After being in Staging for a long time, we felt confident to have ironed out the blocking issues * We were wrong :(
Thanks to our brave Robby Engelmann (who updates a VM to 'Factory' at the time openQA is not yet done testing - so before we call it Tumbleweed), an issue could be identified just in time before the snapshot was released.
The issue in short: mariadb did not like gcc7 that much; even though it built just fine, it actually crashed in situations. Even if you don't know you run a mariadb - if you're a KDE User and run akonadi, then there is a mariadb instance on your system and you would have suffered).
The great Fabian Vogt took all the time needed to hunt down the issue and patched the package together to get this issue solved. The change has been checked in to Factory by now and a new snapshot (titled 0529) is currently building and will be reaching openQA in a couple hours.
It is to be expected that this snapshot will then be ready for prime time and you should be receiving the update in the next day or so.
A major update like GCC always take a lot of preparation time and, as it shows in this case, even then some things can slip through. Luckily, no damage was done to any system in the wild yet.
I'd like to thank all the involved parties that helped chasing this issue today and that helped to get openSUSE Tumbleweed back into a rolling state. Future updates are supposed to roll in at a higher pace again once this gcc7 - based snapshot is out.
Cheers, Dominique
On mardi, 30 mai 2017 16.43:44 h CEST Dominique Leuenberger / DimStar wrote:
The issue in short: mariadb did not like gcc7 that much; even though it built just fine, it actually crashed in situations. Even if you don't know you run a mariadb - if you're a KDE User and run akonadi, then there is a mariadb instance on your system and you would have suffered).
Not always, mine is using my postgresql :-) But anyway I've to remove the postgresql-plperl due to newer perl :-( (sorry not yet boo, tomorrow) -- Bruno Friedmann Ioda-Net Sàrl www.ioda-net.ch Bareos Partner, openSUSE Member, fsfe fellowship GPG KEY : D5C9B751C4653227 irc: tigerfoot -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (3)
-
Bruno Friedmann -
Dominique Leuenberger / DimStar -
Robby Engelmann