Hi, I have just pushed a branch containing a number of bugfixes for the 3.13 branch, that I felt would be useful to be published in various syslog-ng binary packages over the existing 3.13.2 release that we produced a few weeks ago. https://github.com/balabit/syslog-ng/tree/3.13/master The most important change is this: https://github.com/balabit/syslog-ng/commit/7e96705ec0b3570857ed6cd66913999d... Which is bug that causes the elasticsearch2 destination SCL wrapper to fails to start up if any __VARARGS__ is used. (see https://github.com/balabit/syslog-ng/issues/1804). The other patches are just my selection of fixes that were deemed low risk to me. We expect the next official syslog-ng release to be 3.14 at the end of February, and I've felt this set of fixes would be useful even in the meantime. At this point, I take this as an experiment: if publishing fixes this way "works" (e.g. fixes get published in the OS packages that we don't maintain ourselves), then we can get some form of maintenance to older open source releases as time permits, with collaboration from our fellow OS maintainers. Also, power users, who are able to compile source code for themselves get help too. Let me know if this is useful. Cheers, -- Bazsi
Hi, On Tue, Jan 09, 2018 at 08:56:35AM +0100, Scheidler, Balázs wrote:
At this point, I take this as an experiment: if publishing fixes this way "works" (e.g. fixes get published in the OS packages that we don't maintain ourselves), then we can get some form of maintenance to older open source releases as time permits, with collaboration from our fellow OS maintainers. Also, power users, who are able to compile source code for themselves get help too.
Let me know if this is useful.
I love the idea, of course :-) That being said, and as it's related to the other major fix in the upcoming 3.13.2, I'd love to see this resolved: https://github.com/balabit/syslog-ng/pull/1800 In a nutshell, the elastic-v2 destination silently drops messages when the ES cluster is gone. I think this is serious enough to justify a bugfix release on its own. What's blocking the merge is that I can't get *any* java logs to pop up anywhere, thus I don't see what's wrong with the current implementation of the fix. Sorry for trying to bump this into 3.13.2, but I really think this is important.
Hi, I only cherry picked changes from current master, so I can only pick that patch if it is already fixed on master. On Jan 9, 2018 09:51, "Fabien Wernli" <wernli@in2p3.fr> wrote:
Hi,
On Tue, Jan 09, 2018 at 08:56:35AM +0100, Scheidler, Balázs wrote:
At this point, I take this as an experiment: if publishing fixes this way "works" (e.g. fixes get published in the OS packages that we don't maintain ourselves), then we can get some form of maintenance to older open source releases as time permits, with collaboration from our fellow OS maintainers. Also, power users, who are able to compile source code for themselves get help too.
Let me know if this is useful.
I love the idea, of course :-)
That being said, and as it's related to the other major fix in the upcoming 3.13.2, I'd love to see this resolved:
https://github.com/balabit/syslog-ng/pull/1800
In a nutshell, the elastic-v2 destination silently drops messages when the ES cluster is gone. I think this is serious enough to justify a bugfix release on its own.
What's blocking the merge is that I can't get *any* java logs to pop up anywhere, thus I don't see what's wrong with the current implementation of the fix.
Sorry for trying to bump this into 3.13.2, but I really think this is important.
____________________________________________________________ __________________ Member info: https://lists.balabit.hu/mailman/listinfo/syslog-ng Documentation: http://www.balabit.com/support/documentation/? product=syslog-ng FAQ: http://www.balabit.com/wiki/syslog-ng-faq
participants (2)
-
Fabien Wernli
-
Scheidler, Balázs