Due 20 Jan 2010.
vnstat-1.10 contains some fixes for bugs found in 1.9 but not that many new features so it should be considered as a better candidate for a stable version. List of fixes from the changelog: - Fix: Buffer overflow was possible in hourly image output when RateUnit=1 and HourlyRate=1 - Fix: Minor memory leak was possible in the handling of HUP signal in daemon - Fix: Graphical elements weren't correctly aligned in summary image when header wasn't visible (-nh) - Fix: --delete didn't work
1.9 will have been in the tree for 30 days on the 20th Jan and 1.10 for about 17 days. Gentoo QA policy says we need to have a version for around 30 days, so you'd need to come up with something very special to not stabilise 1.9 first. The list of changes[1] indicates that 1.9 fixes quite a few things relative to 1.8, and every older version fixes a number of problems with its predecessor, all the way back to the now ancient 1.6 which is the last one Gentoo stabilised. Arguing for 1.10 to go stable is exactly as good as arguing for 1.9, in that respect, and 1.11 might fix problems found in 1.10, but in your reasoning that shouldn't stop us, which sounds rather illogical to me. So we'll go with 1.9 first and stabilise 1.10 in a different bug later, OK? [1] http://humdi.net/vnstat/CHANGES
(In reply to comment #2) > So we'll go with 1.9 first and stabilise 1.10 in a different bug later, OK? That's ok. I only wanted to point out the known problems with 1.9 so that it doesn't end up being the stable version for the next 2 years like 1.6 did. :)
Arch teams, please stabilise: =net-analyzer/vnstat-1.10
Stable for HPPA.
Marked ppc/ppc64 stable.
tested on x86, looks good here too.
Forgot to remove our CC's
sparc stable
Stable on amd64
x86 stable
arm will pass, closing
