prevalence of "right" time zones without timegm()?

[Tom Yu]

Dave Daugherty <dave.daugherty at centrify.com> writes:

> Adding Zihao from Centrify
>
> Tom, we plan to address this issue in our next round of development fixes - my guess is in the next couple of months.
>
> It will probably be Zihao's team implementing/testing the fix.
>
> As I said we will test this on many different platforms and versions - AIX, HPUX, Solaris, OS/X and many forms of Linux.
>
> Do you have a fix already that you would like us to test, or would you like us to take a run at it?

http://krbdev.mit.edu/rt/Ticket/Display.html?id=6928
http://src.mit.edu/fisheye/changelog/krb5/?cs=25018

The fix should be going into krb5-1.10.  It should not be too
difficult to backport to earlier releases, and we will consider
including it in patch releases to earlier full releases if requested.
More information would be appreciated on the existence of platforms
that have a gmtime() that is aware of leap seconds but that lack a
timegm().