DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://issues.apache.org/bugzilla/show_bug.cgi?id=24417>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://issues.apache.org/bugzilla/show_bug.cgi?id=24417

[PATCH] rotatelogs with local time (including DST) functionality

[EMAIL PROTECTED] changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|CLOSED                      |REOPENED
         Resolution|FIXED                       |



------- Additional Comments From [EMAIL PROTECTED]  2004-06-16 16:19 -------
> Ken Coar recently committed an equivalent feature to 2.1-dev, which I presume
> will be backported to stable branch for a future 2.0.x release.  Please have 
> a look:

Ken's patch won't work correctly on the days when DST is switched on or off, 
because he puts the 
test for the local time offset (utc_offset = lt.tm_gmtoff;) outside of the "for 
(;;)" loop. Therefore, the 
time offset will only be set once after the start of the rotatelogs process, 
and will not be updated 
after a switch of DST occurs.

Ken himself writes in his comment: "NB: Using -l in an environment which 
changes the GMT offset 
(such as for BST or DST) can lead to unpredictable results!", suggesting he has 
indeed not tested his 
patch on days when the switch occurs. However, unless you intend to use DST 
there's no point at all 
in using local time instead of a fixed offset from GMT.

<Sigh> Why don't you just use the patch I committed? It's thoroughly tested, 
especially on days 
when the DST switch occurs, and proven to work in a production environment for 
almost two years 
now.

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to