Andreas Israel
2005-03-18 21:20:59 UTC
Dear list,
today if have explored an "error" in plot_accesslog plugin.
It is not really an error, but in the moment I can't say which
implications this has on logging_engine_plugin.
I browsed my old data and have seen, that in the months in which the
clock changes from summer to winter time and back (october and march
normally), there is a problem in months view, because of the hour which
is to much or less.
The october-chart is ending on 1st November 22.59.59. In the march-chart
this isn't "visible", because the summer-wintertime-change is not on the
last day. But I know its there, the x_label array is one item to long.
So it is not really a bug and if, it is not big. It is "verschmerzbar",
because this happens only twice a year. But what happens with this 1
hour in the logging engine. Or is it not a pot_plot_acceslog_phenomenon,
and it only happens on my computers.
Does anyone have same experiences or solution ideas?
Andreas
PS: I have tested this with current and very old 1.1.5 versions.
today if have explored an "error" in plot_accesslog plugin.
It is not really an error, but in the moment I can't say which
implications this has on logging_engine_plugin.
I browsed my old data and have seen, that in the months in which the
clock changes from summer to winter time and back (october and march
normally), there is a problem in months view, because of the hour which
is to much or less.
The october-chart is ending on 1st November 22.59.59. In the march-chart
this isn't "visible", because the summer-wintertime-change is not on the
last day. But I know its there, the x_label array is one item to long.
So it is not really a bug and if, it is not big. It is "verschmerzbar",
because this happens only twice a year. But what happens with this 1
hour in the logging engine. Or is it not a pot_plot_acceslog_phenomenon,
and it only happens on my computers.
Does anyone have same experiences or solution ideas?
Andreas
PS: I have tested this with current and very old 1.1.5 versions.