[Orca-users] Re: graphs not updating (although the data is) on a sol 2.8 host

[orcausers]

--- In orca-users at y..., "orcausers" <lsaladi at u...> wrote:
> --- In orca-users at y..., Blair Zajac <blair at o...> wrote:
> > orcausers wrote:
> > > The web graphs are being displayed - and the number of web
> servers is
> > > correct - as this is a test system and all the virtual hosts
> defined
> > > in the apache file write to their own CustomLog - I will assume
> there
> > > is no data to present for the other graphs - those are being
> changed
> > > to either eliminate them or TransferLog to access_log.
> >
> > orcallator.se only reads from a single log file, so you won't be
> > able to monitor the different virtual hosts if they each write to
> > their own log file.
> >
> > >
> > > I have a number of other servers I have to do and am trying to
> > > understand whether anything HAS to be done to get the web graphs
> > > other than the addition of WEB_SERVER, WATCH_WEB and WEB_LOG to
> the
> > > orcallator.se script and the yo yo of that on the server.
> >
> > Nope, that should be it.
> >
> > >
> > > do i have to yo yo the orca perl engine on my "repository"
server
> > > when i add the above to any server?
> >
> > If you want it to find the new source input files as soon as it
can
> > get around them, then either bounce Orca or better yet send it a
> > SIGHUP.  This tells it to find all the input files on its next
pass.
> >
> > > do i have to wait for the midnight turnover to a new percol
file
> for
> > > the graphs to be updated after this change? this will determine
> when
> > > i make the changes.
> >
> > Not if you use SIGHUP.  Also, the find_times configuration
parameter
> > tells Orca when to search for new files.
> >
> > Best,
> > Blair
> >
> > --
> > Blair Zajac <blair at o...>
> > Web and OS performance plots - http://www.orcaware.com/orca/
>
>
> life is good - it all works and hangs together. thanks much for the
> help - we appreciate the tool and the data it gives us. we make it
> available to sys admins, dbas, applications analysts - a large
range
> of staff - the more staff look at it - the more problems we can
> prevent. thanks!


when we  implemented the 1.32 ver/rev of the orcallator.se and .cfg 
files we went from percol files to orcallator files - is that
alright?
here are the files in the directory now -

-rw-r--r--   1 root     other      16707 May 29 23:55 orcallator-2002-
05-29-000.gz
-rw-r--r--   1 root     other     189631 May 30 10:20 orcallator-2002-
05-30-000
-rw-r--r--   1 root     other      11966 May 15 23:55 percol-2002-05-
15.gz
-rw-r--r--   1 root     other      42130 May 16 23:55 percol-2002-05-
16.gz
-rw-r--r--   1 root     other      38631 May 17 23:55 percol-2002-05-
17.gz
-rw-r--r--   1 root     other      24184 May 18 23:55 percol-2002-05-
18.gz
-rw-r--r--   1 root     other      25769 May 19 23:55 percol-2002-05-
19.gz
-rw-r--r--   1 root     other      38512 May 20 23:55 percol-2002-05-
20.gz
-rw-r--r--   1 root     other      36874 May 21 23:55 percol-2002-05-
21.gz
-rw-r--r--   1 root     other      37732 May 22 23:55 percol-2002-05-
22.gz
-rw-r--r--   1 root     other      39736 May 23 23:55 percol-2002-05-
23.gz
-rw-r--r--   1 root     other      39272 May 24 23:55 percol-2002-05-
24.gz
-rw-r--r--   1 root     other      24457 May 25 23:55 percol-2002-05-
25.gz
-rw-r--r--   1 root     other      22755 May 26 23:55 percol-2002-05-
26.gz
-rw-r--r--   1 root     other      27138 May 27 23:55 percol-2002-05-
27.gz
-rw-r--r--   1 root     other      38711 May 28 23:55 percol-2002-05-
28.gz
-rw-r--r--   1 root     other     236301 May 29 14:00 percol-2002-05-
29

second ? - can orcallator.se follow a logical link to the httpd
access_log?

$ ls -l /opt/ilcso/apache/logs
total 30
lrwxrwxrwx   1 root     other         54 May 30 07:27 access_log -
> /opt/ilcso/apache/logs/web/2002/05/access_log.20020530

with a WEB_LOG value of
WEB_LOG="/opt/ilcso/apache/logs/acess_log"