So, I turned on some debugging, and it looks as if it's done reading all
the data. However, it's just sitting there now not making any more
system calls (nothing shows up with a truss). I'm assuming is going
through and analyzing the data? Does it do it while it's reading, or
after all the data has been read? If the latter, is there any way to
have it print out status while it's analyzing so we can see if it's hung
or still churning. Right now, the process is is major swap land and
it's probably going to take forever for it to finish. Running anlog on
a bigger box with tons of memory resulted in a 'Ran out of memory' error
-- which is wierd as there was pleanty left. I'm wondering if it's some
OS setting to limit the memory per process, but 'limit' indicated that a
lot of settings were 'unlimited'. We're awaiting a sun 250 to do some
more testing on later this week.
If there is no way to have it print out the status as it's curning
through the data, what's the relevant functions so I can modify the
source to print out debugging infomration?
Thanks,
Reza
Reza Naima sent me..
> It's not running out of memeory, and it's not over NFS -- one box had
> local storage and the other was on a SAN (stoarge area network). It
> just slows down. I'm playing around with various LOWMEM options to see
> how it progresses.. I'll post the results.
>
> -Reza
>
> Jeremy Wadsack sent me..
> > Reza Naima ([EMAIL PROTECTED]):
> >
> > > I'm trying to parse one day's worth of logs (~24G unzipped) on mutliple
> > > boxes, and it seems that the program locks up towards the end of the
> > > analysis. This was tried on several solaris boxes including a blade 100
> > > and a 4500 with 6G of memory. A truss of the process shows that it's no
> > > longer making any system calls -- however, it doesn't seem to have
> > > finished reading in all the logs..
> >
> > > Does anyone have a clue what could be causing this? If you run a
> > > debugger on the code, it seems as if it's not stepping through the code
> > > anymore.
> >
> >
> > Are these logs being read across an NFS mount? I've seen 4500's lock
> > up when trying to move lots of logs across a mount point that couldn't
> > take the load.
> >
> > That, and try the LOWMEM commands (docs/lowmem.html) and/or adding
> > more memory to the box.
> >
> > --
> >
> > Jeremy Wadsack
> > Wadsack-Allen Digital Group
> >
> > +------------------------------------------------------------------------
> > | This is the analog-help mailing list. To unsubscribe from this
> > | mailing list, go to
> > | http://lists.isite.net/listgate/analog-help/unsubscribe.html
> > |
> > | List archives are available at
> > | http://www.mail-archive.com/[email protected]/
> > | http://lists.isite.net/listgate/analog-help/archives/
> > | http://www.tallylist.com/archives/index.cfm/mlist.7
> > +------------------------------------------------------------------------
>
> --
> /"\
> \ / ASCII Ribbon Campaign
> X Against HTML Mail
> / \ and Postings
> +------------------------------------------------------------------------
> | This is the analog-help mailing list. To unsubscribe from this
> | mailing list, go to
> | http://lists.isite.net/listgate/analog-help/unsubscribe.html
> |
> | List archives are available at
> | http://www.mail-archive.com/[email protected]/
> | http://lists.isite.net/listgate/analog-help/archives/
> | http://www.tallylist.com/archives/index.cfm/mlist.7
> +------------------------------------------------------------------------
--
/"\
\ / ASCII Ribbon Campaign
X Against HTML Mail
/ \ and Postings
+------------------------------------------------------------------------
| This is the analog-help mailing list. To unsubscribe from this
| mailing list, go to
| http://lists.isite.net/listgate/analog-help/unsubscribe.html
|
| List archives are available at
| http://www.mail-archive.com/[email protected]/
| http://lists.isite.net/listgate/analog-help/archives/
| http://www.tallylist.com/archives/index.cfm/mlist.7
+------------------------------------------------------------------------
