On Oct 6, 2012, at 4:05 PM, Michal Migurski wrote:
> On Oct 6, 2012, at 3:44 PM, Frederik Ramm wrote:
>
>> If you write it this way, the 2nd line cancels out the 1st. You need to write
>>
>> JAVACMD_OPTIONS="-server -Xmx16G"
>>
>> - if your two lines had been the other way round then that would likely been
>> the cause of the problem, however since in your setting it is the -server
>> that is cancelled out, this likely won't help much ;(
>
>
> I get it, they replace instead of append. Thanks Frederick, I'll try it out!
>
> (settles in for a few-hour wait)
So, I'm still getting heap space failures right at the end of the read process
when the relations have been parsed. I've switched to EC2 for the time being
and tried a variety of settings for the -Xmx option, from 14G on a 16G system
to just 2G.
My command as before looks like this:
osmosis \
--rb planet-121003.osm.pbf \
--log-progress \
--tf accept-ways 'highway=*' --tf accept-relations 'network=*'
--used-node \
--tp user=osm da tabase=osm \
--wp user=osm database=osm
It takes about 3-5 hours each time to get it to fail. Is there a better way for
me to be debugging this? My current ~/.osmosis file looks like this:
JAVACMD_OPTIONS='-server -Xmx8G -Djava.io.tmpdir=/mnt/planet/tmp'
What am I missing here?
-mike.
----------------------------------------------------------------
michal migurski- [email protected]
415.558.1610
_______________________________________________
talk mailing list
[email protected]
http://lists.openstreetmap.org/listinfo/talk
