Having trouble un bzip2 file

I tired to search for this in the search forum section, but I have been unable to locate what I need.
I will explain what I have done and what system I am using and I am hoping that someone can help me.

First I have Debian Jessie installed on my server with no GUI.
I have a quad core AMD FX™-4100 Quad-Core Processor at 3.6Ghz.
I also have 16 Gb of ram and 32 GB of swap.
I ran this command twice to get the world map.

wget -bc -o /home/rehome/osmPlanet/planetosm_wget.log http://ftp.osuosl.org/pub/openstreetmap/planet/planet-latest.osm.bz2

I have tried to unzip the file with the following command:

bzip2 -sdk planet-latest.osm.bz2 > osmbzip2.txt 2>&1 &

I also tried it with out the slim mode.
It has said the the file is corrupt numerous times. I tried re-downloading the file, with the same results.
I read somewhere that if the file is too large that the bzip2 can create segment faults in the memory.
The thing I do not understand is that is is using all my memory for this. I read up on memory management, but it indicated that on decompression it is not relevant.

I recall having this problem a few years ago the last time I downloaded the file and I needed to run it a few times to get it to work. However with the file size I am thinking that maybe it is causing problems with the system.
Is it possible that it is my system and not the download. I run bzip2 -t and it said it was corrupted too.
Wget said that everything was fine when it was downloaded both times. The file is 51G, which I read is smaller then the latest copy. However, both times the download was the same size.
I don’t want to keep re-downloading the file because I don’t want to be hitting their server over and over, if it is not necessary and I only have so much bandwidth allowed to me.
Please let me know what you think.
Thanks,
Donald

P.S. I hope I have put everthing in here that you will need.

You might have better luck at help.openstreetmap.org

May I ask why you’re trying to unzip a large .bz2 file into what would be a humungous .xml file?

Does the md5sum of what you’ve downloaded match what’s at http://planet.openstreetmap.org/planet/ ?

I was unzipping it so I could use osm2pgsql and Nominatim for street locations.
I was unaware that you could use osm2pgsql with .PBF. I thought after I started the import into the database I found out that you can use it.
Please correct me if I am wrong there.
I have not had a chance to check if Nominatim will be able to use that format. I have not looked into it as I was successful in unzipping the file and now have it downloaded and have no intention on dowloading it again, though I will be using the format in the future if I can.
I can tell you that I was able to do it with renice or nice set to 19.
If you happen to know while we are chatting how to do it with .PDF I would love to know, so that when I set up my production server it will speed things up a lot.
The last time I did this a few years ago, maybe more I do not think .PDF was around.
I am still puzzled though why bzip2 kept reporting that the file was corrupt when it was not.
And yes the md5sum was correct and reported OK.
Thank you,
Donald

For info, http://wiki.openstreetmap.org/wiki/Nominatim/Installation says “Using a file in PBF format is recommended”.

As I understand it (from previous mailing list posts) the reason is that the PBF processing is actually faster than bzip2 / xml processing.

Thank you. I will use this in the future.
I will consider this matter resolved.
Thank you for your help, you have been great.
I do not know how to mark this as solved.
But once again thank you,
Donald Boissonneault

Unlike some other forums, I don’t think it has a “mark as solved” option, so don’t worry about that.