Thanks all for the input.<br>The "cluster config" i mean the fs cluster size, i think that is 4k on a default ext2/3 fs<br><br><br><div class="gmail_quote">On Tue, Jan 10, 2012 at 4:52 PM, Phil! Gold <span dir="ltr"><<a href="mailto:phil_g@pobox.com">phil_g@pobox.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">* ciprian niculescu <<a href="mailto:cnicules@gmail.com">cnicules@gmail.com</a>> [2012-01-09 16:30 +0100]:<br>
> Nice thread on space, but what about inode number, on my server i get fast<br>
> to 100% inode used.<br>
> So what filesystems do you use, in what cluster config.<br>
<br>
XFS (as Grant noted) and JFS both do well with large numbers of allocated<br>
inodes. ReiserFS doesn't appear to have an inode limit at all, but I tend<br>
to avoid it these days, mostly for reasons of personal experience: I've<br>
had disk failures that damaged XFS, JFS, and ReiserFS filesystems, and<br>
have had the best experiences recovering data from damaged JFS<br>
filesystems.<br>
<br>
I'm not sure what you mean by "cluster config", but I keep my tiles (for<br>
very low-demand usage) in an LVM logical volume that sits on a RAID1<br>
physical volume. I don't do any disk spanning tricks for them. My<br>
PostgreSQL rendering database, however, spans four 10k RPM disks in a<br>
RAID0 configuration.<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
...computer contrarian of the first order... / <a href="http://aperiodic.net/phil/" target="_blank">http://aperiodic.net/phil/</a><br>
PGP: 026A27F2 print: D200 5BDB FC4B B24A 9248 9F7A 4322 2D22 026A 27F2<br>
--- --<br>
Let's call it an accidental feature.<br>
-- Larry Wall<br>
---- --- --<br>
<br>
_______________________________________________<br>
talk mailing list<br>
<a href="mailto:talk@openstreetmap.org">talk@openstreetmap.org</a><br>
<a href="http://lists.openstreetmap.org/listinfo/talk" target="_blank">http://lists.openstreetmap.org/listinfo/talk</a><br>
</font></span></blockquote></div><br>