<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">2014-08-25 17:08 GMT+08:00 Dave Chinner <span dir="ltr"><<a href="mailto:david@fromorbit.com" target="_blank">david@fromorbit.com</a>></span>:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="">On Mon, Aug 25, 2014 at 04:47:39PM +0800, Zhang Qiang wrote:<br>
> I have checked icount and ifree, but I found there are about 11.8 percent<br>
> free, so the free inode should not be too few.<br>
><br>
> Here's the detail log, any new clue?<br>
><br>
> # mount /dev/sda4 /data1/<br>
> # xfs_info /data1/<br>
> meta-data=/dev/sda4 isize=256 agcount=4, agsize=142272384<br>
<br>
</div>4 AGs<br></blockquote><div>Yes. </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class=""><br>
> icount = 220619904<br>
> ifree = 26202919<br>
<br>
</div>And 220 million inodes. There's your problem - that's an average<br>
of 55 million inodes per AGI btree assuming you are using inode64.<br>
If you are using inode32, then the inodes will be in 2 btrees, or<br>
maybe even only one.<br></blockquote><div> </div><div>You are right, all inodes stay on one AG.</div><div><br></div><div>BTW, why i allocate 4 AGs, and all inodes stay in one AG for inode32?, sorry as I am not familiar with xfs currently.</div>
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
Anyway you look at it, searching btrees with tens of millions of<br>
entries is going to consume a *lot* of CPU time. So, really, the<br>
state your fs is in is probably unfixable without mkfs. And really,<br>
that's probably pushing the boundaries of what xfsdump and<br>
xfs-restore can support - it's going to take a long tiem to dump and<br>
restore that data....<br></blockquote><div> </div><div> Thanks reasonable.</div><div><br></div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
With that many inodes, I'd be considering moving to 32 or 64 AGs to<br>
keep the btree size down to a more manageable size. The free inode </blockquote><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
btree would also help, but, really, 220M inodes in a 2TB filesystem<br>
is really pushing the boundaries of sanity.....<br></blockquote><div> </div><div>So the better inodes size in one AG is about 5M, is there any documents about these options I can learn more?</div><div><br></div><div>I will spend more time to learn how to use xfs, and the internal of xfs, and try to contribute code.</div>
<div><br></div><div>Thanks for your help.</div><div><br></div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im HOEnZb"><br>
Cheers,<br>
<br>
Dave.<br>
--<br>
Dave Chinner<br>
<a href="mailto:david@fromorbit.com">david@fromorbit.com</a><br>
<br>
</div><div class="HOEnZb"><div class="h5">_______________________________________________<br>
xfs mailing list<br>
<a href="mailto:xfs@oss.sgi.com">xfs@oss.sgi.com</a><br>
<a href="http://oss.sgi.com/mailman/listinfo/xfs" target="_blank">http://oss.sgi.com/mailman/listinfo/xfs</a><br>
</div></div></blockquote></div><br></div></div>