Hey guys,<br><br><div class="gmail_quote">On 14 March 2012 08:40, Dave Chinner <span dir="ltr"><<a href="mailto:david@fromorbit.com">david@fromorbit.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On Tue, Mar 13, 2012 at 10:27:48AM -0500, Ben Myers wrote:<br>
> On Tue, Mar 13, 2012 at 04:42:42AM -0400, Christoph Hellwig wrote:<br>
> > On Mon, Mar 12, 2012 at 12:55:19PM -0500, Ben Myers wrote:<br>
> > > > +static int xqm_proc_show(struct seq_file *m, void *v)<br>
> > > > +{<br>
> > > > + /* maximum; incore; ratio free to inuse; freelist */<br>
> > > > + seq_printf(m, "%d\t%d\t%d\t%u\n",<br>
> > > > + 0,<br>
> > > > + counter_val(XFSSTAT_END_XQMSTAT),<br>
> > ><br>
> > > This patch looks great. It might break some PCP scripts, I'm not sure.<br>
> ><br>
> > Why would it break them? The existing stats are still kept around, and<br>
> > new additions to the main stats file are expected to happen once in a<br>
> > while.<br>
><br>
> I'm not sure that it would. You've added a line to the common stats<br>
> procfile, and the pmda might expect the xpc line immediately after ibt2.<br></div></blockquote><div><br>In an amusing twist, it turns out that the quota stats were originally part of<br>the main procfs stats file and were later on moved into their own file as<br>
part of making xqm modular. So, by good fortune PCP already supports<br>a move back into the original stats file for these stats, provided the same<br>prefix ("qm") is used.<br> </div><blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div class="im">
<br>
</div>IIRC the pmda triggers off the descriptor at the start of the line.<br></blockquote><div><br>That's correct.<br> </div><blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
It ignores lines that it doesn't understand the descriptor for (like<br>
all the v2 btree stats). I should probably update the pmda fo all<br>
those stats....<br></blockquote><div><br>That was done quite awhile back (mid-2010), unless there's some other<br>stats come along since then that I don't know about yet? (are these the<br>ones you mean, Dave?...)<br>
<br></div></div>$ egrep '^[a-z]+2 ' /proc/fs/xfs/stat <br>abtb2 4769967 54332939 635853 643428 507 508 282413 22672 61298 45587 1093 1111 1600 1619 393677507<br>abtc2 10166604 113971561 4264119 4272402 1213 1214 77487 9464 29584 35760 1551 1571 2764 2785 1438715558<br>
bmbt2 152523 1231148 52950 42558 0 0 1372 1385 5117 1340 85 48 85 48 1545160<br>ibt2 9475130 102652889 5672 2878 0 0 6549337 15214 801 869 17 0 17 0 1202730<br>$ <br><br>( those are the xfs.btree.alloc_blocks.*, xfs.btree.alloc_contig.*,<br>
xfs.btree.block_map.*, and xfs.btree.inode.lookup.* metrics )<br><br>cheers.<br><br>--<br>Nathan<br>