https://jira.whamcloud.com This file is an XML representation of an issue en-us 9.4.14 940014 05-12-2023 https://jira.whamcloud.com/browse/LU-8747 Lustre <p>In <a href="https://jira.whamcloud.com/browse/LU-6943" title="Get rid of cl_req in CLIO" class="issue-link" data-issue-key="LU-6943"><del>LU-6943</del></a> the <tt>osc_read</tt> and <tt>osc_write</tt> are removed.<br/> This information was very important to understand how an application was using the local cache versus sending data over the network.<br/> This is was an example:</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>snapshot_time 1454067781.979713 secs.usecs read_bytes 100000 samples [bytes] 4096 4096 409600000 write_bytes 100000 samples [bytes] 4096 4096 409600000 osc_read 401 samples [bytes] 4096 1048576 409600000 osc_write 391 samples [bytes] 655360 1048576 409600000 open 3 samples [regs] close 3 samples [regs] seek 2 samples [regs] truncate 1 samples [regs] getxattr 100000 samples [regs] inode_permission 7 samples [regs] </pre> </div></div> <p>After the <a href="https://jira.whamcloud.com/browse/LU-6943" title="Get rid of cl_req in CLIO" class="issue-link" data-issue-key="LU-6943"><del>LU-6943</del></a>, I'm assuming those informations are available on each stats under each osc: <tt>osc.*.stats</tt></p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>osc.zlfs2-OST001f-osc-ffff8820254c4000.stats= snapshot_time 1477130565.324299 secs.usecs req_waittime 1584 samples [usec] 170 7238 2291201 3700860537 req_active 1584 samples [reqs] 1 2 1618 1686 read_bytes 432 samples [bytes] 983 1048576 306500567 291188566310545 write_bytes 1148 samples [bytes] 983 1048576 1181807521 1239149817647387 ost_read 432 samples [usec] 170 4809 575487 981550847 ost_write 1148 samples [usec] 255 6154 1707192 2666192812 ldlm_cancel 1 samples [usec] 264 264 264 69696 obd_ping 2 samples [usec] 247 773 1020 658538 </pre> </div></div> <p>I'm assuming the old <tt>osc_read == SUM(read_bytes in osc.*.stats)</tt>. Could you please confirm?</p> <p>Normally the <tt>SUM(read_bytes in osc.&#42;.stats)</tt> should be = of the <tt>read_bytes in llite.&#42;.stats</tt> when the application is reading data from the network or &lt; when the application is reading data already in the VFS cache. Shouldn't be possible to have a situation where <tt>SUM(read_bytes in osc.&#42;.stats)</tt> &gt; <tt>read_bytes in llite.&#42;.stats</tt> in theory.</p> <p>I did some experiment and I found that the <tt>SUM(read_bytes in osc.&#42;.stats)</tt> &gt; <tt>read_bytes in llite.&#42;.stats</tt>. Could you explain why?</p> LU-8747

OSC_READ/WRITE replacement after LU-6943

Question/Request Major Open Unresolved WC Triage Gabriele Paciucci Sat, 22 Oct 2016 09:56:59 +0000 Wed, 26 Oct 2016 17:19:49 +0000 0 3 <p>I did a simple test:</p> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>lctl set_param llite.*.stats=0; lctl set_param osc.*.stats=0 </pre> </div></div> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>dd if=/dev/zero of=pippo bs=1M count=1000 </pre> </div></div> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>[root@broadwell1 SIDRA]# lctl get_param llite.*.stats |grep write_bytes write_bytes 1000 samples [bytes] 1048576 1048576 1048576000 </pre> </div></div> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>lctl get_param osc.*.stats |grep write_bytes write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 32 samples [bytes] 1048576 1048576 33554432 35184372088832 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 write_bytes 31 samples [bytes] 1048576 1048576 32505856 34084860461056 </pre> </div></div> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>lctl set_param llite.*.stats=0; lctl set_param osc.*.stats=0 </pre> </div></div> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>[root@broadwell1 SIDRA]# dd of=/dev/null if=pippo bs=1M count=1000 1000+0 records in 1000+0 records out 1048576000 bytes (1.0 GB) copied, 0.288931 s, 3.6 GB/s </pre> </div></div> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>[root@broadwell1 SIDRA]# lctl get_param llite.*.stats|grep read_bytes read_bytes 1000 samples [bytes] 1048576 1048576 1048576000 </pre> </div></div> <div class="preformatted panel" style="border-width: 1px;"><div class="preformattedContent panelContent"> <pre>lctl get_param osc.*.stats|grep read_bytes </pre> </div></div> <p>empty....</p> <p>So the application in my first question was reading data using a different call from <tt>read()</tt> and this explain why <tt>SUM(read_bytes in osc.&#42;.stats)</tt> &gt; <tt>read_bytes in llite.&#42;.stats</tt>.</p> <p>In <tt>llite.*.stats</tt> the {<tt>read,write}_bytes</tt> stats only count the number of bytes returned from some variant of <tt>read()</tt> or <tt>write()</tt>. If those reads hit the cache then the corresponding osc stats will remain unchanged. </p> Development Rank 1|hzyt3z: Rank (Obsolete) 9223372036854775807