<feed xmlns='http://www.w3.org/2005/Atom'>
<title>linux-toradex.git/net/ceph/crush, branch v3.19</title>
<subtitle>Linux kernel for Apalis and Colibri modules</subtitle>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/'/>
<entry>
<title>crush: add SET_CHOOSELEAF_VARY_R step</title>
<updated>2014-04-05T04:07:28+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2014-03-19T14:58:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=d83ed858f144e3cfbef883d4bc499113cdddabeb'/>
<id>d83ed858f144e3cfbef883d4bc499113cdddabeb</id>
<content type='text'>
This lets you adjust the vary_r tunable on a per-rule basis.

Reflects ceph.git commit f944ccc20aee60a7d8da7e405ec75ad1cd449fac.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This lets you adjust the vary_r tunable on a per-rule basis.

Reflects ceph.git commit f944ccc20aee60a7d8da7e405ec75ad1cd449fac.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: add chooseleaf_vary_r tunable</title>
<updated>2014-04-05T04:07:26+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2014-03-19T14:58:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=e2b149cc4ba00766aceb87950c6de72ea7fc8b2e'/>
<id>e2b149cc4ba00766aceb87950c6de72ea7fc8b2e</id>
<content type='text'>
The current crush_choose_firstn code will re-use the same 'r' value for
the recursive call.  That means that if we are hitting a collision or
rejection for some reason (say, an OSD that is marked out) and need to
retry, we will keep making the same (bad) choice in that recursive
selection.

Introduce a tunable that fixes that behavior by incorporating the parent
'r' value into the recursive starting point, so that a different path
will be taken in subsequent placement attempts.

Note that this was done from the get-go for the new crush_choose_indep
algorithm.

This was exposed by a user who was seeing PGs stuck in active+remapped
after reweight-by-utilization because the up set mapped to a single OSD.

Reflects ceph.git commit a8e6c9fbf88bad056dd05d3eb790e98a5e43451a.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The current crush_choose_firstn code will re-use the same 'r' value for
the recursive call.  That means that if we are hitting a collision or
rejection for some reason (say, an OSD that is marked out) and need to
retry, we will keep making the same (bad) choice in that recursive
selection.

Introduce a tunable that fixes that behavior by incorporating the parent
'r' value into the recursive starting point, so that a different path
will be taken in subsequent placement attempts.

Note that this was done from the get-go for the new crush_choose_indep
algorithm.

This was exposed by a user who was seeing PGs stuck in active+remapped
after reweight-by-utilization because the up set mapped to a single OSD.

Reflects ceph.git commit a8e6c9fbf88bad056dd05d3eb790e98a5e43451a.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: allow crush rules to set (re)tries counts to 0</title>
<updated>2014-04-05T04:07:25+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2014-03-19T14:58:37+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=6ed1002f368c63ef79d7f659fcb4368a90098132'/>
<id>6ed1002f368c63ef79d7f659fcb4368a90098132</id>
<content type='text'>
These two fields are misnomers; they are *retry* counts.

Reflects ceph.git commit f17caba8ae0cad7b6f8f35e53e5f73b444696835.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
These two fields are misnomers; they are *retry* counts.

Reflects ceph.git commit f17caba8ae0cad7b6f8f35e53e5f73b444696835.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: fix off-by-one errors in total_tries refactor</title>
<updated>2014-04-05T04:07:23+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2014-03-19T14:58:36+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=48a163dbb517eba13643bf404a0d695c1ab0a60d'/>
<id>48a163dbb517eba13643bf404a0d695c1ab0a60d</id>
<content type='text'>
Back in 27f4d1f6bc32c2ed7b2c5080cbd58b14df622607 we refactored the CRUSH
code to allow adjustment of the retry counts on a per-pool basis.  That
commit had an off-by-one bug: the previous "tries" counter was a *retry*
count, not a *try* count, but the new code was passing in 1 meaning
there should be no retries.

Fix the ftotal vs tries comparison to use &lt; instead of &lt;= to fix the
problem.  Note that the original code used &lt;= here, which means the
global "choose_total_tries" tunable is actually counting retries.
Compensate for that by adding 1 in crush_do_rule when we pull the tunable
into the local variable.

This was noticed looking at output from a user provided osdmap.
Unfortunately the map doesn't illustrate the change in mapping behavior
and I haven't managed to construct one yet that does.  Inspection of the
crush debug output now aligns with prior versions, though.

Reflects ceph.git commit 795704fd615f0b008dcc81aa088a859b2d075138.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Back in 27f4d1f6bc32c2ed7b2c5080cbd58b14df622607 we refactored the CRUSH
code to allow adjustment of the retry counts on a per-pool basis.  That
commit had an off-by-one bug: the previous "tries" counter was a *retry*
count, not a *try* count, but the new code was passing in 1 meaning
there should be no retries.

Fix the ftotal vs tries comparison to use &lt; instead of &lt;= to fix the
problem.  Note that the original code used &lt;= here, which means the
global "choose_total_tries" tunable is actually counting retries.
Compensate for that by adding 1 in crush_do_rule when we pull the tunable
into the local variable.

This was noticed looking at output from a user provided osdmap.
Unfortunately the map doesn't illustrate the change in mapping behavior
and I haven't managed to construct one yet that does.  Inspection of the
crush debug output now aligns with prior versions, though.

Reflects ceph.git commit 795704fd615f0b008dcc81aa088a859b2d075138.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Josh Durgin &lt;josh.durgin@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: fix crush_choose_firstn comment</title>
<updated>2013-12-31T18:32:28+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2013-12-24T19:19:27+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=0e32d7126cdf30b610bc131c81f131c717bd0d77'/>
<id>0e32d7126cdf30b610bc131c81f131c717bd0d77</id>
<content type='text'>
Reflects ceph.git commit 8b38f10bc2ee3643a33ea5f9545ad5c00e4ac5b4.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Reflects ceph.git commit 8b38f10bc2ee3643a33ea5f9545ad5c00e4ac5b4.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: attempts -&gt; tries</title>
<updated>2013-12-31T18:32:27+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2013-12-24T19:19:27+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=2d8be0bc8bc2dc7d3fc3d322c4db3fdf6c924660'/>
<id>2d8be0bc8bc2dc7d3fc3d322c4db3fdf6c924660</id>
<content type='text'>
Reflects ceph.git commit ea3a0bb8b773360d73b8b77fa32115ef091c9857.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Reflects ceph.git commit ea3a0bb8b773360d73b8b77fa32115ef091c9857.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: add set_choose_local_[fallback_]tries steps</title>
<updated>2013-12-31T18:32:26+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2013-12-24T19:19:27+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=f046bf92080cbdc4a94c6e86698c5a3f10716445'/>
<id>f046bf92080cbdc4a94c6e86698c5a3f10716445</id>
<content type='text'>
This allows all of the tunables to be overridden by a specific rule.

Reflects ceph.git commits d129e09e57fbc61cfd4f492e3ee77d0750c9d292,
                          0497db49e5973b50df26251ed0e3f4ac7578e66e.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This allows all of the tunables to be overridden by a specific rule.

Reflects ceph.git commits d129e09e57fbc61cfd4f492e3ee77d0750c9d292,
                          0497db49e5973b50df26251ed0e3f4ac7578e66e.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: generalize descend_once</title>
<updated>2013-12-31T18:32:25+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2013-12-24T19:19:26+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=d390bb2a83086f2b79c152e2c1734813bd257d9b'/>
<id>d390bb2a83086f2b79c152e2c1734813bd257d9b</id>
<content type='text'>
The legacy behavior is to make the normal number of tries for the
recursive chooseleaf call.  The descend_once tunable changed this to
making a single try and bail if we get a reject (note that it is
impossible to collide in the recursive case).

The new set_chooseleaf_tries lets you select the number of recursive
chooseleaf attempts for indep mode, or default to 1.  Use the same
behavior for firstn, except default to total_tries when the legacy
tunables are set (for compatibility).  This makes the rule step
override the (new) default of 1 recursive attempt, keeping behavior
consistent with indep mode.

Reflects ceph.git commit 685c6950ef3df325ef04ce7c986e36ca2514c5f1.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
The legacy behavior is to make the normal number of tries for the
recursive chooseleaf call.  The descend_once tunable changed this to
making a single try and bail if we get a reject (note that it is
impossible to collide in the recursive case).

The new set_chooseleaf_tries lets you select the number of recursive
chooseleaf attempts for indep mode, or default to 1.  Use the same
behavior for firstn, except default to total_tries when the legacy
tunables are set (for compatibility).  This makes the rule step
override the (new) default of 1 recursive attempt, keeping behavior
consistent with indep mode.

Reflects ceph.git commit 685c6950ef3df325ef04ce7c986e36ca2514c5f1.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: CHOOSE_LEAF -&gt; CHOOSELEAF throughout</title>
<updated>2013-12-31T18:32:24+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2013-12-24T19:19:26+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=917edad5d1d62070436b74ecbf5ea019b651ff69'/>
<id>917edad5d1d62070436b74ecbf5ea019b651ff69</id>
<content type='text'>
This aligns the internal identifier names with the user-visible names in
the decompiled crush map language.

Reflects ceph.git commit caa0e22e15e4226c3671318ba1f61314bf6da2a6.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
This aligns the internal identifier names with the user-visible names in
the decompiled crush map language.

Reflects ceph.git commit caa0e22e15e4226c3671318ba1f61314bf6da2a6.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</pre>
</div>
</content>
</entry>
<entry>
<title>crush: add SET_CHOOSE_TRIES rule step</title>
<updated>2013-12-31T18:32:23+00:00</updated>
<author>
<name>Ilya Dryomov</name>
<email>ilya.dryomov@inktank.com</email>
</author>
<published>2013-12-24T19:19:26+00:00</published>
<link rel='alternate' type='text/html' href='https://git.toradex.cn/cgit/linux-toradex.git/commit/?id=cc10df4a3a5c34cb1d5b635ac70dd1fc406153ce'/>
<id>cc10df4a3a5c34cb1d5b635ac70dd1fc406153ce</id>
<content type='text'>
Since we can specify the recursive retries in a rule, we may as well also
specify the non-recursive tries too for completeness.

Reflects ceph.git commit d1b97462cffccc871914859eaee562f2786abfd1.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Since we can specify the recursive retries in a rule, we may as well also
specify the non-recursive tries too for completeness.

Reflects ceph.git commit d1b97462cffccc871914859eaee562f2786abfd1.

Signed-off-by: Ilya Dryomov &lt;ilya.dryomov@inktank.com&gt;
Reviewed-by: Sage Weil &lt;sage@inktank.com&gt;
</pre>
</div>
</content>
</entry>
</feed>
