<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
The idea should be to make sure that maintaining the feature
supported matrix is not a nightmare. If we need to go and replicate
entries in 0000_config.sql file for each new version, then this is
an issue. And I think we are all in agreement that this is not the
way to go.<br>
<br>
Either we go with <br>
[1]. default value in ConfigValues, and only changed value in db
script like in patch set 5 (<a class="moz-txt-link-freetext"
href="http://gerrit.ovirt.org/#/c/12970/5">http://gerrit.ovirt.org/#/c/12970/5</a>)<br>
<br>
If this mechanism is broken, do you know where/what is broken?<br>
<br>
or [2] . with the new approach where a Feature supported changes to
true/false <b>*from*</b> a particular version.<br>
(I think, for gluster features, the Feature From works for us as we
do not see it changing from version to version once supported. )<br>
<br>
But if there's a way to fix 1, let's do that to get this moving.<br>
Mike, could you elaborate on what needs to be fixed in [1] ?<br>
<br>
<br>
<br>
<div class="moz-cite-prefix">On 04/02/2013 04:30 PM, Shireesh Anjal
wrote:<br>
</div>
<blockquote cite="mid:515ABA63.4070106@redhat.com" type="cite">
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<div class="moz-cite-prefix">On 04/02/2013 02:47 PM, Mike Kolesnik
wrote:<br>
</div>
<blockquote
cite="mid:1842623267.574910.1364894251176.JavaMail.root@redhat.com"
type="cite">
<div style="font-family: times new roman, new york, times,
serif; font-size: 12pt; color: #000000">
<hr id="zwchr">
<blockquote style="border-left:2px solid
#1010FF;margin-left:5px;padding-left:5px;color:#000;font-weight:normal;font-style:normal;text-decoration:none;font-family:Helvetica,Arial,sans-serif;font-size:12pt;"
data-mce-style="border-left: 2px solid #1010FF; margin-left:
5px; padding-left: 5px; color: #000; font-weight: normal;
font-style: normal; text-decoration: none; font-family:
Helvetica,Arial,sans-serif; font-size: 12pt;">
<div class="moz-cite-prefix">On 03/27/2013 05:48 PM, Mike
Kolesnik wrote:<br>
</div>
<blockquote
cite="mid:36623975.7044697.1364386718971.JavaMail.root@redhat.com">
<pre>----- Original Message -----
</pre>
<blockquote>
<pre>On 03/20/2013 08:20 PM, Yair Zaslavsky wrote:
</pre>
<blockquote>
<pre>----- Original Message -----
</pre>
<blockquote>
<pre>From: "Shireesh Anjal" <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:sanjal@redhat.com" target="_blank" data-mce-href="mailto:sanjal@redhat.com"><sanjal@redhat.com></a>
To: "Mike Kolesnik" <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:mkolesni@redhat.com" target="_blank" data-mce-href="mailto:mkolesni@redhat.com"><mkolesni@redhat.com></a>
Cc: <a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:engine-devel@ovirt.org" target="_blank" data-mce-href="mailto:engine-devel@ovirt.org">engine-devel@ovirt.org</a>
Sent: Wednesday, March 20, 2013 4:47:08 PM
Subject: Re: [Engine-devel] FeatureSupported and compatibility
versions
On 03/18/2013 01:11 PM, Shireesh Anjal wrote:
</pre>
<blockquote>
<pre>On 03/18/2013 12:59 PM, Mike Kolesnik wrote:
</pre>
<blockquote>
<pre>----- Original Message -----
</pre>
<blockquote>
<pre>Hi all,
The current mechanism in oVirt to check whether a feature is
supported
in a particular compatibility version is to use the
FeatureSupported
class. e.g.
FeatureSupported.networkLinking(getVm().getVdsGroupCompatibilityVersion())
Checks whether the "network linking" feature is supported for
the
the
VM's cluster compatibility version. This internally checks
whether
the
value of the corresponding config (NetworkLinkingSupported) for
the
given compatibility version is true/false.
I'm not sure if this is a good idea, since a feature is
typically
supported "from" a particular version. E.g. Gluster support was
introduced in 3.1, and it continues to be available in all
subsequent
versions. So I see no point in adding configuration for every
version
indicating whether the feature is supported in that version or
not. I
suggest to use either of the following options:
</pre>
</blockquote>
<pre>You can "merge" the configs into a single config when older
versions
go out of the supported versions for the system.
i.e. in 4.0 you can have upgrade script that merges all
GlusterFeatureSupported to one entry instead of several.
</pre>
</blockquote>
</blockquote>
<pre>Why are we even storing this information in config? Is this
something
that can be "configured" at customer site?
</pre>
</blockquote>
<pre>As previously explained (but off list :) ) , Config gives you the
ability to have a cachable "map" of entry (i.e - "feature name")
per version and value.
I guess it was convinient for the developers to use that.
I also mentioned that customers/oVirt users should config the
entries of vdc_options using engine-config tool only.
Not all entries are exposed via engine-config.properties (and no,
not just "is feature supported" entries are hidden).
</pre>
<blockquote>
<blockquote>
<blockquote>
<blockquote>
<pre>1) Instead of using a boolean config for each version, use a
single
string config that indicates the "supported from" version e.g.
GlusterSupportedFrom = 3.1. There could be rare cases where a
feature,
for some reason, is removed in some release. In such cases, we
could
use
one additional config for the "supported to" version.
2) Continue with the boolean approach, but do not have entries
for
every
version; rather make use of the "default value" for majority of
cases,
and add the explicit version mapping for the minority e.g.
GlusterSupported = true by default, and false in case of 3.0
(only
one
config required for 3.0)
</pre>
</blockquote>
<pre>I'm not sure why we would want to complicate this simple
mechanism?
Is there much to gain?
</pre>
</blockquote>
<pre>I think option 1 suggested above is simpler - to implement as
well
as
to understand.
Let me give you an example of why I don't like current mechanism.
I
introduce a version check for a feature that was introduced in
3.1.
I'm being asked now to add three entries in config
3.0 - false
3.1 - true
3.2 - true
It will also mean that when 3.3 goes out, someone has to make
sure
that another entry is added for 3.3-true. I think it is not
logical
as
well as scalable if you have more versions. And it sounds far
more
complex (to maintain) than just having
<Feature>SupportedFrom = 3.1
So I would like to know if there are any objections to my
proposal.
I
intend to use this for at least the gluster related features.
</pre>
</blockquote>
</blockquote>
</blockquote>
<pre>I've sent a patch (<a moz-do-not-send="true" class="moz-txt-link-freetext" href="http://gerrit.ovirt.org/12970" target="_blank" data-mce-href="http://gerrit.ovirt.org/12970">http://gerrit.ovirt.org/12970</a>) with following
changes:
1) Introduced CompatibilityUtils that provides utility methods for
checking if a given feature is supported in the config. One method to
check based on boolean values (as is being done today for virt
features), and nother to check based on a range (from, to) which I
would
like to use for gluster features.
2) Renamed FeatureSupported to VirtFeatureSupported, and made it use
the
first utility method from CompatibilityUtils
3) Introduced GlusterFeatureSupported for gluster features, which
uses
the second utility method from CompatibilityUtils
Key advantage here is that
- we don't have to touch any virt specifc source for adding
compatibility checks for gluster features
- virt features continue to use the existing boolean config check
Any comments / suggestions / reviews will be highly appreciated :)
</pre>
</blockquote>
<pre>I think splitting to two classes is OK, but the underlying mechanism IMO should be as Omer suggested:
Use the default value from the java config file, and if in the DB there is a version specific value then use it for that version only.</pre>
</blockquote>
<br>
Review comments here are on the contrary:<br>
<a moz-do-not-send="true"
href="http://gerrit.ovirt.org/#/c/12970/5/backend/manager/dbscripts/upgrade/pre_upgrade/0000_config.sql"
target="_blank"
data-mce-href="http://gerrit.ovirt.org/#/c/12970/5/backend/manager/dbscripts/upgrade/pre_upgrade/0000_config.sql">http://gerrit.ovirt.org/#/c/12970/5/backend/manager/dbscripts/upgrade/pre_upgrade/0000_config.sql</a><br
data-mce-bogus="1">
</blockquote>
<div>The comment in the review simply states that the
mechanism is probably broken, not that that's the way it has
to be done.<br>
</div>
</div>
</blockquote>
<br>
The comment explicitly asks me to add entries for every version.
What you have looked at is my response to this comment, which
suggests that the current mechanism is not great. In fact, what I
had done in patch-set 5 is exactly what you are suggesting : true
as default value and explicit entries in config for the "false"
values. But it was not accepted. <br>
<br>
<blockquote
cite="mid:1842623267.574910.1364894251176.JavaMail.root@redhat.com"
type="cite">
<div style="font-family: times new roman, new york, times,
serif; font-size: 12pt; color: #000000">
<blockquote style="border-left:2px solid
#1010FF;margin-left:5px;padding-left:5px;color:#000;font-weight:normal;font-style:normal;text-decoration:none;font-family:Helvetica,Arial,sans-serif;font-size:12pt;"
data-mce-style="border-left: 2px solid #1010FF; margin-left:
5px; padding-left: 5px; color: #000; font-weight: normal;
font-style: normal; text-decoration: none; font-family:
Helvetica,Arial,sans-serif; font-size: 12pt;"><br>
<blockquote
cite="mid:36623975.7044697.1364386718971.JavaMail.root@redhat.com">
<pre>I don't think "From, To, etc" is a good design, it's not a standard way and is very restrictive.</pre>
</blockquote>
<br>
Can you please explain in what way is it restrictive?<br>
<br>
Also, what is the "etc" you are referring to?</blockquote>
<div>What if for certain version it is not supported, you add
"except"? Or do you specify 2 ranges?<br>
</div>
<div><br>
</div>
<div>Starting to add from/to creates a limited design of one
range, which would be difficult to tune if necessary.</div>
</div>
</blockquote>
<br>
Really? Does someone really think that there will be a feature
that will be supported in multiple different ranges of versions? I
see zero possibility for this. I would love to see some +1s on
this concept before I can accept this argument.<br>
<br>
<blockquote
cite="mid:1842623267.574910.1364894251176.JavaMail.root@redhat.com"
type="cite">
<div style="font-family: times new roman, new york, times,
serif; font-size: 12pt; color: #000000">
<div>I think the design generally for config values is very
simple and suits us well - use the default value, unless a
specific version is configured differently.<br>
</div>
</div>
</blockquote>
<br>
I think the current design is wrong. A feature gets supported
"from" a particular version, and that's all that is required in
most of the cases. Expecting developers to add version-by-version
mapping for features is bad. The "to" part in my patch is just to
handle rare cases, if at all they come up. I'm willing to even
remove that if such a case doesn't exist today.<br>
<br>
Also, even though I have followed it for the sake of consistency,
I don't think these values need to be stored in the config (db) at
all. Only explanation I've got for it is that it was probably
'convenient' for developers to use the config mechanism. I'm for
having this check purely in code in a central place, and not the
config (db).<br>
<br>
<blockquote
cite="mid:1842623267.574910.1364894251176.JavaMail.root@redhat.com"
type="cite">
<div style="font-family: times new roman, new york, times,
serif; font-size: 12pt; color: #000000">
<div>This way you can specify the feature is supported, and
disable it for specific versions.<br>
</div>
</div>
</blockquote>
<br>
So one has to look at both code (FeatureSupported) as well as db
(config) to get an idea of what versions the feature is supported
in. Not great. <br>
<br>
<blockquote
cite="mid:1842623267.574910.1364894251176.JavaMail.root@redhat.com"
type="cite">
<div style="font-family: times new roman, new york, times,
serif; font-size: 12pt; color: #000000">
<div>I think this direction gives us the flexibility that we
would like to have.<br>
</div>
<div><br>
</div>
<div>Currently it doesn't work that way, but I think it's not
impossible to change, and more worthwhile than introducing a
new mechanism.<br>
</div>
</div>
</blockquote>
<br>
I disagree, and would like to use the "supported from" mechanism
at least for gluster features.<br>
<br>
<blockquote
cite="mid:1842623267.574910.1364894251176.JavaMail.root@redhat.com"
type="cite">
<div style="font-family: times new roman, new york, times,
serif; font-size: 12pt; color: #000000">
<blockquote style="border-left:2px solid
#1010FF;margin-left:5px;padding-left:5px;color:#000;font-weight:normal;font-style:normal;text-decoration:none;font-family:Helvetica,Arial,sans-serif;font-size:12pt;"
data-mce-style="border-left: 2px solid #1010FF; margin-left:
5px; padding-left: 5px; color: #000; font-weight: normal;
font-style: normal; text-decoration: none; font-family:
Helvetica,Arial,sans-serif; font-size: 12pt;"><br>
<blockquote
cite="mid:36623975.7044697.1364386718971.JavaMail.root@redhat.com">
<blockquote>
<blockquote>
<blockquote>
<blockquote>
<blockquote>
<blockquote>
<pre>Thoughts?
Regards,
Shireesh
</pre>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
</blockquote>
<br>
</blockquote>
<div><br>
</div>
</div>
</blockquote>
<br>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
Engine-devel mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Engine-devel@ovirt.org">Engine-devel@ovirt.org</a>
<a class="moz-txt-link-freetext" href="http://lists.ovirt.org/mailman/listinfo/engine-devel">http://lists.ovirt.org/mailman/listinfo/engine-devel</a>
</pre>
</blockquote>
<br>
</body>
</html>