<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<meta name="Generator" content="Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
span.EmailStyle17
{mso-style-type:personal-compose;
font-family:"Calibri","sans-serif";
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri","sans-serif";}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal">Mike Bayer wrote:<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">> Well let me reiterate the idea, which is that:<o:p></o:p></p>
<p class="MsoNormal">> <o:p></o:p></p>
<p class="MsoNormal">> 1. we add features to oslo.db so that the use of a custom stored
<o:p></o:p></p>
<p class="MsoNormal">> function is not a big deal<o:p></o:p></p>
<p class="MsoNormal">> <o:p></o:p></p>
<p class="MsoNormal">> 2. we add features to oslo.db that are based on using triggers, special
<o:p></o:p></p>
<p class="MsoNormal">> constraints, or Gist indexes, so that the use of a database constraint
<o:p></o:p></p>
<p class="MsoNormal">> that needs this kind of thing is not a big deal<o:p></o:p></p>
<p class="MsoNormal">> <o:p></o:p></p>
<p class="MsoNormal">> 3. the first proof of concept for this, is a CIDR function / trigger for
<o:p></o:p></p>
<p class="MsoNormal">> this one reported issue in Neutron.<o:p></o:p></p>
<p class="MsoNormal">> <o:p></o:p></p>
<p class="MsoNormal">> Now the question is, "can I think of any operation in openstack, besides
<o:p></o:p></p>
<p class="MsoNormal">> this one, that would benefit from a custom stored function or a
<o:p></o:p></p>
<p class="MsoNormal">> specialized constraint". The answer for me is "not specifically but i
<o:p></o:p></p>
<p class="MsoNormal">> bet if I started looking, I would". Anytime there's an application
<o:p></o:p></p>
<p class="MsoNormal">> loading some rows of data out of a table, doing some calculations on it,
<o:p></o:p></p>
<p class="MsoNormal">> then dealing with a subset of those rows as a result, is a candidate for
<o:p></o:p></p>
<p class="MsoNormal">> #1 (in fact I have some vague recollection of seeing some patch in
<o:p></o:p></p>
<p class="MsoNormal">> Neutron that had this issue, it was the reason that compare-and-swap
<o:p></o:p></p>
<p class="MsoNormal">> could not be used). Anytime an application is trying to insert rows
<o:p></o:p></p>
<p class="MsoNormal">> into a table which should be rejected based on some criteria beyond
<o:p></o:p></p>
<p class="MsoNormal">> "unique key", that's a candidate for #2 - perhaps the plethora of
<o:p></o:p></p>
<p class="MsoNormal">> UUID-based recipes throughout openstack in some cases could be better
<o:p></o:p></p>
<p class="MsoNormal">> stated by more data-driven constraints.<o:p></o:p></p>
<p class="MsoNormal">> <o:p></o:p></p>
<p class="MsoNormal">> If we were to decide that the Neutron issue right here doesn't need any
<o:p></o:p></p>
<p class="MsoNormal">> changes, then I would be fine abandoning this initiative for now. But
<o:p></o:p></p>
<p class="MsoNormal">> as it stands, there seems to be a need to either do this change, *or*
<o:p></o:p></p>
<p class="MsoNormal">> add a new UUID column to the subnets table, and basically I'm hoping to
<o:p></o:p></p>
<p class="MsoNormal">> start steering the boat away from the island of <o:p></o:p></p>
<p class="MsoNormal">> add-a-new-column-everytime-theres-a-concurrency-problem.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Hello Mike,<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Thanks for your idea, If this feature can be moved to Oslo_db then
<o:p></o:p></p>
<p class="MsoNormal">that's better because Oslo_db has more people with experience of Database
<o:p></o:p></p>
<p class="MsoNormal">than Neutron has. Do you have any plan for this in Newton cycle?
<o:p></o:p></p>
<p class="MsoNormal">If yes would you mind sharing me about your plan?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">In case we couldn't move this one in Newton cycle then we should have
<o:p></o:p></p>
<p class="MsoNormal">an alternative solution which Kevin mentioned as "compare-and-swap".
<o:p></o:p></p>
<p class="MsoNormal">In next cycle, we will come back your idea. How do you think?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Kevin Benton wrote:<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">> we can solve this particular bug by doing a compare and swap<o:p></o:p></p>
<p class="MsoNormal">> operation on some network scoped value (at the cost of all subnet creates<o:p></o:p></p>
<p class="MsoNormal">> on the same network becoming serialized via conflicts and retries).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Hello Kevin,<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">In my understanding, after your patch set [1] is merged, we can add one
<o:p></o:p></p>
<p class="MsoNormal">line of code so that the systems will increase the revision_number value
<o:p></o:p></p>
<p class="MsoNormal">of a network which contain a subnet before adding/deleting the subnet,<o:p></o:p></p>
<p class="MsoNormal">Am I right?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">[1] <a href="https://review.openstack.org/#/c/303966/">https://review.openstack.org/#/c/303966/</a>
<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Thanks and best regards,<o:p></o:p></p>
<p class="MsoNormal">NamNH<o:p></o:p></p>
</div>
</body>
</html>