Re: [controller-dev] [yangtools-dev] Check patch

Home Messages Hashtags Subgroups

#2091

Re: [controller-dev] [yangtools-dev] Check patch

Robert Varga

#2091

Hello Chris,

Unfortunately that is not the case -- the datastore performs structural validation, but it does not check leaf values and leafref validity. That capability is coming in Lithium (BUG-2362). Full-blown validation of conditionals (must/when statements) requires an XPath engine, which probably makes it a Be item.

For Helium, we decided to close at least the path from Java binding, as that was just an omission requiring a few hours of work -- hence BUG-1953.

Bye,
Robert

On 11/26/2014 08:25 PM, Christopher O'SHEA wrote:

Show quoted text

Hey Robert,

I'll update the test

One thing I’m not sure of is why does the data get accepted into the config datastore (we get a 200) if it doesn’t follow the model?

My (limited) understanding is that there is validation against the model when install into the datastore.

Chris

From: "ecelgp@..." <ecelgp@...>
Date: Wednesday, November 26, 2014 at 11:06 AM
To: Robert Varga <nite@...>
Cc: ODL Controller Development <controller-dev@...>, "integration-dev@..." <integration-dev@...>, "openflowplugin-dev@..." <openflowplugin-dev@...>, yangtools-dev <yangtools-dev@...>
Subject: Re: [controller-dev] [yangtools-dev] Check patch

Hi Robert,

Thanks for the quick answer, we got these test vectors long time ago from openflowplugin devs. So it seems the yangtools patch uncovered some flow issue or misunderstanding. We will fix that.

BR/Luis

On Nov 25, 2014, at 11:58 PM, Robert Varga <nite@...> wrote:

Hello Luis,

Your test vectors are not consistent with the model (opendaylight-match-types.yang), specifically you are using a plain IPv4 address where an ipv4-prefix is required.

Either the test vectors or the model needs to change.

Bye,
Robert

On 11/26/2014 03:34 AM, Luis Gomez wrote:

Actually, sorry to blame controller project, it seems the patches that broke some openflow flows are these ones in yangtools:

https://git.opendaylight.org/gerrit/#/c/11319/

https://git.opendaylight.org/gerrit/#/c/13070/

Current flows that show broken (in master and stable/helium): f2.xml, f13.xml and f14.xml

This is a bug in openflow or yangtools?

f2.xml:

<?xml version="1.0" encoding="UTF-8" standalone="no"?>

<flow xmlns="urn:opendaylight:flow:inventory">

<strict>false</strict>

<instructions>

<instruction>

<order>0</order>

<apply-actions>

<action>

<order>0</order>

<drop-action/>

</action>

</apply-actions>

</instruction>

</instructions>

<table_id>2</table_id>

<id>125</id>

<cookie_mask>255</cookie_mask>

<installHw>false</installHw>

<match>

<ethernet-match>

<ethernet-type>

<type>2048</type>

</ethernet-type>

</ethernet-match>

<ipv4-source>10.0.0.1</ipv4-source>

</match>

<cookie>2</cookie>

<flow-name>FooXf2</flow-name>

<priority>2</priority>

<barrier>false</barrier>

</flow>

f13.xml

<?xml version="1.0" encoding="UTF-8" standalone="no"?>

<flow xmlns="urn:opendaylight:flow:inventory">

<strict>false</strict>

<instructions>

<instruction>

<order>0</order>

<apply-actions>

<action>

<order>0</order>

<flood-action/>

</action>

</apply-actions>

</instruction>

</instructions>

<table_id>2</table_id>

<id>136</id>

<cookie_mask>255</cookie_mask>

<match>

<ethernet-match>

<ethernet-type>

<type>0x806</type>

</ethernet-type>

<ethernet-destination>

<address>ff:ff:ff:ff:ff:ff</address>

</ethernet-destination>

<ethernet-source>

<address>00:00:00:01:23:ae</address>

</ethernet-source>

</ethernet-match>

<arp-op>1</arp-op>

<arp-source-transport-address>192.168.4.1</arp-source-transport-address>

<arp-target-transport-address>10.21.22.23</arp-target-transport-address>

</match>

<cookie>13</cookie>

<flow-name>FooXf13</flow-name>

<priority>13</priority>

<barrier>false</barrier>

</flow>

f14.xml

<?xml version="1.0" encoding="UTF-8" standalone="no"?>

<flow xmlns="urn:opendaylight:flow:inventory">

<strict>false</strict>

<instructions>

<instruction>

<order>0</order>

<apply-actions>

<action>

<order>0</order>

<output-action>

<output-node-connector>CONTROLLER</output-node-connector>

<max-length>60</max-length>

</output-action>

</action>

</apply-actions>

</instruction>

</instructions>

<table_id>2</table_id>

<id>137</id>

<cookie_mask>255</cookie_mask>

<match>

<ethernet-match>

<ethernet-type>

<type>0x806</type>

</ethernet-type>

<ethernet-destination>

<address>ff:ff:ff:ff:FF:ff</address>

</ethernet-destination>

<ethernet-source>

<address>00:00:FC:01:23:ae</address>

</ethernet-source>

</ethernet-match>

<arp-op>1</arp-op>

<arp-source-transport-address>192.168.4.1</arp-source-transport-address>

<arp-target-transport-address>10.21.22.23</arp-target-transport-address>

<arp-source-hardware-address>

<address>12:34:56:78:98:AB</address>

</arp-source-hardware-address>

<arp-target-hardware-address>

<address>FE:DC:BA:98:76:54</address>

</arp-target-hardware-address>

</match>

<cookie>14</cookie>

<flow-name>FooXf14</flow-name>

<priority>14</priority>

<barrier>false</barrier>

</flow>

BR/Luis

On Nov 24, 2014, at 7:17 PM, Luis Gomez <ecelgp@...> wrote:

Can anyone in controller check this patch?

https://git.opendaylight.org/gerrit/#/c/12544/

Not only it got unstable merge but also broke the system test in helium/stable branch:

https://jenkins.opendaylight.org/integration/view/CSIT%20Jobs/job/integration-helium-csit-compatible-min/

BR/Luis

_______________________________________________ yangtools-dev mailing list yangtools-dev@...https://lists.opendaylight.org/mailman/listinfo/yangtools-dev

View All 2 Messages In Topic

#2091

Join {integration-dev@lists.opendaylight.org to automatically receive all group messages.

Home Hashtags Subgroups Terms