On 18.02.2026 15:22, Dumitru Ceara wrote:
> Внимание: ВНЕШНИЙ отправитель!
>
>
> Будьте осторожны с вложениями и ссылками.
>
>
> Hi Alexandra,
>
> On 2/17/26 6:58 PM, Alexandra Rukomoinikova wrote:
>> + "Logical_Switch_Port_Health_Check": {
>> + "columns": {
>> + "protocol": {
>> + "type": {"key": {"type": "string",
>> + "enum": ["set", ["tcp", "udp", "icmp"]]},
>> + "min": 0, "max": 1}},
>> + "src_ip": {"type": "string"},
>> + "port": {"type": {"key": {"type": "integer",
>> + "minInteger": 0,
>> + "maxInteger": 65535}}},
>> + "address": {"type": {"key": "string",
>> + "min": 0,
>> + "max": 1}},
>> + "options": {
>> + "type": {"key": "string",
>> + "value": "string",
>> + "min": 0, "max": "unlimited"}}},
>> + "isRoot": false},
> Is it a use case that "protocol" is not explicitly set? Should we make
> it mandatory, i.e., type "string"?
>
> With the current code lsp-hc-list crashes if there's a health check
> assigned with a NULL 'protocol' (no value). That's because of:
>
> int protocol_cmp = strcmp(lsp_hc_1->protocol, lsp_hc_2->protocol);
>
> The same question stands for 'address', will there ever be a case when
> address is not specified?
>
> I'm still reviewing the rest of the series but if nothing else pops up
> and if you agree I can change the schema here to make protocol and
> address mandatory without the need of a new revision from your side.
>
> Looking forward to your feedback.
>
> Thanks,
> Dumitru
>
Hi! Yes, it's my mistake( I can resend version if that's more convenient
for you.
--
regards,
Alexandra.
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
