Hi Michael, I do agree that with a rebase the log would be nicer (everyone dislikes those merge commits). However, I think that a merge strategy is better since (1) doesn't rewrite the commits

Hello Neutronians, would any of you object to me raising a helpdesk@... ticket asking for this project Gerrit Strategy to be changed from "Merge if Necessary" to "Rebase if Necessary" ? This avoids

I'd like a summary meeting on the 12th, but it could also be replace by a summary mail.

I'm assuming you mean JAX-RS here, not https://ws-rs.org ? :-) sure - but I'm confused now; jersey/CFX/resteasy has nothing directly to do with the goal of c/68415, which is to replace the web.xml

Michael Vorburger <vorburger@...> wrote: IRC only without audio/video. As most attendees come from openstack side, we're following openstack way and topics are mostly on networking-odl(and

is this and next meetings purely on IRC on the general #opendaylight channel, like https://wiki.opendaylight.org/view/NeutronNorthbound:Main says, without any audio Zoom/Bluejeans? - Just checking. I

Vishal, just to clarify/not let this loose end imply something unsaid: This "Neutron-with-MQ" was an "idea" more than a real current plan, on my side. If there is any real interest in pursuing

Hello Robert, thank you for your interest on https://git.opendaylight.org/gerrit/#/c/68415/, the proposed new API for programmatic registration of web Servlet, Filter, etc. in infrautils to be

Hello Michael, It boils down to what ws.rs implementation we choose and how we integrate with it. There are at least three choices: - jersey - CFX - resteasy When we are talking about OSGi/BP

reboot or upgrade of ODL cluster is valid use case. several minutes disconnection sounds reasonable. What are requirements/assumptions? analysis of this use case is necessary. For example I can think

Hi Isaku, Thanks for quick response.. It is possible that Daya will be attending ONS.. Inline [Vivek]

Hello. Who will attend OpenStack PTG? Due to OpenStack PTG and ONS, ODL neutron irc meeting schedule will be as follows: - Feb 26, 2018: skipped due to openstack PTG - Mar 5, 2018: maybe skipped?

Valentina, I totally missed when looking at your email yesterday that it's a DEBUG not an ERROR log! Thanks Robert. Why are you concerned about this DEBUG level log? Is there an actual real problem?

Hi Michael ! Thanks a lot for explanations. Here is a full stacktrace: https://paste.fedoraproject.org/paste/Xzonb2HZWcdp~ZMUeqxWhA To reproduce, just enable TRACE log in MDSAL: karaf@root()> log:set

Hi Michael, In one of my earlier projects [that was before ODL even started] we were using a similar design of SDN Controller using MQ. Plans for Flourine? Regards, Vishal. From:

This is fully expected and is inevitable due to how unions work. Do not enable whole-sale logging on org.opendaylight.mdsal, or specifically exclude binding-dom-codec. Bye, Robert

FYI: That's what we do. But personally I still think this entire architecture of that driver writing into a Journal DB and then sending those events to ODL via REST is somewhat sub-optimal. One fine

the key to solving this is figuring out why in mdsal it *deserializes* what is clearly an IPv6 as an IPv4 object... deserialize(EncapsulatedValueCodec.java:65) is constructor.invokeExact(input), but

Hi MD-SAL Team ! Hi Netvirt Team ! When I add IPv6 subnetwork or manipulate a port with IPv6 address I constantly see this exception in log (MD-SAL TRACE log was enabled): openstack subnet create

Cool - thank you for expresssing interest and support for pursuing this. I'm guessing I'll get an enthustiatic +1 from you on https://git.opendaylight.org/gerrit/#/c/68415/ then? ;-) That's just the