 Hi everyone, so good morning. So I'm Adriano. I'm Mr. Core and here's the project update So For anyone that doesn't know what Mr. Does. Mr. Is a workforce Workflow engine it's provides the workflow as a service. It allows people to write workflows without in Iaco or to To write stuff that you don't really have to write code and allows you to have a high-level language It was fun and have an up Havana release of open stack right now it has over 50 contributors Pretty much every red hat triple the every triple deployment uses it since Newton and It's a point on 5% of the clouds As as work flow engine for open stack, but it's also used on Other place like sir, which makes me really happy because I'm a physicist. So So some of the features that were landed on rocky One of the most important ones One of the most important one. It's the execution events and notification server that allows you to create webhooks to your workflows and Split like huge workflows that sometimes have a test that takes too long into more manageable manageable workbooks and workflows It's defined per flow execution. So each execution can have its own notifications and And can notify all events of that are listening for for that that call We also added services for Tacker vitrej manila killing and zoom some of them don't know what you do. Sorry And another thing that landed on rocky that was good for people that have huge workflows with many actions Is that we are able to cash workflow workflow actions now? so many of the large workflows that have They have a huge amount of reasonable Repeatable actions and those are caching that got them huge improvement for some of the large workflows in performance We have a new heartbeat mechanism that the text when the action fails it's In rockets too based on a scheduler, but that changes on time and That allows us to know when when the task is we don't have to wait that much for a test to fail we can get it and and Reported easily of Esther There were a huge number of tolerance and for infrastructure failures improvements to so a Lot of them land on rocky. There are some more bug fixes load of bug fixes right loaded first time And that's that's allowing from our robust service In the previous version you could delete executions when they were running and that would cause all call all kinds of problems for For the environment for make sure to recover So now on from rocking on we have a force parameter when you want to delete executions that are running You have to you have to pass the force parameter. Otherwise, they will not allow you to delete so can still do that if you know what you're doing, but removes the chance of random or Or just messing up her environment Another thing that landed and that was really important for It wasn't that important for first in triple But that was a massive improvement for some of the other use for me. So it's that the namespace support for workbooks It was usually just for workflows, but now a full workbook can have a namespace That allows you to have if you're open to users They don't have to have a unique name for their workflows because your friend and move to use environment That could be a problem like you're trying to create an workflow that someone other person created same name That was not allowed. Now you can and Unless you specify it for backwards compatibility, it defaults the full namespace. So if you're just not using it It's completely transparent to you Yeah, also from Rocky on the something may your action can do much more Stuff to relate it to how we may you should behave with CC BCC and engine HTML formatting That gives us like better Appearance when we're dealing with external users or our processes So in Stein I'm a little going a little bit fast. Sorry In Stein what has already merged is as I said, there are huge improvements in performance Anything that has too many tasks waiting for a join That was too much pulling on that It's that's that's been solved and that's huge improvement for some of our Users that have massive tasks with like thousands of actions There's integrity checker that was When it ran it was not the best performance and that was optimized and Also the heartbeat checker that I said before now it doesn't run on schedule. It does in a separate way It which was a department schedule that when we start and stop and start again it would heck out problems you might have Two schedulers running or two test cadres looking for that task and that when you have Loads of them you have overload of tests doing the same thing. That's not the case anymore in Stein huge amount of bug fixes so in progress We have workflow error analysis that's been in progress for a while It might not be read first time But that allows you to drill down in the bug What went wrong with your workflow with your actions and get more visual and specified way instead of like using mist documents to dig down and Get what you what you do because the usual flow is that you get execution ID You probe the results then you get the actions that get the execution's ideas of the actions that you can do down a lot but with the Workflow error analysis that's much more visual and easier There's more support for execution event and execution servers right now. We just support webhooks but we're We're working on more ways to do that Probably that's probably gonna land on Stein. We depend on on other factors, but that's in progress A good thing for if you're running clusters or if you're running more than one missile server engine nodes is that now we It's it's in review. It's the graceful shutdown off service that allows you to scale down If you have if you're trying to run a J better even better tolerance to infrastructure problems And oh It's an effort and there was like a tooling and to developing custom actions like Cookie cutter recipe that you can if you want to do a custom action You can like kickstart you and just makes you almost there And there's improvements on docks at emerging beyond Stein There is for the integration that is a schedule that's on a roadmap Kubernetes and Sable Tower There's more We still have some we got all the low hanging fruit from for performance and stability But there's still some high hanging fruit that we have to tackle And the one of the biggest text that that we have right now is that blocking mode We're using Oslo message blocking mode because the my sequel driver Doesn't quite work well with event let for us So we are forced to use that and We try some experimentals with we event let it's better than it was on rocky, but still gives us some some Some problems with the database, so that's still ongoing We've been talking to the Oslo people They're deprecating removing on Stein. They got to keep it a little bit further for us to use it But that's probably the main the main Issue that we have to tackle in the future So to find us it's easy. We have our own ERC channel on open stack MISROL free node We are always welcome to new contributors We do instead of having Meetings on open stack meetings out or something we have office hours That's three times a week someone is going to start a meeting on the open stack MISROL channel They have like times or times zones for everyone that Even wanted the middle of the night for us but for us here in the Europe But we have we try to reach all time zones And that's it any questions. Thank you