Intersight Managed Mode (IMM): Architecture & Day - 2 Operations

I let me introduce myself and my teammate Mark and Javier Galvis, customer success specialist.00:07For compute and AI in Europe with me is Mark who has the same question I have, and we are fully focused in in EMEA, ok?00:17This is the the flow of the session for today. You will discover, we will discover the day two operations with Cisco inter site. This session is fully focused in those sections we can execute, let's say with inter site.00:33Once has been deployed, we will we will have some slides about the the the the architecture the the principal architecture of of Cisco intercept but you will get access to those new things we have, ok?00:52And at the end of the of each section we will have a demo for you. Okay, so before we start we have we would like we would like to introduce our Cisco AQ which is our AI powered platform.01:09That brings together the the access to the to the Cisco support and and our professional services, ok? It helps the the organizations.01:28To understand the infrastructure and health anticipate the the the problems we we can those problems we can have in our day by day life and resolving issues through a unified architecture.01:43Okay. This is the the the process to onboard in Cisco IQ you you need to obviously login in or create that new account if you don't have it.01:59Migrate your data and and you can obviously download the, the appliance to deploy in in your in your infrastructure. This is the agenda for today. We have four main different sec sections, the hardware of.02:15Operation and diagnostics.02:34Firmware actions we can perform in in our in our servers, in our switches et cetera, the topology views how to get a clear vision about the connectivity in your data center. And finally the the metrics explorer. There are some minor sections in.02:36In all of those in all of those topics for today, but this is the main, the main goal for today. Here I will give you a couple of seconds to join us in in our slido. There's some because we have some questions for you to know more about your, your experience and your.02:55Your point of view about the UCS management infrastructure, ok?03:16So I will give you a couple of seconds you have the BD code and the passcode.03:23And for you guys is that if which UCS operations that task do you perform most frequently? We are asking this team to know more about.03:52Your your point of view about the the the UCS infrastructure, what kind of challenges you are facing of, of course, we have several operations and hardware replacement. That's, that's good. We have a couple of things for you guys today.04:04Hey we have fabric interconnected topology and monitoring, yeah, that's good. Good I think maybe main server operators good that this is the best, the best.04:19For you guys 67 %. Okay, I think we can continue. Yeah, this is the perfect session for you guys. So yeah, 1st of all for those newbies with inter site and this the next two slides.04:40Management with intercite and the and the more and the IMN architecture. So if you don't know if you are new in with Cisco intercite, our Cisco inter site is pro pro it is trying to provide a centralized.05:02The centralized management for for UC for our UCS servers, chassis, fis et cetera, et cetera. The policies and and profile standardized configuration and it's absolutely simplifying the deployment that's very important for the.05:20For the UCS administrators and networkers, networking administrators, we integrate the inventory, the health monitoring, the metrics, the the topology views, the support, troubleshooting. That's the most important you you will get access to the.05:39Cases you can create your own tech cases with us. You can it's a completely easy to handle the firmware management and the and the life cycle operations and is helping to maintain the the consistent infrastructure.05:59And reliable infrastructure.06:17Okay, so this is the the one slide that explain for all of you the inter site managed mode architecture IMM, ok? With this with this the the different versions we have, we have one SaaS version in cloud.06:18For an appliance you can do loading in in your in your on premise, ok?06:38And so the IMN is trying to combine the centralized management with that local control we need through the the fabric interconnects, so the device connector as you can see, is securing this connectivity link between the.06:44The UCS the UCS domain and the and the if you have it appliance, so the domain, the chassis, the, the, the server profiles applies apply the those policies to standardize the infrastructure.07:02So that's the main goal of inter site. All of this is the device connector is able to connect the the UCS and our inter site on our UCS servers because redfish.07:20Provides as a standardized, a scalable scalable and secure alternative to IPMMI and this this is a model replacement for the UCS XML API, ok? It's using RESTful.07:38Interface with with Jason format data as you can see in the slide and the OM extension enabled those capabilities such as the the big e.g. and and firmware management.07:54Plus custom actions like e.g.. So that's the most important to know in this connectivity between inter site and the rest of devices. So let's move to the.08:10And to the several operations we have for today the the the most obvious one of the 1st one is to know the differences between those several power operations, distinguish the what we are doing when we press the button of our.08:27Of our server virtually Bottom is extremely important for us. So the the 1st one obviously that the power on and power off is turning the the server on and off the power cycle turns turn the server off and then back on.08:47The hard reset is like obviously like the the pressing 3 s, the the bottoms of the server is shutting down immediately. And finally, the last one, the shutdown OS is requesting a graceful, a graceful shutdown through the operating system.09:05What is supported of of course but you can perform this this action with Linux, with the Linux, with windows, with VMware, all of them, ok? So this operational consideration is.09:23If we can say we prefer that graceful OS shutdown, whatever is possible and of course to protect the running workloads we have in, in that server. So the next operation would be the the recent semos. The th the.09:38This will restore the server BIOS settings to the factory default.09:58By cleaning the the current seamless configuration and it it's typically used as you can see in the slide is typically used to recover from a incorrect via setting.10:04Okay, that's configuration is configuration issues or or or maybe good related problems e.g. the reset KVM very important if we during this presentation you will see the differences between the KVM and KBM.10:19We will explain it later, but if you have different data centers e.g. that's extremely important for you. So the the recent KBM we start the virtual KBM service running in the Cisco IMC.10:38So to restore access to the server remote caution, ok? So is use it, is used use this action when the virtual KBM console fail to launch or if there is an existing session, e.g., becomes an responsible.10:54Okay so this operation is supported from the framework for the two as you can see one A and and later, ok, for early versions use the the reboot Cisco IMC, ok?11:14Let's move to the rediscover and what rediscover stands for for us for as a UCS administrator. The the rediscover is refreshing the the managed server inventory in, in Cisco inter site to reflect the latest hardware.11:34A configuration firmware versions and operational states, ok?11:53Status so use it after making after making changes to the server to on when inventory information is updated or out of things Sometimes happened. So rediscovering the server is refreshing all of these.11:58Information from before, ok? The next one would be the decommission the decommission it's important to remove a device from Cisco intercept management and here, it's inventory configuration and historical.12:18Let's say associate let's say association within the within the platform, ok? Meanwhile, we are leaving the the physical hardware and change it. That's important. This this operation is typically used when a device is, is retured.12:38Or replace it or is NO longer required management through inter site when we are trying to remove from our UCS infrastructure, but needs to be in the in the proper way, ok? Let's move to the security race and the the security.12:55Is the is a server action.13:15That truly clears the server settings and and the local drives and can can be triggered alongside the the the the commission. So it support M5M5 and and newer and then and the newer servers.13:18In in ISM in the in the standard in the standalone mode.13:38And IMM, of course. So it needs the the the NAST SP 800. That's that's important to note in the requirements so the process can take approximate approach 24 h. That's important to know.13:43If you need to secure the race and particularly when eration large LFF drives when we have a full configuration of of the drive, ok? That's important to note if you need to secure the the erase.14:04Yeah in the scrub policy we have the the whole easily we can repropose the the server is used to is is used the host server utilities there, it's.14:23Clean the the server during a profile and deploying, ok? Preparing for reuse the OS reinstallation. The this performs the.14:42Of the local drive and and including the NVMe of course and delayed the virtual drives. That's important to note if you need to repropose because the the virtual drives configuration will be the will be deleted. Anyway, you will have a warning.14:59A warning option about about this and it supports M five and and newer versions, ok? In, in IMM only, that's that's important to note. So the the requires.15:18A con compatible firmware and inter site appliance 11.one.two you will you will find out these binaries in in Cisco.com to download the the the Cisco appliance virtual machine.15:33Let's move to the server diagnostics with our server diagnostics in in inter site we detect the the hardware defects on on on arrival, ok? Identify the fault under the under the load and help the troubleshooting.15:53Server hardware issues.16:13So, things to consider in this server diagnostic we have the the flexible testing too quick and just 1 h or compressive 24 h, ok? With individual component selection.16:15A single SDU image, the compatibility with the M six and newer versions monitoring and control very important it it's detailing the process through the KBM console with the, with the option to cancel diagnostics.16:34And the results in in inter site, you will get access to detail error messages and suggested resolution, ok? You will see here in this last yeah.16:53Screenshot the the diagnostic test result. Let's move to the server remote connectivity, how to launch and disable the virtual KBM and the and the internal at virtual KBM.17:10And you will see the differences between those two options to administer or to get control of your of your server. With the normal, let's say normal virtual KBM, we provide a secure remote access to the fabric interconnect attached to the standard eh.17:26UCS server, ok? So it's requiring a local network connectivity to the endpoint.17:46So it needs to be recognized by the the server needs to be recognized by the inter site inter site KBM is using the single sign on.17:52Our part to grant the access without additional Cisco IMC or UCS managers UCS manager credentials, so that's fantastic and encrypted. The console is using the server configured management IP, so we need to provide that a management IP.18:09Into the server.18:27And and obviously the remote port, the prompting the users to select the AP when multiple management address are are available, ok? And then let's move on to the tunnel and virtual KBM, this is fantastic if you have.18:28And service on the edge or if you have more than one data center because the the KBM is opening a secure remote keyboard video and, and mouse session through in Cisco intersight. This traffic is fully encrypted even if you are not connected by.18:48By, by VPN, so he's providing.19:08Auto band access, to the server console this allows allow us to to the administrators to interact with the with the server remotely from for troubleshooting and and maintenance, even when the as I said, when the operating system is is unavailable.19:11So that's fantastic because we will have always access to our servers to disable this KBM disable this remote access when it's NO longer required so you will, we can enable and disable.19:31And this section is up to you so you can enable and disable whatever you want. So that's fantastic. Once again, if you have distributed data center, ok, data centers or sites.19:48In the system event logs we have to the the obviously we we we will we will give you the ability to download the and clear the system events for, for every single device in the.20:05Intercite doesn't matter this switch switches, servers, 3rd party tools, fabric interconnects, all of them and those logs will be available for you.20:20Now let's move to the tech support bundle in our eh in our operations which is the the is collecting the diagnostics data to, to help the Cisco tech.20:39A team to investigate the issues. As I said before with inter site you have access to the your own to create your own tech cases and and and generate and you will get access to generate a bundle.20:53From the equipment action menu and download it, ok? Under the the the system tech support bundle option in the device console is using the the equipment action menu. You can use the, the.21:11And that those option under inventory and download it from the diagnostic date so you can provide all of this information to our tech team, get data straight on connectivity or communication with our.21:28Team and start this process. Doesn't matter if you are using or our connected virtual appliance, our CBA inter site. Now let we we added this slide for today the hardware.21:44Ability list which is important for you guys in the in Cisco inter site because it's automatically checking the server hardware, the firmware, the the drive against the the public UCS HCL.22:03To get full compatibility of our service with with inter site. So it report let's say reports validation results for the server hardware, the server software and and obviously the adapters so it's helping the administrators.22:17To identify unsupported combinations and compatibility issues. So as you can see in this in these screenshots we have different alerts to upgrade the the the firmware or the drivers.22:34Of our components, which is extremely useful to get everything under control in one specific server e.g. in terms of the devices, the the the the beats, the drives, all of those things. If everything is in green everything is under control under the hardware compatibility list.22:53So this is the extremely important for e.g. for virtualization environments to get everything in the same level of drivers or verses whatever, ok? Here in this in this second slide and we are trying to explain the.23:12The process of the HCL validation is using the operating system and collected through the UCS tools by for in this case ESXi included in the Cisco custom ISO for our servers.23:29Or the, e.g., the OS discovery tool for for the ESXi and obviously Linux and windows. This is just an example with with VMware. So the the the tools send inventory data locally to the to the to the Cisco IMC over IPMI which.23:48Forwards to the inter site through the the the its device connector and that's why device connector is it's so important for for inter site.24:08And then inter site then automatically checks the, the hardware framework and driver's compatibility against the, the, the public UCS HCL list. Now we are providing after that we are providing the local co.24:18Configuration backup and and restore which creates a full backup of of our inter site account logical configuration, with this option we are trying to prevent preserve.24:38The identity of the identities and a full list of included profiles and and template. All of our infrastructure will be back up and in case of in case of disaster, ok? So we are restore we the restore can cover the.24:53Share configuration or be limited or.25:13To selected organizations, ok? Anyway remember with the with inter site PVA with the private virtual appliance, you will you will be able to deploy a cluster of up to four nodes for private for our inter site.25:16Infrastructure so you will.25:36Have full covered infrastructure configuration, the the PVA and obviously our logical configuration. Here we would like to show you that the whole compare.25:38The policies in in inter site lets the administrator to compare all settings across up to five policies. So it's helping to identify the differences.25:56In the in case to the busy.26:14To edit the settings and align configuration where it is needed. It's very important to get a nomogeneous infrastructure for for the UCS deployment, ok? Let's move to the chassis operations. As you know, Cisco is one of the leaders in the.26:33In the blade world, in the blades, in the blade server market. So in, in this case we would like to show you the the different operations we have with with the with the chassis operation with our X series, e.g.. The so the the rediscover is initiated.26:52A fresh discovery of the of the chassis and and its own components, updating the the Cisco inter site with, with the, with the latest hardware inventory and operations status. So it's pretty similar to the server but fully focused in the in the chassis.27:12It's a little bit more easy easy than the, than the server because we have less components but we will discover the PSUs and all of these things in the included in the in the in the chassis operations.27:29And here in discovering you can see the, the, the, the process and and what is execute when we rediscover a new chassis operation and the different options you have, the rediscovering, decommission, removed.27:45You can turn all the locator to locate the chassis in the data center. As I told you, you can open the tech case for this for this specific UCS chassis.28:05Let's move to the decommission which removed the chassis and it's I/O modules inventories from the, from the active Cisco inter site and and configuration while the the hardware remains physically present and and connected.28:20Okay, then, after the decommission we can remove the the the chassis that deletes the configuration of a physically remove it chassis from.28:40From Cisco inter site it's important to know that before disconnecting the chassis and configure the server port which is connected we can add the chassis back later if you, if you want of course and reconnect it and run a rediscovery, so we.28:57You can and remove the the chassis server if I can say for for Cisco interest. And then one of the most important operations we have in our day by day life with fabric in interconnect.29:19If I move, on, this is the the different views we have with with with our fabric interconnect which provides a centralized view of our FI health, identity and and connectivity.29:36It display if you can see the management IP address, the models, the firmware version and and and the port availability, that's important to get our fis under control.29:52So the device, the the device details include peer information and the associate and the associated with the with the switch profile, ok? So these views this view help to the administrator to identify the affected.30:08Devices and prepare maintenance activities. That's important for every single network administrator to get a clear vision of the in this case, as I said in the in the port connectivity.30:27For our servers. This is the detail the detailed view where you can see the the the the the ports available or occupied by by our configuration is displaying the the device's specific information.30:42In including the health, the, the, the model, of course, the serial number, all of this.31:02And let's say bureaucratic information we have with our, with our server, the management AP, the the firmware versions, e.g., and you can execute actions and base it in that in that information. This is the inventory view.31:08This is the typical view you can see when you are configuring from scratch your, your, your FI, ok? It's providing the details about the the Ethernet and fiber channel ports, the port channels and fans and and power supplies.31:25This graphical representation and health indicator is helping us to locate the affected components, so you, you, you will have a clear vision about what is going on in, in our.31:41Fis, so this information supports the hardware troubleshooting and and maintenance planning, e.g.. Okay, here you can see the connect the connections view and if you already have.31:58Intercite deployed you can confirm this in this information with us that this connections view is identifying the associate servers chassis, fabric senders, e.g..32:12And decommissioned devices, ok? And finally the UCS domain the UCS domain profile view is presenting the policies governing ports.32:28And port channels VLAN and the the of course so this deploying status help the administrator, the network administrators to validate the configuration and the configuration changes. So everything is.32:44Everything is under control from here and you will see the the different changes. And the the the last one slide is is dedicated to the different actions we have if we are using inter site, the with UCS manager.33:02And obviously we are providing support to migrate all of our customers from UCS manager to, to IMM or at the bottom you can see the the screenshot with the, with the different options, the differences between UCSM and and IMM. As you can see, there are a slightly different.33:20And options and and and how those options are.33:40Completely different compared with with other with other versions, ok? Here you can see in this slide the the how to get access to this to these actions, as you can see, we have those in in our demo we have.33:48Those two modes, the UCS inter site with UCS manager or inter site with IMM both with, with fis, ok? You will see here the different options we have. And now we have a demo provided by by by Mark.34:06About this these actions. Mark. Yeah, thank you. So now the idea is to show you a bit of what Javier explained, but to show it to you with a real use case example with a real inter site in hand. So let's start with.34:27The basics so all most of the actions that were that were shown by by here.34:45Are on the operate menu, so you go to to operate and then you can select either the server, the chassis or interconnect depending on what you want to do. So let's go to the servers and the 1st sorry and the 1st thing to notice is that specific dashboard, so I think you're familiar with it.34:50The the the the 1st important thing with it is that maybe you are not aware of of that, but actually there's two ways to go to the action panel. So we show we've shown to you a lot of actions, but there is two way to access those action panels either from the specific dashboard. So if you go to the to the far left.35:09Far right, sorry, of that specific server e.g., you will be able to see like a lightning icon to basically type the actions. So if you just click on the three dots on the far right, you will get access to all the actions that were mentioned by of course it depends also on the type of server and type of.35:29Deployment of server but from there we will be able to see everything related to the power, to the system so the the the the the the the the enablement of the example that's something that that was mentioned before.35:49That's something that we can enable from there. Of course, it also, if you enable it, you can also access you can also limit the access of it with the with airbacks ok so with the user can create some more rules and you can limit what what user can access the.36:06Also for that specific menu we can also access the diagnostic and such, so I just scroll down a bit. I can see all the other things that were mentioned, so e.g. the diagnostics, so let's e.g. click on that specific action.36:26So now I will run the diagnostic for that specific server, the one that I've selected, and if I click on next, I will be able to have to select what kind of diagnostic I want. Do I want a quick one or the comprehensive one? The quick, the quick one is around 4 h, something like that, and the comprehensive one is maybe 24 h.36:42Of course it depends on what is on the on the server exactly if there's a lot of of of storage less storage, a lot of of memory, less memory, of course. Then in that specific diagnostic you can select whatever you want. Do you want to see everything? Do you want to diagnostic everything?37:02You can see that we can even diagnostic the GPUs. So then you just click on next and it will launch a specific action on the panel on the specific request panel. So that's what I have done yesterday just to show you an example because at taking a bit of time, e.g. mine was almost.37:21If I just click on the server diagnostic just right there.37:42I can I can follow up what was what was happening. E.g., I'm in the inter site so we were downloading the SDU, that's what was explained before by the diagnostic ISO file that was that was automatically downloaded by by inter site sent to my server and booted.37:45On it, then all the tests were done and if I want to access it afterwards it's pretty simple. So just going back to my to my server, just go back to the action menu. If I scroll down again you can see that there is a view last diagnostic reports action, so let's click on it.38:05So I can see the last diagnosis that was called diagnostic one and you can see the result of it, everything was, was completed with green checks. So yeah, pretty good. Let's click on that specific server so.38:242nd way to access all of the actions is of course on the specific server, on the top right, we can have access to the actions. As you can see exactly the same menu of action. And something important is the system. So the system.38:39Can are really easy to access. So you can just go to actions, you go scroll down and you can click on.38:59Download system, event logs. Just do it, just do that wait a few seconds and then you will you will be able to to to load it directly on your browser like any other file and you can then access it just from there, right?39:06So that's the way to access the system then I want to mention the the HCL, that's something that was meant that was mentioned also by javier. Actually I have not NO complete HCL to show you but not unfortunately, so I'm not a good example of that. I hope that.39:22Your inter site is a bit more better under mine. But if you go e.g. on that specific server and you go to HCL.39:41You can access the HTML menu. You can see the 1st test that will be of a test of compliance, hardware compliance. So we are checking if the firmware is the right one for your specific model of server in the specific CPU.39:50Then checking the if the OS installed on it and the version of OS is compliant with the s the firmware and the specific model and then checking if the adapter adapters are compliant with the other two sections, inside that adapter compliance check there is compliance of the driver so the.40:06The driver of the.40:26For in my example the driver of the storage controller and the driver of the network yards the the so we are checking that complaints on the driver's side and on the firmware side. So actually I've done some updates yesterday to be honest with you and yeah, I broke my HCL for some reason.40:27I need to double check that, but that's that's nice because I actually can show you a specific thing that is called get recommended drivers. So in there you can actually see what are the drivers that are expected of you to be compliant on the HCF site.40:47So now I want to show you the policy because that's something that is really important. So what we will do is actually create a new one from scratch, right? So let's go to configure and go to policies.41:06And we will create a new policy. So I will check back down scroll, so clicking on starts, I will call it new scope because that's what I want a new scope policy. And then I can select from there if I want to have that to have to to.41:21Take any action on the disk or the BIOS or both of them. So I wanted to erase the disk and to erase the BIOS configuration. So I will create it.41:42And now what I will do is go to a specific profile and I will apply it to any of my server profiles, right? So going to my server profile, I will go to that specific server profile that is that is nice. There is NO.41:51A scrub policy as we can see there, so going to actions, going to edit, and I will just add my scrub policy. So just just there scrub, select policy and I will select my new scrub that was created a few seconds ago.42:06So clicking on clicking on on select and now I can already close the I can already close the edits settings. It was saved. And as you can see, the specific server is still in status ok so there is NO.42:23Need to to redeploy anything because actually the policy is only in effect when you, when you remove when you unassign a server when you unassign server from that specific policy. So that's something that will only be used at a specific moment, so there is NO need to redeploy.42:39Anything, there is NO need for maintenance window to add the specifics for policy or anything. That's something that is really easy to use and with NO impact.42:59On the, on the, on the specific server at the moment of course the impact is massive if you unassign the server. So let me show you just an example of that. So if I just click on unassigned server, I have a lot of warnings saying that specific server will be scrubbed so be careful of what you on what you do.43:07But I'm I know what I want. I will just unassign, and now actually there is two actions that will be created one that will be the undeploy of the sub profile and 2nd one that will be a scrub so we can just see just a few hours ago I have done the same thing, so we have the scrub action that is just right there, so if you just click on it.43:26We can see the same as before a specific download was, was made to to to to get this to get just the image that will actually erase the configuration.43:46And then we can we could we could follow up what was happening and what was scrubbed and such. So the good thing also they wanted we are in the policy menu, something they want to show you.44:00Is actually the the compare of policy. So that's something that is really interesting, so let's go to policies and let's change the type. I want to show you something interesting. So let's go e.g. select multiple policies, right? Let's select those three policies.44:17These three specific BIOS policy.44:36Let's select those ones. And if I just call up again, I can see now that there is a new a new menu with more actions available and I can see I can compare those. So the idea here is that instead of opening e.g. multiple tabs, which can be a bit.44:41A bit difficult. The idea is just to.44:59To have an easy way to compare to compare some policies. So those are biased policy so we are really complex to read, but that could be interesting to have that specific policy for those e.g.. So in just a glimpse of an eye you can see exactly what is different between between multiple policies. And actually you can also change the display because.45:02The display now is only what is only showing me what is different, but you can change that to to show me everything. So.45:22With that I can also see what is the the usage of the specific policy, why should I use it or not? So that's, that's really interesting and really useful to have a quick view of everything that is inside a specific policy and compare it with others.45:31Something that I also want to show you is the the back to the servers. So tech support are really important especially in day two operations. So there's multiple ways to to to to to to to to generate tech support.45:49Easiest one is of course to go to the action of a specific server and click on collect support bundle. That's the easiest way. But of course there's also other ways so let's go to the device console, so maybe some of you are not familiar with with the device console, so basically device console is just a console that's.46:09A user interface that is hosted on deploying interconnect.46:28Themselves. So you just go to the IP of the interconnect and then you can just go there and if you go to inventory, you will be able to see the server that are connected. You even have a few actions available and one of them is to generate the so you can generate also directly from directly from there and then.46:32And access it through the diagnostic data tab.46:52So I created one earlier, so I can just I can just download it. But what I've shown to you is how to generate the bundle for the servers, but if you want to do it for the interconnect, it's exactly the same way you click on generate bundle and then you select if you want the chassis, if you want interconnect.46:55Let's say I want to connect I want only the local switch and I can just generate it. Of course, an inter site is also exactly the same way if I go to connect, I have the exact same view as before. I just click on the three dots, I can just scroll down and click on collect exports.47:14So yeah, basically we have the same kind of of interface, the same kind of view, whatever it's a chassis server of course we have more or less actions available. But yeah, the idea is to make it as simple as possible and of course you can even do multiple actions.47:33At once when when it's possible so yeah that's that was all for that specific demo, so I think we can move on right here. Yeah, I was mute. Okay, fantastic thanks mike. Guys, this is 2nd question we have for you today.47:48And to know more about your.48:08About your day by day life. So we are asking you about what is your biggest operational challenge when you are managing Cisco UCS. We are trying to improve, we are providing we are improving every single month our inter site, our.48:12Our UCS servers and we are listening our customers and as a CSS here in Europe we can provide feedback from our customers to the product management team of inter site and UCS.48:30UCS portfolio so plenty of requests of our customers is known by the is known by the by our product management team. So it's a fantastic thing. As you can see the 1st one is the firmware upgrade.48:49That's that's good because we have a great news for you, how to upgrade the firmware for your fabric interconnects and with for your servers in in Cisco inter site. This is one of the most.49:08With intersight, this is one of the most easy things to perform with, with inter site, ok? In just two clicks, you will be able to upgrade your your the firmware in this case of of fabric interconnect or with your servers, doesn't matter.49:25From from inter site so upgrading those framework is initiated from the fabric interconnect and upgrades the entire domain by default with an option to upgrade a single FI if you have already.49:43The one of course, the traffic back evacuation and and I/O and path connectivity is tracking the enabled by default. So this traffic evacuation is automatically automatically done by the, by the inter site.50:02So the user acknowledgement is permitted once the data path alarms is clear on the 60 timer eh 60 60 min timer expires even if the alarm.50:20Remains the connectivity tracking can be disabled under, under the advanced options, ok? But as you can see in this screenshot is just to select the the the easy the how easy it is to right click the the.50:37Right click the button in in our FI and upgrade the the firmware. Here we are providing the the sequence operations of this of this of this action of course is complicated in the in the backend, of course.50:56The the infrastructure bundle is downloaded to the FIA if you have if you have two and synchronize it with the with the FIB. In the B then side the images are stated 1st followed by the traffic evacuation I/O path alarm, recording.51:14And, and the user.51:33Acknowledgement to activate the the images and and reboot the FAB. So here, as you can see, once the B site data path alarm is clear, a second acknowledgement allows the the upgrade to continue and on the F.51:36I FIA the A site is image images are then stated traffic is evacuated and the the the images are activated in the FI reboot.51:56Okay, and here in the 3rd step the the the waiting for the B site data is to clear, I said this thing before, upgrade the A side infrastructure, upgrading the FIA and without activation or without activation in the case of NXOS.52:14And evacuate the FI.52:34And then the the FIA ok and remove the FAA then sorry here this is the process, long story short, long story short, this is the process and then.52:37We would like to award you about the cache which is automatically stored in the infrastructure and server bundle use it for the upgrade and sync and synchronize it between the both fis, ok?52:58So the bundle are downloaded from from the inter site.com, that's obvious for the for the SAS or connected virtual appliance or the software repository if you have the PVA. Remember that with the PVA you are completely isolated, so you need to download these binaries.53:14And this upgrade these firmwares to your data center, ok? So the monitoring, the the, the monitoring utilization under the FI inventory as a full cache is preventing for the door nodes, ok?53:32To free space is using the least framework catcher to identify the bundles IDs and clear and clear framework catching to remove the selected bundles on on both fis, ok?53:50In this case, we are moving on to the firmware for, for chassis, ok? Here you have a pretty similar process. The the firmware cache upgrade is, is in.54:09X series. Let's, let's talk about the X series chassis is action that is updating the PSU and XF XFM framework with advanced mode allowing either either components and and.54:26To upgrade the separate separately, ok? So the PSU upgrad upgrades are non disruptive when power is redundant, of course, using parallel staging and and rolling activation if you are able to provide this.54:46More than one PSU and take approximately approxim 30 min if I'm not wrong, the XFM upgrades in involving the pcies switch on, on board controller firmware can.55:05Interrupt server connections to GPUs, e.g. or or or smart needs requiring affective servers to power off and restart afterwards, ok?55:21In the case of servers, how to do it with our servers. This is the process about how to upgrade the firmware of our server. The process is pretty similar, just right click in a, in our selected server.55:37And and and and oh sorry and clicking the in the in the the right click button, ok? Or for our servers. Here we are upgrading.55:54If the reboots that it's mandatory, of course you can follow up the process and on top of the by inter site you can you will be able to follow the process and with the KBM console, you can see what is going on in the server and how.56:11How Many times it's rebooted all of this section executed by the, by, by intercept. And in the policy, that's important because here's where you can automate all of those, all of those things is automating the the management through the server.56:30A policy and and policies and template is replacing the host firmware package policy in UCS manager if you have experience with UCS manager or UCS central. And it support upgrades and and don't downgrades and downloading the the staging.56:49During the policy, the the during the profile deployment and applying during the activation, applying the activation. So this policy supports up to six model specific firmware versions, ok?57:10Allows storage controllers and drivers to be executed in. In the advanced mode, e.g., and works with the with the with the standalone and IMM IMM servers, but you can add to automate this critical action of firmware.57:28Great from your inter site, so that's fantastic for when you need to maintenance windows, ok? And we have our one demo for you, right? Mike about this thing. Yes yes exactly.57:48So exactly as you explained, there's multiple ways to to do to the server. So as you mentioned, the easiest one is of course to select the actions of one server, select firmware and then you will be launched on the.58:06Somewhere.58:26Window. So you can see your server, you can see what is the current firmware, so the 02:26 something. And then when you go to the next step, it's showing you what is the suggested one, what's what are all the other that are available.58:27For you, so e.g. if you want to downgrade, it's not only the ones that are above your your your specific version.58:47And then you have the advanced mode, so e.g. if you want to exclude the drive or the from the upgrade process. So that's really really important and and the good thing is that you can select actually one server, but you can also select multiple server at the same time and do the specific action.58:54So e.g., if I select multiple servers, the only requirements is to that that is that the server needs to be the same model. So in my example, it's all X series M six, so I can select those, click on more action on the top right and click on create firmware.59:13So now we are all on that specific version, the 96 at the end if I click on next, now it's showing me the newest suggested release I could just select it, click on next, and then upgrade those servers easily.59:34Right? Of course it will require reboots, but that's another story. The other way to do it is as was mentioned by is to use policy. So let's create a firmware policy. So basically the policy, you can put it into.59:52You can put it into your server profiles and the most important thing that you when you create it is to select the right type of targets so you have to select if it's for servers to alone or for.01:00:14If I attached a server, so let's let's click on if I attach server clicking on next. And as mentioned before, you can select, you can select whatever version you want for multiple for multiple kind of models, so.01:00:32Let's say in my template of profiles of the template of of profiles, I have C 20C2 25 on that specific version I want to recommend it one. Let's say I also have in my data center, I have some x series and eight.01:00:45Same story I want the recommended version and let's say I also have some C 2020C 02:20, sorry, and seven and yeah I want the latest recommended one. So I can just create that, apply that specific policy to my template of servers or to all of my.01:01:06In one with only one policy, you can actually easily deploy that to all of your server and and do a bunch of batch, sorry, you can do batch of upgrade at the same time. So let's say that operation and go just quickly on the.01:01:29So same, same story, so if you want to interconnect, there is NO but you still have access to the action button, so you can just select the three dots, the action menu.01:01:50And select firmware. So as you can see, I only have selected once one for interconnect, but if I click on start, now I can see both interconnect in the upgrade menu. That's because actually you cannot upgrade one interconnect without the other because it's always a pair of interconnect.01:02:02So always both of them that needs to be needs to be applied at the same time. So if you just select it, clicking on next, you have the same view as before with the suggested release and all those are release available for you. The suggested release just as information is basically the same as the gold star.01:02:22Release on the software.cisco.com website.01:02:42It's not something that is related to your current configuration, right? So it's only suggested release based on the gold all the gold star release on the software.com website. So if I e.g. if I select this one and click on next.01:02:47It will start downloading the the the image, putting it into both of interconnect, starting with rebooting and apply it to the to one of the interconnect, then apply it to 2nd one for as little impact as, as possible.01:03:05So I think we can continue on your side. Okay, yeah, yeah. Let's continue. The last, the last question for you guys is a very personal question. If you could improve one aspect of your UCS operation, what would be.01:03:23What for you is the most important use case, let's say in our in your personal or or professional scenario because for some people the automation is extremely important for.01:03:42Many of them, the faster troubleshooting, as I said before, one of the main things of inter site is to provide you automation and and and intuitive eh troubleshooting. Okay, so you don't need.01:04:00With more external help to open your tech cases or get a full troubleshooting reducing the the downtime, e.g., for those virtual administrator for.01:04:20All of those administrator is important for all of them, this this kind of things. So the the the winner is the more automation. We have all the sessions about how to automate these sections you will learn you are learning today with intersite.01:04:39With ansible, with terraform, all of those things. We have other sessions about this, so check it out after this session.01:04:59Because we have plenty of documentation and and our GitHub to automate these these actions, ok? So, as I said in the previous slides, the get a clear vision of your infrastructure and what is going on.01:05:07What is going on in your topology in your infrastructure, it's it's very important for the for the faster troubleshooting e.g..01:05:27As you saw in the previous slide. So, here we would like to about the different topology views you have in in the fis with the chassis, in the rack servers if you have a standalone servers, all of those things. So.01:05:34Inter site is providing this visual as you can see in this screenshot, the visual representation of the of of those components are the connections in in inside of the inter site and the UCS domains especially when you have a.01:05:52Super huge infrastructure eh that's fantastic for you. So these interconnected views of, of fis chassis and and servers is helping the administrators to navigate through the infrastructure relationships, the, the health overlays.01:06:09And the port connections level is make it is making easy is making easier to identify the the issues and the access in the, in the con in the environment conditions. So it's very important for those for this troubleshooting.01:06:27Here you can see, e.g., in those fis when the topology view in those fis and the relationship with the with the servers, ok? It's providing the super high level overview of all connections.01:06:45Within a single IMM domain. So once again, it's helping the administrator to understand how the infrastructure components are interconnected especially when you are making troubleshooting or if your infrastructure is.01:07:04Pretty complicated, ok, is helping to those customers. Here you can see the information if you click in the, in the port, you can see the information of those ports e.g. and the status and all of this information needed by the network administrator.01:07:21Here in the network utilization view, e.g., you can see the, the, the operational status, the network utilization, the metrics, very important. If you know if you need this figure, these metrics.01:07:39Of your of your fabric interconnects, ok? In this case with the chassis topology view, it's very useful in in virtualization environments when we have plenty of servers with plenty of.01:07:56Of connectivities with the fis of course and is providing this detailed representation of of this connectivity in this case with just one chassis that you can extend it for more than one of course, and includes the visibility into the internal ports.01:08:12Is helping the administrator to understand the component connectivity once again for those people who is using VMware or investigate if if something is going wrong or it's going right in in your in your in your.01:08:30Chassis, ok? Here you can see the the information whole distinguished from the from the other connectivity. In this case, it's pretty it's pretty easy in this case we we are using a rock server, a pizza box server well.01:08:48You have we have this representation from a single C series server.01:09:07Which includes the connectivity with the with the fis, ok? And so it's pretty clear this configuration and the information provided by each of those, of those ports here we have a card with four ports.01:09:13Plus those ports in the in the in the motherboard so we can see the information and where is where it's connected. And very important to understand this, the the green and red and red lines are pretty clear, ok, but in the case of of the of the.01:09:30Of the yellow that those are visual indicators to have the administrator to locate.01:09:50The potential issues and and prioritize the troubleshooting and always checking the selected views laying and the colors is is helping to the network administrator to identify.01:09:57The the status. Obviously in the yellow it means that with the the partial down means that there is a mix or inconsistency inconsistency links the which is going up and going down.01:10:14So you have this clear vision to make the troubleshooting, ok? Yeah, you can see the the the this step one, the the connect the connections in the, in the fis.01:10:29For the topology view. So, it's once again helping to the network administrator the yellow or red connect connections helped to identify or locate the the between the problem between the the fis and the and the device or.01:10:45Of course and the fis though those callers may also reflect the FI port FI port alarms, that's important even when the physical connectivity.01:11:05Remains intact, ok? In 2nd in 2nd steps and 2nd step I I highlighted this the the 1st bullet though because if the problem, is detected of any of those links to verify if the carry.01:11:18If the cabling is is correct you can double click in the in the company chassis to see the information. That's the most important that that's why I I highlighted.01:11:37And the the the graph, ok? In the storical event it's given the format the let's the administrators.01:11:49To revisit the the past time interval to inspect to inspect the topology and in the infrastructure state as, at the at the moment, ok? At at that moment. So it's helping to to investigate the intermittent link.01:12:04Flapping and and and spikes to to make this troubleshooting easier for us after the systems returned to normal, e.g., without recreating the the issue. So the overlays of active acknowledge and and and clear.01:12:22Terms provide provide us the context for understanding events and and state transitions ok this is a new feature of of of inter site is pretty new, so use it because it's really useful to get a clear reason of this historical.01:12:42Events, ok? And here in the metrics in the metrics explorer, we have the data collection process to gather the metrics through the device connector for for IMM and and.01:13:01Remember for the if you are using, we have the transition tool to help you to migrate from UCS manager to, to inter site to help you in this in in this in this process. So he's providing.01:13:18The visibility into the device's state, behavior and and and performance, ok? In the deployment in the in the inter site on cloud deployment that the the those metrics are retained for up to two years, ok? The process supports outage.01:13:33With reduced.01:13:53Collection granularity for you how to extract that that metrics is combining then an instrument name and and measured value and statistics, ok? So e.g., you can in this example, you can.01:13:56Get the vision of the statistic in the fan select the fun speed option and finally a minimum to get this this metrics, ok? And the different instrument names you have available in in inter site.01:14:16With fun of course with the single power, the temperature, very important if you are using virtual GPUs which generates a super high temperature in your data center, all of those things, ok? And eventually metrics.01:14:34It's, available directly and on each supported inventory object in the in Cisco so doesn't any kind of of object, ok? So they provide.01:14:51That visibility into the selected component condition, behavior, and, and performance. Okay, so the administrator can use this information to monitor individual individual resources and investigate issues.01:15:06Within the context, the, the, the context of the or the affected device or devices in total, ok? With the port metrics in in topology you will have access to the topology view of one specific port view on on on the fabric interconnect, so it's fantas.01:15:22SD.01:15:42To to to create your own capacity planning for your fis to know the the the amount of traffic we are handling. So you just need to click in the desired port in the in the port you prefer, you want to see and and say.01:15:46The the metrics of those ports. So that's fantastic in when your FI.01:16:03It's full of with all of your parts occupied, it's fantastic to create your own capacity planning for the future if those fis needs to be extended or if you need to get a new one or new ones, a new cluster.01:16:09Yeah, that's fantastic for the capacity plan, ok? And finally the explore the explore menu here the the the is let letting the the administrator to customize the that that metrics they want to analyze or and and how to displace them.01:16:25Using the different chart type, ok? Here you can see how to add those metrics and the and the filter to to group the the different possibilities. So it's allowing to.01:16:43To the to select one or multiple metrics for for that analysis, let's say, ok? And finally the exploration widget remember that that your dashboard in in inter site can be.01:16:59Fully customized for your, for your experience. You can customize your your dashboard to showing up just your preferred devices in in inter site. Finally, the exploration widget is allowing us and saving the the exploration.01:17:15To be added as a widget in the in the in the dashboard, ok?01:17:34So you can configure it by your own in your your main dashboard. And the last one is the operation, the open sorry the open telemetry API, which provides access to the inter site time series metrics is using.01:17:39The open standard, the open telemetry standard, so that's fantastic because it's supporting the, the integration with OTLP compatible solutions and the metrics can be created using a native native queries, ok? So we are using this.01:17:57Open telemetry is fully open.01:18:17So that's fantastic for the client. Mark, you have another demo for this? Yes yes exactly. So yeah, I want to show you a bit of demo of the topology and the metrics. So let's start with the topologies. So to access it is quite simple. You go to the e.g..01:18:19It can be also the same for the servers, you go to connect, you go to topology.01:18:39And then you will have access to that specific view. On that specific view, what you can do is actually to click on different things. So let's click e.g. on that 1st interconnect, and then from there you will get some metrics of, e.g., here the host energy, the power, the memory utilization, the temperature and stuff.01:18:45And such, so you can get those information from that specific topology. Also if you overwrite some of the ports, the, the ones that are enabled on the other ones actually you have the specific view and you can even disable some of the ports from there.01:19:04So it's a bit like the view that maybe some of you knows from UCS manager so you could also easily enable or disabled admin state of of ports. From there you can also click on view metrics to have access to the metric of that specific port.01:19:24And something that was mentioned by Ravier a bit earlier is the historical event timeline. So that's a feature that was added, if I'm not mistaken exactly one week ago, but it's really, really useful so let's e.g. enable it. So as I can see there is not much activity as.01:19:41You can see we have not much activity, but e.g., let's go back a bit more in time. Let's, let's go two days before. I can see that on the specific date I had a lot of alarms and a bit of requests, so maybe interesting to see exactly what was happening there yeah what was happening there?01:20:00So by clicking on that, I can see now that my topology has changed. Now we can see that I have a red red link between the the between my chassis and the B one. So if I just overlight the the times, the time frame where I selected 1st, I can see that I have.01:20:19Many alarms and a few requests so let's click on the alarms.01:20:39So that way I can see all the alarms that were available at this specific time. So I can see a lot of interface down. I can see networking missing and stuff like that. So that's really strange. Now let's see maybe also the request because there's a lot of requests. So let's click on the request and I can see actually.01:20:43It was because of upgrade of firmware, so I can see the firmware of my blade of my of my chassis of if I think if I just go next, I can see even more things related to to firmware. Yeah, so actually I can easily see what was.01:21:03Two days ago.01:21:24What was happening on that specific time frame? I can see why I had a red link and I can see why everything was not working or working strangely on that specific time frame. Yeah so also just if I click on the metrics.01:21:25Of course I will have access to all the metrics of interconnect. You have exactly the same view for the servers, so in there you can see things related to what we call environmental metrics. So we can have the fan speed, the energy for each PSU and stuff like that.01:21:44You can even make it bigger if you want to have a better view. So e.g., let's click on the arrows just right there. You have a bigger view. You can change of course the the time interval let's say e.g. for 24 h. If you have the you can also change for the granularity so.01:22:03Yeah, that's something that's possible in there you can also change the time frame. If you go to the to the network matrix, you have the view related to the metrics related to the network. And same as before we can open it a bit bigger and you can see exactly what was happening.01:22:23Multiple times before we have seen that specific button that is called view and explorer. And if I just click on it, it will open up the metric explorer that was mentioned before. You can also just access it easily on the right on the left side of the screen if you click on analyze and then on Explorer, you can have access to this. And the good thing is that.01:22:42When you do that, when you open an explore something that was already created by.01:23:02You can actually customize the view. So let's say now my view is showing me the the the network bytes of that specific internet ethernet port on that specific interconnect is showing me the average, the maximum and the minimum. Let's say I don't care about the.01:23:08Minimum and the maximum, so.01:23:28I can just delete that specific metrics so one line less and then also the maximum I want to delete it. So now I only have the average, ok? But maybe I want to see more than just the average of the specific ports. So from what I can see there is a bit of filtering just down right.01:23:30There so actually I will just remove everything that is not the port type because what I want to see is actually metrics of of all the of all the internet port in all of my inter sites.01:23:50Okay let's do that so now what I'm seeing is actually that it's the the statistics the the of of of it, but really not clear so maybe I want to add some more information into it. So I will just click on group by and actually from there I can add more things.01:24:02Into it, so e.g. I can add the host name. So now I can see actually that it's related to I can see that more I can see more lines related to the network interface network bytes of internet ports in multiple type of devices, but.01:24:22If I want even more details, I can also add the name into it. So now I can see exactly what, what port is doing what, right? And as you can see there is only five ports right now, so actually I can up it to ten.01:24:39Thought to ten to ten lines, right? So, now I have ten lines available in my view, but actually you can of course go bigger than ten, ok? So I could you could just select all but the thing is that that specific view on the right side of the screen is only able to show you ten things at the.01:24:56But the good news is that actually you can create your own API calls to get actually more than ten things at a time. So, the 1st thing that you can do is to export those data. So if you just scroll down a bit below the the view, you can export the whole data as.01:25:18SV file.01:25:38And also you have something that is really interesting that is called the query card. If I just go to the query code and if I scroll down just a bit, I can see a JSON request that is actually the Json request that was sent that was sent to the API by my exploration. So that's what I've done on the left.01:25:39On the left side of the screen, so what the the table that I have completed just on the left side of the screen was created as a as a request. I can just select it and go to the.01:25:59To any automation tool or to the embedded clients, so let's go to the client that is under developer center, so let's go in there and then go to API reference and just type time series and if you.01:26:11Yeah sorry yes, if you go to and click on clients, you can just copy and paste the the Json thing and you can have access to all those information.01:26:32Using as mentioned before, some standard writing so you can easily export it to any other telemetry tool like. So yeah, that's all from my demo we can go we can finish on the last few words.01:26:47Yeah. Okay, so this is the end of the session for today guys. Some people is asking about the records and the the slide deck of this presentation of course it will be available for you don't NO worries at all. If you need something just drop us an email. We.01:27:06Yeah, Mark and myself we are available for one to one consultation or if you have any kind of issue or doubt or whatever, ok? So you can.01:27:26And come to us and and ask whatever you want, ok? So NO problem at all. This is, this is the the for today. Remember that in October we have these those new sessions that we will provide our teammates, the latest few.01:27:35Features and enhancement. This is very important. This is a very recurrent sessions we have with the new features of.01:27:55Of inter site of inter site more are coming and the ecosystem integration very important if you have 3rd party tools or 3rd party solutions like NetApp, VMware, Itachi, et cetera, et cetera, all of those, all of those big guys, ok, all of our friends.01:28:02And so we have another session in 1 October and 8th of October, ok? So this is this is all for today guys. Thank you very much for joining us. Once again, if you need something with your inter site administration or your inter site.01:28:21Task, drop us a message. Let me go to the to the to the 1st slide if I can, sorry here, just a second.01:28:41Yeah, you can see.01:28:55Our emails, ok?01:29:16My name@Cisco.com so drop it send send us an an email if you need something, if you need to know more about our UCS management solutions so that's all for today, guys.01:29:20We'll give you a couple of seconds. Mark, thank you very much for your demos, very useful guys. I hope you learned a lot from those demos. Try these things in your inter site and thank you very much once again for, for joining us today, ok?01:29:37Thank you so much. Thank you very much. You, bye bye.01:29:59