ask-annie/out/850846395/transcript.txt

# Transcript: 850846395
# URL: https://vimeo.com/850846395
# Duration: 5637s (93.9 min)

[0:02] Is don't hesitate to just ask at any point. I'll try to answer at the point at the time.
[0:10] But
[0:10] if it will disrupt the whole scenario,
[0:13] obviously, we'll try to figure it out in a different way. There is very minimum requirement here. If you know what advanced routing is, you shouldn't be get lost into the presentation. It will be technical. I'll be showing things on the server.
[0:26] But
[0:28] if
[0:29] someone
[0:30] really get lost, just let me know. We'll catch you up as fast as possible. So let's talk about secure transport.
[0:37] In the last years
[0:42] okay, my presentation is moving a little slow, so I'll just wait for it to catch up with me.
[0:48] Yep. Here it is.
[0:50] If you had been with Secure Transport for long enough,
[0:54] you had seen it changing and adding more and more features, it to routing files more and more effectively.
[1:01] We started with the script base before four six. The four six to five two gave gave us the application framework, which was the basic application to advance route and the non advanced routing with the kinks that could route files around without conditions.
[1:16] In the five three world, we started getting the advanced routing.
[1:19] And then in the last year and a half, r and d had been adding a lot of additional new features that allow you to move from conditional routing into full blown orchestration.
[1:31] And that's what we'll be talking mostly about today,
[1:35] the orchestration part of it, and how you can improve things.
[1:40] Those are not things that you
[1:43] need to use if you don't need them, but they are to the arsenal of what SD can do.
[1:50] When we talk about conditional routing and we had head sessions on that a few years ago,
[1:55] and you probably are using that, that can make the server very, very powerful as it is. But with the new features that were added in the last couple of years,
[2:07] things had progressed even more.
[2:11] Modern MFTS, we know has a lot of big challenges.
[2:15] Most of them ask what to do about the old workups, what you had built something years ago.
[2:23] It is working. You haven't touched it in years, but it's not effective.
[2:27] Part of the idea of this meeting is to show you what is now available in the product so you can rethink some of those old scenarios,
[2:35] think about how the old world had been living, what you had been doing,
[2:39] and try to use make better use of SD server.
[2:44] We're not lost about new protocols. That part, everyone on the call know how to deal. If a new protocol is supported, it's a new connector, nothing interesting there.
[2:54] Security enhancements,
[2:56] as you know, every single update we have has a couple of those.
[3:00] You need to stay on top of them. Some of the things we'll be talking about today might help you with that. Some of them might not.
[3:07] The unconnected system and the lack of global view, see that's where some of this will help as well.
[3:13] One of the biggest challenges when I work with a new partner or with new customer with with that don't have gateway
[3:21] is usually that they have multiple products doing different things and not talking to each other and trying to connect them becomes a nightmare.
[3:28] And then, of course, the big conversation between what is managed bridge and what is orchestration.
[3:33] Is it
[3:35] can you set up everything in the same place or not?
[3:40] So how did we got to where we are? Initially
[3:44] so there were three big steps
[3:47] that were done into a secure transport that allowed us to get to where we are and build what I want to show you today. The transfer site definitions,
[3:56] we're not talking about the fact that we're adding a lot of them. So, you know, you can get the connector with protocols that we never even talked about until a couple of years ago. But, also, inside the feature transfer side, you can use dynamic parameters to the to to actually
[4:12] put something in the site, what the download folder, support folders, whatnot.
[4:17] The applicable framework allowed the adding of all of those new ones.
[4:21] The number of connections for each site can be controlled individually
[4:25] inside of the site, inside of the subscriptions.
[4:28] We added the ability to do the history of the site, which means that if you have partner that doesn't want to add to down for you to delete files after you download them, we can keep track of that.
[4:40] We can provide a list of what is on the site so you don't mess up and download thousands of files you didn't expect.
[4:47] And then for some specific protocols, SSH,
[4:51] we actually can now reuse the connections,
[4:54] making it even more effective to pull a lot of files.
[4:58] We added more enhancement on the scheduling. We moved away from the pure admin UI single screen that allows you a lot of things, but not everything, and gave you the complete
[5:09] Chrome jobs so that you can schedule in each way as long as you know what you're doing.
[5:14] And then
[5:15] now
[5:16] and that's one of the things we'll talk about on this meeting.
[5:19] We can do
[5:22] invoke an on demand pull from inside of a routing the routing engine itself,
[5:27] allowing you to do interesting
[5:30] things like
[5:31] do a pull based on something that happens with other files.
[5:35] And for the routing, from conditions inside of the route, passing the calculated values,
[5:40] passing various values between different scenarios
[5:44] going into orchestration.
[5:47] And that's what the five scenarios we'll be talking about are,
[5:51] and I'll be moving mostly to the server after that.
[5:57] The the five scenarios,
[5:59] we'll be doing an on demand pull based on something that happens inside of a route.
[6:04] We will be doing a little bit of work with
[6:08] passing variables between different,
[6:11] routes,
[6:14] retrieving
[6:17] out incoming
[6:20] details into the outgoing for outgoing transaction,
[6:23] something that our asset,
[6:25] the protocol could always do, but now it can be done in other protocols,
[6:29] the ability to react
[6:31] on something going wrong during the transmission. And, of course, at the very end, we'll talk about monitoring and how all of that works together and how you can pull together a full blown orchestration.
[6:42] And with that, let's talk about the first of our scenarios, the conditional pool.
[6:49] As you know, in ST, until
[6:51] a few versions ago, a few updates ago, there was only one there were if you wanted to pull a file, you had to have a scheduler for it or you could do an on demand,
[7:01] but there was no way to implement scenarios
[7:03] such as
[7:05] if the partners put you a dot ready file or any other file
[7:10] that tells you that a set of files is ready, go and actually pull the correct files from there.
[7:17] Or something as simple as if something fails during processing, go and pull something else to go and work on it.
[7:26] Or,
[7:27] if you have a back end application that produces and writes a huge file,
[7:32] There was no way for that application to tell you easily that the file is now ready to be pulled besides renaming the file itself, which not all applications can do.
[7:42] In order to implement that, what r and d implemented
[7:45] and what r and d built into the product, sorry about it,
[7:49] was the ability to call the pull from inside
[7:54] of a route.
[7:56] So you can go into the route with one file and call pull into the other to to get other files. Now a couple of warnings,
[8:04] and I'll be slowly switching to my
[8:07] other screen so I can show you the server itself.
[8:11] But
[8:12] one of the things to remember is that this is just a scheduling of the pool.
[8:17] It's not going to pull the files into the same
[8:26] into the same sandbox.
[8:27] So you cannot just go and grab other files and put them and process them. If you want to do something like that, current file, the one you started with, will need to get it published somehow so both of them go together.
[8:39] The other important part, if you're pulling multiple files, they'll still be treated as single files. It is not the it doesn't give you the ability
[8:48] to pull all of the files into a single sandbox to process them together.
[8:53] So,
[8:54] but other from that, it's pretty straightforward.
[8:57] And
[8:58] to show you how that works,
[9:00] and you should be seeing my browser by now,
[9:05] It's really, really straightforward,
[9:07] so this will be the easiest of the use cases. What I have implemented is if the file that arrives into the routing has the suffix of dot ready,
[9:17] go to this specific site and pull pull the files you find there.
[9:22] And here is where you need to be careful because there are two different things that you can actually specify. The first one on the local is
[9:30] where you want to put the files when they launch into ST. It can be in the same folder where you're into now. It can be a separate folder under the account.
[9:41] It just needs to be under the account that is running the subscription.
[9:44] And then you can override the download folder and download pattern on the site you are pulling from.
[9:52] Before everyone asks, please note that this is only available for the generic for the old protocols, SFTP, FTP, HTTP, generic HTTPS. Those are the nonpluggable
[10:03] protocols.
[10:05] This is not available if you have a pluggable site yet.
[10:08] If someone needs it for for a pluggable site,
[10:12] please open an idea so r and d can prioritize.
[10:16] The but this the reason why it is for these four sites, but not the others is pure technical.
[10:22] These four are still built in a different way behind the scenes,
[10:26] which gives R and D a little bit more power on what they can and cannot do. But they are looking into the pluggables as well.
[10:34] The
[10:35] override of the download folder essentially
[10:38] turns
[10:39] means that
[10:41] if
[10:42] this can this this on demand pool can be even used for things like pulling from multiple folders from the same place without needing separate sites.
[10:51] Although, we'll talk about it a little later with the difference tonight.
[10:56] So
[10:58] as I said, the pretty easy configuration,
[11:00] and now all I need to do is to go and upload the file
[11:05] so we can see it working.
[11:07] And then I'll open for questions before moving to the next one.
[11:11] Okay.
[11:13] Just on time. So this morning, when I was testing, I had the JPG file. It's not a dot ready file, so nothing happens. So let's go upload,
[11:21] ready file.
[11:24] Rename. Where is my rename?
[11:34] Oops. That was the wrong name. Wasn't it?
[11:39] Alright.
[11:45] And then let's grab the ready file.
[11:49] And it will be pulling from the test folder over here
[11:53] where I already have a file.
[11:56] Oops. Sorry.
[12:01] Maybe I should have picked a smaller file.
[12:07] Okay.
[12:08] While this is going, any questions?
[12:11] Anything
[12:13] on this scenario that
[12:15] is unclear or you have a
[12:18] question or you want to
[12:23] I have one thing. Will it pick up all the files present in
[12:29] the subscription based on the ready file
[12:32] or a single file? How it is?
[12:35] Oh,
[12:36] okay.
[12:36] So it depends on how you configure it.
[12:39] If you saw when I so let me go back into the account. Give me a second. So when you spend when you pick when you kick off the in this case, I'll put everything it finds on the other side
[12:54] because
[12:55] I put my pattern as star dot star.
[13:00] Okay. So
[13:01] and that's where you can specify when you do come on.
[13:08] When you
[13:10] when you configure the step over here at the bottom,
[13:13] where you have to download for the download pattern, you can leave them as they are on the site if you want to,
[13:19] and then it will do whatever the site says.
[13:22] Or you can override any of them. In this case, I I just put everything that has at least, you know, start off start,
[13:29] which is
[13:30] as file globing, basically, everything that have a extension.
[13:35] But
[13:36] you control it here, and you can even use variables.
[13:40] So you can build the new pattern from the old from the current file name or from the date for all we know. You know, you can do it any way you want.
[13:51] Okay. And another thing
[13:53] Yeah. How this is different from the adding just in the partner
[13:59] and attaching and transfer
[14:01] site
[14:03] rather than using this? So what's the main
[14:06] difference between these two approaches? I wouldn't
[14:10] send to partner and transfer site to do same thing then.
[14:14] This is pulling from the partner. We're not sending the file to them.
[14:18] This is issuing a pull attempt.
[14:21] So we'll be bringing in more files.
[14:23] It's the opposite direction.
[14:27] A partner is a you are saying an external partner. Right? Yeah. We're connecting to an external partner to get more files without the second scheduler
[14:36] so that we're monitoring for the single file for the ready file.
[14:40] And once the ready file is there, that means the partner has all the files we need. So we will go grab them
[14:45] instead of
[14:47] having a schedule running at 5PM and hoping that they'll be there for us.
[14:53] Is that So monitoring
[14:54] Yeah. Go ahead. Means monitoring
[14:56] means we are continuously monitoring
[14:59] or only for that particular time we are checking your file is there? No. No. No.
[15:05] This is part of routing,
[15:07] it will be based with the ready files. So here is how it happens. See, we have a pull. It happened.
[15:13] So in this case, I uploaded the file.
[15:18] This is a non scheduled action. Someone from someone uploaded the file. Right?
[15:23] At this point, we go and pull the files from the other place
[15:29] as opposed to having the schedule running every five minutes to look for the files.
[15:34] But who is uploading the file? The external file or or No. No. No. The application in internally,
[15:41] an application, maybe a business
[15:44] will upload the file in order to trigger the multi gate.
[15:48] It can be anyone. It can be your second application telling you, I am ready to process. Go get me data.
[15:55] It can be your manager saying, guys In in order to to in
[16:00] order to avoid to use the scheduling.
[16:03] Yes. So it's not like a watch job.
[16:07] Anytime
[16:08] someone can upload the file and it will it will trigger the pulling from the remote site. Yes. And it can even be another scheduler.
[16:17] You can have a scheduler looking for the ready files every five minutes and only pull all of the other files when the ready file arrives.
[16:25] It doesn't matter.
[16:27] It can be an upload to the API.
[16:30] It can be a schedule to control him or whatever externally you are using to say go start processing.
[16:37] It can be a phone call from someone saying, we have all the files. Go grab them, and someone goes and put the file on.
[16:44] And you can even tie it with the folder monitor by putting the file just on the file system and grabbing it from there.
[16:51] Right?
[16:52] The idea is that instead of having the pool
[16:55] specifically tied just to a scheduler,
[16:58] now you can use the pool inside
[17:01] of the routing
[17:02] when the file arrive. You can even make it more intelligent. You can do things like, if we pull from one site and there are no files there, go pull from another place.
[17:13] Or if a pull errors out,
[17:15] go into the conditional route for the error and pull from there.
[17:20] Right? You can play with it any way you want.
[17:24] I have one question.
[17:27] So current subscription
[17:28] will be pulling a file which transfer site is mapping. Right? It it will be downloading the file from the only one source. So in this approach, can we map multiple source to a single subscription folder?
[17:41] And can we pull those files from a different source location?
[17:46] Yes.
[17:47] Okay.
[17:49] So
[17:50] Annie, I I have a question about the scheduling part.
[17:54] You can only set that scheduler to go pull files
[17:58] from
[17:59] a location
[18:00] at a
[18:02] one time. It's in this configuration.
[18:04] So, like, if I wanted to look for files every hour, that would be global for the entire for every whole job that I would have in in secure transport. Correct?
[18:15] No. So this is a single
[18:17] one for this specific case, only one time. Mhmm. But if I had multiple
[18:24] sources,
[18:25] you know, different vendors Yeah. And one vendor once I look every hour and another vendor I look every four hours or something like that, is that is that possible to make that type of, you know Yep. File block? I and because I was reading,
[18:40] I thought I could only do that one time at the configuration
[18:42] at the server configuration. Yeah. I think you're mixing up a couple of different features here, and that's why they're getting little mixed up. So for this one, it's really just a reactionary pull. It have nothing to do with scheduling per se. So if you have two separate subscriptions, one of them looking every hour, the other every four hours Sure. That's okay. You're still okay. Okay. It has nothing to do with this. This is really if something happens, go get me more files kinda thing. So don't think of this as a replacement
[19:14] for scheduling.
[19:16] It is really giving you additional way to dos to do pulls
[19:20] that we didn't have before.
[19:24] Okay. So scheduling still stands.
[19:26] You still can use scheduling, and that's why we have the cron jobs there, and they're a lot more powerful powerful than they had ever been. Right?
[19:35] But
[19:36] if you need to do something on demand, if you need to do something based on what happens inside the fraud, now you have that option this option as well.
[19:46] And, yes,
[19:48] in a specific case sorry. Go ahead.
[19:51] And here, I have a question. And after when you pull the,
[19:55] multiple file
[19:56] Mhmm. You have to to create,
[19:58] another subscription in order to send to a partner
[20:01] as a Yes. A safety safety monitor?
[20:05] Yes. So it arrives
[20:07] as if it was so when you when I created the route, and I'll go back to it.
[20:15] When we created the route, you specify which folder to put it into.
[20:19] Right?
[20:21] If this folder is a subscription folder,
[20:25] then it will do whatever the subscription folder tells it to do.
[20:30] So in this case,
[20:34] I'm
[20:36] pulling into test.
[20:38] Oops. Sorry.
[20:40] Where is my test? Yeah. I'm pulling into the test into the test folder. This is the one where I'm pulling internally.
[20:46] If test is a subscription folder, when the files arrive, they'll start getting processed based on whatever the test folder is supposed to be doing.
[20:54] So if you want to send them to a partner, you put the subscription on the test folder, and it will send them.
[21:00] For ST, this pool will look as if it is a scheduled pool.
[21:05] It's just or ST, you press the button pool now.
[21:08] But it happens not based on an admin on an admin UI and not based on a scheduler,
[21:14] but based on something happening in another route.
[21:20] Okay.
[21:21] How
[21:22] we can specify this source location multiple source locations here?
[21:28] You do multiple post one after another and select the different sites.
[21:34] It's a step.
[21:36] Oh, okay.
[21:37] So in the same step,
[21:40] we cannot mention
[21:41] two source locations. Right? No. We need another step
[21:45] to do it. Okay. Got it. Yeah.
[21:48] Well, I'm not sure if this is not a multi select, but even if it if it is, then it might work. I haven't tried quite honestly,
[21:55] but it will tie you through the same download folder, download pattern. So so I will do it with separate steps.
[22:02] So if you want to use
[22:04] one scripture and pull from, like, multiple
[22:08] transfer site locations, you could just add pull the partner step for each location you pull it from within that road. Correct? Yeah.
[22:15] Okay. You you can use it for that. Yes. Yes. So it'll it'll it it so it'll pull from one one step after the other.
[22:23] How how about the scheduling? How do you how do yeah. My my question is around the scheduling of the pull in the route. What once you activate the route, is does it pull
[22:33] all the time, or can we define
[22:36] down to the minute
[22:37] minutes and hours? It's one yeah. It's one time pool. It says pool now. So when the file arrives, go and pull me the files. It works the same way like the
[22:47] pool on the admin UI. You know, when you set up a subscription,
[22:51] you have the pool now. Yep. That's what it does. It's basically the same idea. It's a single pool. It's not a schedule. But
[22:59] if your
[23:00] if you your
[23:04] at the moment, I'm doing it when the file arrives. Right? But you can also do it when, for example, you pull from somewhere where there are no files and give this external schedule or is every hour,
[23:16] then this will become every hour.
[23:20] Okay.
[23:21] Right? So think about so I'm showing you some pieces, and I'll show you a few a few more and we'll tie together a lot more things.
[23:29] But the idea is not to give you complete scenarios all the way through, but to give you a little bit more pieces and show you how things that you might have not talked about might be connected together. Right? That's the idea.
[23:45] So
[23:46] okay. Any questions on this one before I move to the next one? I Just one one one question. Yeah. Regarding the send send
[23:56] trigger on failure.
[23:59] Yep.
[24:00] Oh, it work.
[24:04] If there is no file to pull
[24:06] Oh.
[24:07] Oh. Oh.
[24:09] Okay. So
[24:11] over here, when you create the pull,
[24:15] you know that we you can do on
[24:22] on success or on error. So if you pull files and there are no files, this is basically a success.
[24:30] But you can also execute the route when the remote server returns no files.
[24:35] And if you
[24:36] get if you use that,
[24:39] then when you create the route, you can do a conditional route that says and the wild card is empty,
[24:45] and then create the route specifically in this case.
[24:50] This is one of the very early features of the product that we built it all the way in 05/03. It was there. It's
[24:57] but if that's what you want to do. So remember that when you create the subscription, you have a lot of other places where you can enable the routing and then send it into the routing.
[25:07] That's why, as I said, I'm looking to do with the positive because it's the easiest one,
[25:11] but it also can be done with the,
[25:14] other one.
[25:16] And we have something in the chat. How many how many maximum steps can we set in an advanced routing for a single account to avoid performance issues?
[25:24] I don't know. My biggest one is about a 100, and I haven't had troubles. So
[25:31] it depends.
[25:32] Now the problem with pools, as we know, is not about how many pools we have, but how many files it finds in each pool.
[25:40] So,
[25:42] you know,
[25:44] that's the problem.
[25:46] Okay. Other questions.
[25:48] How do we get the pool from route step? Nope. That is actually out of the box.
[25:52] As long as you are on, I think, February of later or later later, I don't remember when we added it, it's out of the box feature now.
[26:00] It is hiding.
[26:03] If you see my browser. Right?
[26:06] When you're building the route,
[26:09] it is
[26:13] yeah.
[26:22] It is over here now on between the published account and sent to partner. So this is the May release.
[26:28] I it was already there in the on the February release for sure, maybe a little earlier. I think it's from December,
[26:35] but I don't remember the exact version. If you're on an older version, come come to the new version, please.
[26:43] Okay. Other question.
[26:46] It
[26:51] can be from one source, the pool for partner, yes, we answered.
[26:56] Multiple defined location. Yep.
[26:59] And that yes. And to answer again that question, it came up in in the chat, but I will also voice it, and I can sure see Razvan is trying to get the question in. So, he's next, and then we move to the next scenario.
[27:13] But the answer
[27:14] the yes. This is one way to actually have multiple
[27:19] tools to go into the same subscription fold,
[27:22] something we could never do before.
[27:25] Right? You run one of them, and then inside of it,
[27:29] you call the others. Just be careful not to do it too often. Keep in mind, those pools are independent,
[27:35] which means that if you have one that takes three hours and you run another one, they'll step on each other toes. They are not like the scheduler, so they will not protect each other.
[27:44] So you need to think about that. That's very similar how you need to think about it if you do on demand
[27:50] to the API, for example. Razvan, what's the question?
[27:55] Hello.
[27:56] Can you retrieve the listing of the remote directory so that we can see what files we can download?
[28:05] Yes and no. You cannot from the step itself.
[28:09] However, because it is a transfer site, you can just open the transfer site and
[28:14] fill in the download folder and click on the list from there.
[28:19] So
[28:21] You cannot have the list on one file?
[28:27] What is Basically,
[28:29] we need to so we need from so for so our our internal customer would like to have the listing.
[28:35] So he sends the message,
[28:37] retrieve instead of retrieve the file, retrieve the listing from the external partner. Yeah. Then we then we want SD to retrieve the listing
[28:45] from the external partner and forward it back to our internal customer so that our internal customers knows, okay.
[28:52] These files are now available. I will,
[28:55] issue
[28:56] retrieve messages
[28:57] for each of them. Okay.
[29:00] Not from this step. However,
[29:02] this is already available in the API.
[29:05] So
[29:06] you can go to the API, grab the list from there, send it to your partner, and then they can use it to issue whatever they need to issue.
[29:17] But not from this app.
[29:19] With with API, it's
[29:21] it's available?
[29:23] Yeah.
[29:24] It's basically
[29:26] okay. Hold on.
[29:28] When
[29:30] you come to a transfer site
[29:34] come on.
[29:38] My server is a little sleepy.
[29:43] You you have this list button over here on the download folder?
[29:47] That is also API able.
[29:50] So So if you go to the API, the list is also available to the API.
[29:59] But that has nothing to do with what we were just talking about. It's just basically one of the things we added into a transfer site earlier.
[30:06] So if I click on that
[30:09] and wait it a little bit,
[30:13] see, that's my list.
[30:22] And if I go to the swagger, I will have the same
[30:26] Yes.
[30:30] Let me park that. Remind me towards the end of the meeting so I can go through the rest and we'll go through the swagger if we have time at the end.
[30:38] Because if I switch there, I'll never get to the rest of the scenarios.
[30:43] I like APIs.
[30:45] You know? But yes.
[30:48] So for the ones that haven't realized that while I'm here anyway,
[30:52] you know, we always had the test connection for some of the sites. Right? The test connection are just connecting.
[30:57] Now that you have the list as well, you effectively can use it to test if the download or upload folder are the correct ones you're thinking about.
[31:06] Because if you can list them and give them the correct files, then you're in the correct place.
[31:11] But yes.
[31:15] And, yes, if you have the list,
[31:17] then you can use the
[31:19] you grab the list, you have now the file names, and you can feed them into with dot radio or whatever to go through them one by one. That's a valid scenario.
[31:28] Okay.
[31:29] Any other questions before I switch over?
[31:33] We have Vishnuda
[31:34] and Morita that raised their hands.
[31:38] Okay. Vishnu, go ahead.
[31:40] Can we stage the download of from the
[31:44] routing?
[31:45] Like, pull 10 files first then then and so on instead of holding? No? Nope. Nope. It is basically go grab whatever you find. It will
[31:55] if the site itself is carrying how many per time, then it will grab the first end and 10 connections. You know, you can't control how many connections, but you cannot control which files come
[32:05] unless they are file name based. The only the only thing we have is file name based.
[32:11] And the other one that is raised up is,
[32:16] Marita.
[32:18] Hey, Annie.
[32:20] This question's
[32:22] regarding the destination folder and the download folder from the
[32:26] the pull itself.
[32:28] I'm not fully
[32:29] grasping the the use of I mean, the use of it.
[32:33] So the the the pull transfer site is still assigned through a through a subscription. Right? Yes. So what so I
[32:42] think I missed which subscription you're assigned to and why
[32:46] for the destination folder and the download folder,
[32:51] is that a different subscription that they have to go to?
[32:55] It can be or it can be the same? If it is the same so it gives you better control. If you leave those empty,
[33:02] it will pull them into the current subscription folder and into the using the whatever the site says the download folder is supposed to be on the other side. But you can override them during the pull based on where you are and based on what you are doing. So for example, you can pull a file from the mainframe
[33:21] on the main on the main subscription.
[33:23] It tells you that you need to go process something else, and then you can connect to someone else's SSH site and pull a file with a totally different pattern
[33:32] where the mainframe file name might be the pattern for the other site as well.
[33:37] Okay.
[33:38] Okay.
[33:40] That part, I I I get the the pull side the pull part. So if you if you define a destination folder, does that mean it will not gonna save a copy of that file within the subscription that's using that pull?
[33:53] It will will it just send that file within the destination folder that you specify?
[33:58] It will point in whatever
[34:00] folder you specify
[34:03] on the pull
[34:05] screen.
[34:06] And then what happens to the file will depend on what
[34:10] subscription you have on that folder. On that folder. Well in our subscription, it will be processed through the same way that we just processed. It will just not be dot ready. It will go to the next steps here.
[34:21] It is in a different folder, it will be a different subscription. Think of that as Daisy changing subscriptions in a way. And that will it this becomes a nightmare for monitoring, yet we all understand that. Right? It can be very, very, very complicated.
[34:37] But
[34:39] it also gives you a lot of power.
[34:42] So
[34:43] fully configurable.
[34:45] Very annoying sometimes.
[34:47] I I will admit that at one point, I spent an hour chasing down the file because I kept these changing things one after another to play with them.
[34:55] It made about 25 transfers before I figured out what I had done wrong. I basically created a loop.
[35:01] Not very good idea.
[35:03] Okay.
[35:05] So
[35:05] that was the easy one. So let's talk about so this was for how to get more files in, and let's go do something more interesting for the rest of the meeting, which will be how to process the files we got a little better. It doesn't matter if the file landed by upload,
[35:20] if it landed by pull, or it landed by this new on demand booking gate. Once the file lands into the folder, what we can do?
[35:31] One of the things that had always annoyed me historically
[35:35] was that oops. Sorry about that. Is that SD is was very good at doing multiple processes one after another, but then when you try to send a mail notification,
[35:46] you don't know exactly what happened or
[35:48] even worse, you don't know
[35:51] when something happens. If you have sentinel, if you have if you look at the tracking table, it will tell you what started when.
[35:58] You know? But chasing down, if you have a process that takes a couple of hours to complete, for example, PGP on a very, very big file,
[36:07] there is no way when you send a notification for your end partner to tell them this file got uploaded at 3PM, for example, it got processed,
[36:17] got into processing at 3PM, and then it got PGP encrypted by 6PM.
[36:22] So here is the file.
[36:23] If you try to take the time, of course, you can grab the time, put it on the file name, that becomes a problem because now you are encrypting on zipping, so the name becomes embedded.
[36:34] So what
[36:36] R and D came up with was
[36:39] a new step allowing you to set parameters
[36:41] inside of a route and then reuse them later inside of the route. And the easiest scenario that I came up to show you how to do that and the one that I think is going to show you the things easiest before we move to harder things is
[36:56] what we'll do is when the file arrives, we catch the timestamp.
[37:00] It can be a timestamp. It can be anything else from the environment, including what protocol was being used. We go process the file, and then we use the previously
[37:10] thought information
[37:12] at the very end together with the current time.
[37:15] So that you can send an email
[37:18] to, explain when the file had been received
[37:23] or
[37:24] to allow
[37:25] the mapping.
[37:26] And you can even reuse other variables
[37:29] so that you can build a full blown expression using an email or rename operation.
[37:34] And that makes a lot more sense on the live server, so I'll just go to that.
[37:40] Now unlike
[37:42] the
[37:43] previous the pool partners,
[37:45] for this one, you actually need to go and download the step. The step is a pluggable step.
[37:51] It's called,
[37:54] it's called set five set something. Sorry.
[37:57] Let me go and find that dates.
[38:01] It's the one.
[38:08] It's get called set flow attributes.
[38:12] And the example I have is pretty straightforward
[38:15] just because I didn't want to deal with a lot of examples. But what I'm doing on the set flow attributes is just to get a new variable called received.
[38:25] And here I put the date of when exactly
[38:27] this route starts.
[38:31] The example I built, again, not as you so what you can do is to add it as a new parameter. You can override an existing if something is coming from the subscription. You can do the dates here. You can use pretty much everything from the expression language. So you can use other variables,
[38:47] including the variables that are already on the flow parameter, the flow attributes.
[38:52] You can use the account name of the user that that send it and so on. And then in this case,
[38:58] because it's just for illustration on my sent to partner,
[39:04] which is the next step,
[39:08] all I'm doing
[39:11] is to create
[39:13] after I sent the file to the partner,
[39:16] then I create a dates TXT file as a trigger for them, put the date that we I just set up, the one oops. The one where that the file arrived in,
[39:28] and then send it to them
[39:30] so that when they open it, they know when the file enter processing on our side. That means that if the push itself
[39:37] took three hours because it's a huge file, they know when we started
[39:42] it without the need to try to figure it out based on their server logs.
[39:47] This can be also you can used in emails,
[39:50] obviously.
[39:51] It can be used for anything else, and it doesn't need to be a date. I just like playing with the date because it's very visible. If
[39:59] you need multiple of those, you can string them together. You can set up multiples of them, but you have multiple steps. You can have one of each between the steps, and then you can put them together as a
[40:11] bread crumb to explain to the partner when something was processed.
[40:15] Of course,
[40:17] that allows you more flexibility than what you can do with something like Sentinel, for example, where the whole routing comes as a single package.
[40:24] But it also,
[40:27] gives you the clean way
[40:29] to add to to catch additional data without tagging it onto the file name. Until now, the only way for us to do things like that was to basically tag everything on the file name over and over and over and then pass the file name.
[40:45] Okay. So let's see how that works. So the way it's built at the moment is anytime I upload the file, it just set the flow attribute
[40:53] and send it to the partner and then send dates file after that.
[40:58] And which folder that might be? Let's see.
[41:03] Send date, I think.
[41:06] Okay. It it kicks me out. While I'm logging in, any questions on this one?
[41:30] No questions?
[41:31] Okay.
[41:35] Or did I confuse everyone?
[41:38] Okay. Let's go and check into the upload folder.
[41:44] So in case it's not clear with my browsers, that server one where I'm playing on, that's my receiver server, and those are the two clients on both sides.
[41:59] And I probably should have changed at least the colors on one of them.
[42:03] And if you see that,
[42:06] here is the dates file.
[42:13] And I'm not sure. I don't think hold on. You're not seeing my notepad, are you?
[42:19] Let me share that.
[42:27] Notepad.
[42:27] Here it is. Okay. Can you see the notepad now?
[42:34] Yes. Yes. And that's it. It basically just put it into the file and send it over. And, again, this can be used in various ways. I just prefer to work with files and with mails, you know, less clients.
[42:48] But, in the real life scenario, you probably will want to send that to a mail,
[42:52] or you will want to send it to another server, or you will want to call it somewhere else and so on. So any questions on this one?
[43:02] It's pretty straightforward
[43:03] once you figure out how to use the variables,
[43:06] and it's like it's more like waking your appetite for the next ones because I'll be using the set flow parameter a lot for the rest of the scenarios.
[43:18] So
[43:19] any questions on this one?
[43:24] No?
[43:25] Let's see. Something in the chat. Is there any any
[43:30] any place where we have
[43:32] I know, like, you know, we have variables and and hours, transfer size, account. Like,
[43:38] sometimes, you know, it's it gets very confusing which which one is available where.
[43:42] You know? I know.
[43:44] So it's like, it would be helpful because I use this, you know, flow attributes and subscription attributes a lot. And and it would be helpful if you can have, like, some kind of guideline put in one place.
[43:55] Because it for not right now, it's like, you know, trial and error for me. You know, if I need a variable, you know, for the transfer side, I know it's a variable name in caps,
[44:03] and others are literally, you know, try it out kind of thing.
[44:08] Unfortunately,
[44:09] no. Because
[44:10] if you think of ST at the moment, it's not linear in any way or point. And especially with some of the things I'll be showing to you in about five minutes, you'll see it gets even worse.
[44:21] Because with the flexibility comes the inability to actually see what is where. And the more variables you use,
[44:27] the less likely is for us to be able to give you a meaningful report.
[44:31] Can we show you what is there? Sure. But especially with conditions,
[44:36] it becomes a nightmare.
[44:39] Yeah. Is that's
[44:41] go ahead. Is that set float attributes available on specific version of,
[44:47] ST?
[44:49] Yes.
[44:50] Again,
[44:51] it is about a year old, but you need to download. It's a pluggable step.
[44:57] Okay.
[44:58] So you'll need to, download it and put it on your server. It's one of those, plugins thing is,
[45:04] but it's pretty straightforward and a very useful one.
[45:08] So if you don't see it, you haven't downloaded it. You'll need to grab it.
[45:13] Unlike the
[45:15] pool tank. The pool tank came out of the box. This one, they decided to keep out. I think that this one is more useful,
[45:21] but that's just me.
[45:22] Okay.
[45:27] So I'm not going to switch back into the presentation. You'll get the presentation. There is two, three more
[45:33] screen scenarios,
[45:35] but I'll just stay here because it makes more sense. So this was the easy one. So we just set the parameter,
[45:41] and then we just send it to another user.
[45:45] Right? Send it to the partner
[45:47] in one way or another, or we used it. Don't forget that once we set the flow parameter,
[45:53] now
[45:54] and that's one thing that I kinda
[45:57] used in the previous example and never pointed out. Don't forget that now the step themselves can use expression language.
[46:04] So if you have a set variable, set parameter,
[46:07] you can actually use it to build expression language for the next step as well.
[46:14] And we'll get to back to that in a second. So
[46:17] we have the day. How
[46:21] what happens if instead of we're sending to a partner,
[46:26] we want to publish to an account
[46:29] and use
[46:31] that parameter as part of the processing on the other side for the other account. For example,
[46:36] you have multiple files arriving into a single account that get distributed based on the file name or based on folder. It doesn't matter.
[46:45] And then you have a subscription triggering on the other side,
[46:49] on the other account
[46:51] that
[46:52] might that needs to use the information from that this one.
[46:58] For example, telling you which protocol the original files came on.
[47:03] It's pretty straightforward if you are still in the routing for this file. Right? You have,
[47:09] transfer dot protocol, which will tell you what had been used.
[47:13] However, once you pass the file to the next user in the chain to your receiver,
[47:19] that variable is not there because the protocol used to get the file to them is
[47:25] routing.
[47:25] Right?
[47:27] That's where using the variables actually helps because once you set the flow parameter, if then you trigger another subscription directly on the same server,
[47:38] all of those parameters will get moved to the other subscription as well so they can be used.
[47:45] But I know I lost everyone, so I will go show you what I'm talking about.
[47:55] So
[47:57] what I'm doing here,
[48:00] again, setting a flow parameter. I don't even remember what I set on this one. So
[48:05] the current date and the sender email.
[48:08] I hard coded the email, but this might this is also the way to pass the email of the current user.
[48:14] If it was a flow parameter on the flow itself, it will automatically be sent to the other flow.
[48:20] But
[48:21] if it was the the specific email or the,
[48:26] standard account dot email,
[48:29] then it will not. So you can remap it from here. If you want to put multiple of those, you can add them one after the other, one per line. Don't put anything at the end. Just the end of line is what telling us there are different variables.
[48:43] And add new and override
[48:45] so that if there are any already set through the flow param through the hard coded flow parameters,
[48:51] they will also be able to be done here.
[48:54] And then I'm publishing to an account into the receiver, another account on the same server.
[49:01] Nothing interesting here.
[49:03] Absolutely just triggering the subscription.
[49:07] The so that on the other end,
[49:09] when the subscription is triggered,
[49:13] we can see what happens.
[49:16] And then if I go to the receiver account
[49:21] I have too many accounts.
[49:23] And I go to the receiver account,
[49:27] into that route.
[49:31] I'm just
[49:32] being lazy
[49:34] and
[49:35] just renaming
[49:37] the file so it's very visible.
[49:40] But, of course, you can use other things.
[49:46] So what I'm doing couple of things. When the user downloads a file, the file so when the end user, the one we just push the file into, downloads the file,
[49:56] What I want to do is to either send the mail. In this case, I am renaming
[50:02] the file
[50:04] with the sender email
[50:06] instead of so that I can see it visually
[50:08] and then publish it back
[50:12] into the receiver account so that we can see it.
[50:16] And this is the conditional one.
[50:18] The non conditional one
[50:21] is
[50:22] when the file
[50:25] arrives.
[50:29] I'm doing another rename
[50:32] with the current date that I sent.
[50:34] So the way that is built is when the file is published into the account, it will auto it will be renamed
[50:42] to use one of the variables I sent. When the user downloads the file, it will use the other variable.
[50:48] And, again,
[50:49] I'm doing that with renames just so that we can see it very fast, very visually on the screen. You usually will want to use that in sending emails or for notification
[50:59] purposes or for reporting to Sentinel eventually.
[51:03] But what it will show you is how variables we set in another account
[51:09] when the file arrive get used, and it can be days later
[51:13] in another one. And this is where I will stop and actually mention something that I don't think we had documented very well.
[51:21] Flow parameters,
[51:22] when you set them, are saved into the dot STFS folder for this file. You know, every file we have on the file system in a user's folder had the dot STFS folder. Right?
[51:36] When we set the flow parameter
[51:39] through the ST server through the through this new step. It does not save in get saved into the database. It goes into dot STFS folder structure.
[51:49] If you delete dot your dot STFS
[51:52] folder structure
[51:53] and some
[51:55] then those disappear.
[51:57] So
[51:58] especially in this use case where we pass them from one to another, they'll be available during the download for as long as
[52:06] dot SDFS survives.
[52:08] So if you're going around and cleaning your dot SDFS folders, please don't.
[52:14] Okay.
[52:16] Questions while I'm testing?
[52:21] So is it
[52:22] is it possible that when we set up a flow attribute, let's say, account and, you know, we place a file there and it it gets passed on to,
[52:31] you know, account two
[52:33] Mhmm. Is there any way to maintain like, receive the values of flow attributes in the second account? No? That's that's exactly what we are doing here. Okay. So I'm in the Annie account,
[52:44] and I'm uploading
[52:46] a file.
[52:47] Right?
[52:48] Okay.
[52:50] Let me go into the receiver account.
[52:58] Log in.
[53:05] That's exactly the scenario here.
[53:08] K?
[53:09] And if we go into
[53:14] the test where they arrived,
[53:17] here is our file.
[53:18] See? It changed the name with the time we caught when the file arrived.
[53:23] Remember what it was doing
[53:26] was when the file arrives
[53:29] unconditionally,
[53:31] use the current date we called earlier.
[53:33] So, basically, here, can hard code any value in you you have used it, but I can hard code anything. Right? Yes. The the attribute okay. That that's wonderful. Thank you. Whatever you set up in the one account, you just need to use the same name in the other account, the same variable name. It basically changed.
[53:51] And now the more interesting one, if I download this file
[53:56] and let me just delete this one because, you know, I didn't do a properly named cleanup. So as you can see the list, I don't have anything that has email on it. Right?
[54:06] Right?
[54:07] So this is also the original file because I don't delete it. So the file write discovery eleven GPG, see the time. Right? It arrived at 85753.
[54:17] But when we renamed it, it was
[54:21] I've done well,
[54:25] my expression is wrong, but it's basically 5751.
[54:28] It's the original time of the original arrival. If I download this file,
[54:38] and refresh,
[54:42] if I haven't messed up something,
[54:46] here is how the file got renamed again with the other with the other variable that will came all the way from the other
[54:53] the other earlier.
[54:54] It was not connected to the receiving itself.
[54:58] So it could have happened three days later for all we know.
[55:06] And, again, I'm playing with renames because it's easier to show it on a screen, but you could have used this email to send notification to the original
[55:15] uploader of the file that the final recipient downloaded the file.
[55:20] And I would imagine all of these, transfers would have same core ID in from tracking?
[55:26] Let's go check.
[55:28] That's why I have a live server
[55:30] at DocBook PowerPoints.
[55:38] So
[55:42] okay.
[55:46] And, of course,
[55:48] then this file got renamed one more time because I have an unconditional rename or receiving that I'm in the same folder. So I mentioned something about not doing things like that. Right?
[55:58] Yeah. But,
[55:59] yeah. Put it in a different folder when you are renaming. I forgot to to change that. So but I don't think I created. Let me see if I need to stop something.
[56:10] No.
[56:11] So but, you asked for a core ID? Core ID will be the same
[56:18] yeah. Because it's a published account. It's the same file. But
[56:24] we can go check.
[56:26] Yeah.
[56:30] So on this one,
[56:32] core ID is this.
[56:35] I really don't know the answer to that one. So we'll go check. So copy,
[56:39] and then let's open the original flow,
[56:43] which was this one.
[56:48] Yes.
[56:52] We can't see anything. I'm I'm not sure if you're doing anything.
[56:57] Yeah. Because it's in a different screen. I apologize.
[57:00] It's the same one.
[57:02] Let me
[57:04] okay. Give me a second. Let let me share in a different way. I know what's going on because I have a lot you know, they're the small screens, and I'm just sharing a single screen.
[57:12] So sorry about that.
[57:15] And when you don't see things, please let me know. Okay. Do you see my screen now completely?
[57:20] Yeah. Yes. We do.
[57:23] Here it is. So let me repeat that because it's important. So this is the last step. So as you can see, a lot of things were happening because I was doing a lot of weird things.
[57:33] But this is the very last of them,
[57:39] And this is the transfer ID. Not we are talking about transfer ID. We're talking core ID. Right? So let me click on the core ID,
[57:47] and you can see all of them are under the same core ID.
[57:52] Awesome. That's great. Okay. Yep. It's the same file. We just moved it from one place to another. So we and, yes, you can also put the core ID. So you don't need to put the core ID into one of those variables because as you can see, it's the same one.
[58:06] But if you want to pass it through a transfer site, that's another way to send the core ID.
[58:11] The idea of this was to show you how
[58:14] you can actually daisy chain the file, receive into one account, publish to another account, and then give information to the second account about what happened in the first one. I did date and email, but you can do original protocol. You can if it was a pull, you can do which side it came from.
[58:32] Or you can send some information about processing, you know. Or even you can send send a string, an email
[58:38] that then
[58:39] a string for an email that then can be using into email on the second one. They download the file or when they do something to be sent back to the original center.
[58:51] Okay. That's makes sense? Yes.
[58:53] Yes. Does. Thank you. Okay.
[58:56] I know I'm a little behind on chat,
[58:58] so sorry about that, guys. I'll try to get to it in a second.
[59:03] So any other questions on this one before I go to my favorite scenarios?
[59:13] And are you going to use that in your daily life? Maybe. But if you are using published to account and especially if you're using distribution accounts where one account pulls the file for multiple accounts and then distributes them,
[59:26] that's a that's a very good way to actually do things like that.
[59:31] It also
[59:33] in some case so
[59:35] this email, for example. Think the scenario when one account pulls from your mainframe, for example,
[59:42] and grabs file for 10,000 users. If you want to notify the mainframe team that file was downloaded from the end user, the only way to do it until now was either to hard code the email
[59:53] inside of the route itself
[59:55] and to trigger everyone through the same route, which
[59:58] they might be doing different things. Right?
[60:01] Or you could do
[60:03] a hard copy inside the feature of the account to reach of the flows. In this way, you can just put it inside the flow parameter on the before you publish the file to them, and then it's there. And again, it is in dot STFS.
[60:17] It is not
[60:19] in the database.
[60:20] So if you do it that way,
[60:23] that's the only requirement.
[60:25] If I download if you download this file again, it will do the same thing because the variable is still there until the STFS file is still there.
[60:34] Okay?
[60:37] Okay.
[60:39] Any questions on that one before I move to the next one,
[60:43] which is something that you might have done a lot? So
[60:48] one of the challenges in SK is,
[60:53] as you know, if you have a
[61:00] routing and
[61:02] everything goes well, you go through all the steps, everything looks beautiful,
[61:06] now I have trouble. What happens if you fail on step one, but you still need to do something
[61:13] while only on error,
[61:15] but you don't want to continue on success.
[61:18] As you know, in the server
[61:20] and I'll just go and open my route that I'm playing with and start talking based on that.
[61:29] When you have two steps,
[61:31] as you know, when you specify the very first step at the very top, you could say continue on error,
[61:38] but it will always continue on success. So there was no way to do just an error condition, do something if something fails.
[61:45] That's one of the big changes they did actually.
[61:48] R and d added a new checkbox
[61:51] at the very top that says proceed later on both success and failure so that every step says what will happen when it succeeds or fails.
[62:00] This way, you can only continue on failure or you can continue on success, whatever you want to do in both cases.
[62:08] But not only that, but as we'll see in a second,
[62:11] you also can
[62:13] check-in
[62:14] the next step what happened in the previous one
[62:18] so that you can run a step based
[62:21] on the exit
[62:23] of the previous steps.
[62:25] So in this case, what I'm doing is I'm pushing a file.
[62:29] And then on the next step, if this push fails, I try another push to a different place.
[62:35] Basically, like a backup server or as a
[62:39] restaging
[62:40] the file to in a different place for being processed again or something like that.
[62:45] This scenario could have been done in the old world by just proceeding on error and doing nothing on success, obviously.
[62:52] But
[62:53] if you have a lot of steps in a different way, this couldn't be done because you once you continue only on error, that means nothing after that will happen. While here, we can proceed both on success and error, and they have different steps catching both the success and the error.
[63:10] The first step here is nothing interesting. I'm just sending the file.
[63:14] Nothing happens.
[63:16] Always
[63:17] proceeding on both success and failure. I want to go to the next step regardless if I fail or succeed.
[63:24] But then on my second step,
[63:30] when it opens,
[63:32] I only want to run this one if the previous step was a failure.
[63:37] So if the previous step managed to push the file, I don't want to do this step.
[63:42] But if something went wrong,
[63:44] I want to do that. And this might be anything from center partner. It can be a set flow variable.
[63:51] It can be published to account. You can do a rename. You can do
[63:55] any custom step if you had built any or anything like that. That gives you the nonlinear
[64:02] implementation inside
[64:03] of the whole route by allowing you to actually control what happens next and not just go down down the route.
[64:14] Any questions on this?
[64:24] And
[64:25] those are actually documented pretty well over here.
[64:31] So you don't need to,
[64:34] try to remember what we call them.
[64:42] No questions? Let me go send the file while somewhere if no one has a question.
[64:47] Oops.
[64:50] Where am I?
[64:51] I'm in the wrong account.
[64:56] And, again, I'm point playing with two pushes because it's easier to fail one of them,
[65:02] but it doesn't need to be a center partner that fails initially. It can be a PGP failure. It can be
[65:09] anything you want. And then the second action
[65:12] can be, again, anything you need it to be.
[65:16] In one use case I built for a customer in the last few weeks,
[65:20] what they needed is if the if the push fails, they just wanted the file to be restaged
[65:26] again for processing so that it can be processed later. So put it into a account and trigger another subscription to go do something else.
[65:37] Okay.
[65:39] Backup servers,
[65:40] upload the file.
[65:44] I'm uploading the same file over and over. That will be fun.
[65:51] Okay. Let's go see what happens.
[65:58] Oh, in case you haven't seen this UI before, so now
[66:02] my here is the it's failing now.
[66:05] Right? That's the first push. I make it fail because I need it to fail.
[66:10] And then once it fails twice because that's I I've changed my retries to be just one retry.
[66:17] And then because it fails, it went to the other place. If we open this one, you'll see that the upload folder over here was success one.
[66:26] And it failed because
[66:28] the path doesn't the path is invalid. Basically,
[66:31] one doesn't exist.
[66:33] And then when it figured out it cannot do it, it went and send it to the correct place, which was the errors folder.
[66:41] So that's where I wanted it to go.
[66:48] Any questions on this one?
[66:51] This is probably the most straightforward forward of the above them all.
[66:56] And, of course, if I go to my other account, here is the file. Oops.
[67:02] My other server where I'm sending the files.
[67:11] Here is the file.
[67:13] And don't look at the timestamp here. It's
[67:17] this server is in Europe somewhere.
[67:22] Okay.
[67:24] Can can we send maybe a status
[67:27] takes the file, like, send them, for example, back to the
[67:32] initial sender?
[67:33] Or the same situation?
[67:35] Yeah. Because I'm on a different server now. So if
[67:39] you if this was a subscription folder,
[67:43] it can send back whatever it needs to do.
[67:47] That's my partner server even though it's ST. I just don't do other servers. But yes. No. But I meant when so I meant to in the routing, when it sends back Ah.
[67:59] Yes. You send back you
[68:02] send back in the error. You sent the the actual file that was sent.
[68:06] Yes. But I
[68:09] But I meant to to send instead of the actual file to send back
[68:14] they still file with the status something. Uh-huh.
[68:17] Well, yes and no. You cannot send the status file on its own. However,
[68:23] what you can do is to do a send to partner to a local site and then use the trigger to send to the external.
[68:30] So
[68:31] and, yeah, this will mean one more file,
[68:34] but
[68:44] funny stuff
[68:45] anyway.
[68:46] So
[69:00] Audio is speaking.
[69:02] We don't hear you very well, or I'll at least I do not hear you. I don't know.
[69:08] We can't hear you
[69:11] any can you hear me now?
[69:13] I'll be back.
[69:19] So while Annie is trying to
[69:23] fix the sound,
[69:25] prepare your questions.
[69:35] Okay. Can you hear me now?
[69:38] Yes. We do.
[69:39] Okay.
[69:41] Well, I had to switch to the other line back to the other line. So sorry about that, guys. Technology.
[69:47] So
[69:48] if I heard the question correctly, I mean,
[69:52] you're asking if you can send a status file as opposed to the actual file to the second place. Right? Yes. Yes. Exactly. Yeah.
[70:01] Not natively because SD doesn't have a way to just send any file at all.
[70:07] However,
[70:10] if
[70:11] there are two tricks that you can use here, one of them is
[70:16] here in the center partner,
[70:19] instead of sending so send this file, our file, not to the original,
[70:24] but to a folder monitor locally somewhere.
[70:28] So they just drop somewhere on the file.
[70:31] And then you can use the sent trigger file to actually send to the SSH.
[70:39] Right?
[70:41] And I know it's a little roundabout and not exactly what you're asking, but we just don't have a step that just sends a random file, not the one we are processing at the moment.
[70:53] Okay. So, basically, use the trigger file sending it to another
[70:57] Yeah. You yeah. But in the opposite way. So send our file to somewhere locally or, you know, to us again, which, again, that means one more push. So if it is a big file,
[71:09] it might not be the best thing. So use a protocol that will not eat a lot of your processing processing time. I a folder monitor is great for that. You know, the push site on the folder monitor site.
[71:20] And then anytime you push successfully, now you can send a trigger file. How I did it with the dates file in the previous example.
[71:29] Right?
[71:32] Yeah.
[71:33] So that's one way. Another way will be to actually put the file somewhere else in a different to publish to account and then do something
[71:44] subscription, on the publish to account subscription,
[71:46] and do it from there.
[71:48] Or you can write a custom step that does whatever you wanted to do.
[71:53] It doesn't
[71:55] it doesn't need to be another center partner.
[71:58] Right?
[72:04] Makes sense? A little bit?
[72:06] Maybe? Yes. Yeah. Yep. K. Okay. So now the file went somewhere.
[72:12] But can you tell me so I have two different sites. It was sending to two different places. Right? Can you tell me where it went?
[72:21] If you look at the tracking table, obviously, you can. But if I want to send the mail notification
[72:27] or, you know, rename the file and put it in our hiding with telling me where I am,
[72:34] did you go through
[72:35] Door 1 or Door 2? Where did the file go?
[72:40] No one knows. Right?
[72:42] We know that it might so if it was a if if the whole thing ends in an error, we know it didn't go anywhere, so we know both sides failed. Right? But if the route ended in success,
[72:54] you don't know where it went through.
[72:57] It might have gone to the first server and never gone to the second, or it might have got to the second. Right?
[73:03] So how do we and if you have three or four or five of those, so you have multiple steps, you never know where exactly it failed. So
[73:12] now you know how to get the steps conditionals
[73:16] so that they work
[73:18] only on successes, failures, whatever. Right?
[73:21] But how do you know at the very end of the route what happened in that?
[73:27] That's where our good friend, the set flow parameters, come into play, And that's my fifth and last scenario, and we're almost on time.
[73:36] So let's make that a lot more complicated.
[73:44] So what
[73:45] I'm doing
[73:46] is
[73:49] keeping track
[73:51] of things,
[73:52] of course. At the very beginning of the flow,
[73:56] I'm stating the main server and the backup servers to know. I'm a developer. I always want to initiate
[74:01] my variables. Plus once they're initiated, it's a lot easier to check for them
[74:06] than
[74:07] to try to figure out, are they empty? Are they no? You know, anyone that had ever done any work with strings knows that trying to figure out if the string is there or empty or no is a pain.
[74:19] So
[74:20] we say we didn't send to anyone yet.
[74:23] Then we do our center partner.
[74:25] Always, everything is clean.
[74:27] Then we do a set flow attribute,
[74:32] which is if the previous step succeeded,
[74:35] set domain server to yes. We sent to domain server. Again, it can be a date here. It can be the server name. You can set up whatever you want. I'm just doing yes and no just as markers.
[74:46] Right?
[74:47] So it will be sent to yes.
[74:50] Then the next step is our conditional sending to the second site,
[74:55] which is on failure.
[74:56] Now and here is where it becomes interesting. Could I have used the same expression if the previous step is an error continue?
[75:04] I cannot.
[75:05] Because
[75:07] the previous step now is my set flow attribute.
[75:10] But the set flow attribute sent me a variable that will tell me where it went from. So if the main server is still a no, then it didn't set the variable to yes. That means I'm in the error condition, so I'll now send it to the wrong to the second place.
[75:25] Another set variable here on success of the previous,
[75:30] and only if the main server is a no. Because I'll always get to this step
[75:35] because it's next in the line, But I only want it to run if the previous one is success and the previous one was actually the previous push.
[75:48] And at the very end, guess what? I'm doing query name.
[75:53] No. I'm doing a push up. I'm doing a public publish to account
[75:57] where I'm just putting the two variables one after another so I can see what happened.
[76:02] And, again, this can be used in an email. It can be used for conditions downstream.
[76:07] It can be used to send the,
[76:09] the file somewhere else or just to let you know. But by setting the two variables properly,
[76:15] you actually now know where the server went from. So if it comes out as one yes, no, that means it went to the main server. If it comes at once no, yes, it went to the second server. If it comes at one no, no, it didn't go anywhere.
[76:34] Questions on this one while I go and play a little bit for Anita?
[76:45] Did I lose everyone?
[76:51] No.
[76:52] Can hear you.
[76:53] Did I lose everyone logically on what we have time to achieve?
[76:58] I'm
[77:00] I don't I
[77:01] don't know where Okay. Can I download that plug in,
[77:04] actually?
[77:05] Oh,
[77:07] we'll set of attributes.
[77:09] Yeah. It check go
[77:11] to the support site and just look for it. It's one of it's it's the same like you download s three, for example, or any other transfer sites.
[77:20] It's one of the plugables. It's just a plugable step.
[77:24] I will
[77:26] Nicole remind
[77:27] me to send you the link. We'll send you I will put the link into the mail that Nicole will send after the meeting. Okay.
[77:33] Thank you.
[77:34] Mhmm. Okay.
[77:40] K. And let's go over that. Was just saying, okay. Send me the link Yep. And I'll add it to the email. Yep.
[77:47] Yeah. Just remind me in about three second, you know, after we finish the meeting because I'll forget.
[77:54] Okay.
[77:55] So I'm uploading the file. It will try to get pushed.
[78:00] And then
[78:02] does anyone remember where I put the file after it pushed?
[78:05] I need in the test. Okay.
[78:10] And I have the
[78:11] that's an old file. Okay. We're still waiting for it. Let's see what happens now.
[78:17] Yeah. Yeah. Yeah. Yeah.
[78:19] Annie.
[78:26] What am I doing here?
[78:34] Sorry.
[78:35] Warnings.
[78:38] Yeah. So while,
[78:40] your test is running,
[78:42] there had been quite a lot of that activity
[78:44] in the background on Teams, so this is great.
[78:48] From what I've read, it was more customers
[78:54] answering
[78:55] to each other. Mhmm. But I I I'm not certain
[78:58] if there are remaining
[79:00] appointments.
[79:01] Yeah. That's the last thing that I'm showing. So after that, I'm going to go read chat while listening. So, guys, here is my file. One no yes as expected,
[79:10] matching what we saw here.
[79:12] File arrived,
[79:13] failed on the first, succeeded on the second.
[79:17] So
[79:18] questions on that
[79:20] or anything so far? Because that's the last thing I have. By the way, my
[79:26] presentation also have a couple of diagrams, you'll get them.
[79:30] And I will also send an export of the Annie account and the receiver account
[79:36] together with the PowerPoint. So you can just import the thing and
[79:41] play with that if you want. So questions?
[79:44] I'm going to go with
[79:46] So chat.
[79:47] So any
[79:49] does this flow attribute is represented, let's say, in any monitoring band to, like, central or ADI? So let's say we set up some. Yeah. It doesn't unless you send it. So if you go to your so you will not see it in Sentinel at all because it's not a standard variable.
[80:06] However,
[80:08] if you go to here, remember that you can remap variables.
[80:13] Right?
[80:14] Yeah. I can I can create variables here, but they will be, like, fixed ones? Right? Nope.
[80:20] Nope. Nope.
[80:22] That's the whole point.
[80:23] It's actually an attribute,
[80:25] and the name of that one will be
[80:28] a okay. I'm drawing a blank. I'll I'll put it in. Let me put a note. I'll put it in the notes. But, basically, you can put it over here.
[80:39] It's I think it is dollar sign and then flow
[80:43] dot
[80:45] nope. It's
[80:48] no. It's just a variable. So so, for example, the sender ID, if that's what I called it, this how is you call it here. You need to use a field that already exists, so pick one of the ones we are not using.
[81:00] And that's it.
[81:02] And then if it is in the four parameters for the specific file that goes through, it will fill get filled in. If it's not, it will not. That's all you need to do.
[81:14] Okay. Okay. Understood. But it's not con but it is not conditional.
[81:18] And, again, I'll I'll look up the exact syntax, and I'll put it in the notes or just, put it somewhere. Sorry. I was thinking about it this morning and forgot. But it's
[81:29] keep in mind that it's not conditional. So if you have a lot of those, you'll need to publish all of them. So it might be a good idea to reuse them between scenarios or not use a lot of them.
[81:40] Yeah. And also, like, it will become some to find, like, unused variable in center. I know there are a few, but it's it's not ideal. But not that many. Yeah. Right? But there are still a few you can use. Yeah. But yeah. That that that makes it even more powerful because that also means you can do your reporting on the other side based on that. I personally use the flow parameters mainly for emails or for things like that, but that's also an option.
[82:04] Okay?
[82:05] Okay. Good. And there is one more question I had that was regarding the the
[82:10] you know, we have this override upload folder. Like, in the route, we can upload. Like, let's say, we have a service account and you have a transfer site there and there is a Mhmm. Application also where we upload a lot of files. We can override the folder
[82:23] right there in in the out step. Right? Yes. To partner. Can we have something like an override download folder? Because every time we have, like, it is different. On the pull
[82:33] well, that only with on demand pull.
[82:37] Can you say it again? I like it. Okay. Actually, there are two ways for for that. Let me share the browser again because that actually is so one of them is with the on demand pool that we saw earlier. Right? But also,
[82:51] if you go to the transfer sites, depending on so depending on the protocol you're using
[82:59] Where is my transfer set?
[83:06] Am I in the wrong account?
[83:13] Oh, I'm in the wrong server.
[83:15] Sorry.
[83:20] That's what happens when you have two servers next to each other.
[83:27] So
[83:30] as long as you are new enough, see that this is now
[83:36] of you can use the date. Yeah. You can use an expression.
[83:40] You cannot use
[83:42] flow parameters are a little
[83:46] icky here, but you can try because some of them are actually usable.
[83:51] So put the flow parameter and try to use it. It will come as flow dot attribute something something.
[83:57] I I think this works fine. But if you close this, you know, tooltip,
[84:01] you don't there is, in upload
[84:04] Here, you see this override upload for our our folder. Right?
[84:09] I need something like that.
[84:11] It's automatically checked on. That's why it doesn't exist. For upload folder, you need to enable it. For download folder,
[84:18] it can always be over it from the API or from the on demand pool.
[84:24] But but we don't have this option. Like, when you go to pull or something. Right? Let's say you are scheduling a pull
[84:31] in subscription. Not on a schedule. Yeah. Correct. You cannot do it on a schedule, but you can do it on the on demand.
[84:38] Yeah. That's what, like, I was thinking. Because what happens, let's say, have a scenario where there is a one server which runs, like, a lot of middleware,
[84:46] and there are, like, 50 folders we have to download files from for different purposes. We have to create all those stuff inside every single time, like, we go and set set it up.
[84:56] Use the flow. So no. Use the flow parameter over here directly, you know, dollar sign and so on and see what happens.
[85:04] Okay. Yeah. Yeah. We'll we'll we'll try that. But I was thinking, like, maybe, you know, there is some design limitation. That's why we don't have something like this. There used to be. So yes.
[85:13] If you if we had this conversation a year ago, that would have been.
[85:17] Now it's a lot easier to do things because they had to open it a little bit for the on demand. But
[85:23] k. Okay. Thank you. Mhmm.
[85:26] Okay.
[85:27] Question.
[85:28] Update.
[85:30] Actually, I activated the flag file. It checks for the presence of subkey with PGP public key. Sorry, Thomas. We are not very good with PGP keys occasionally when you do changes.
[85:40] The plug in name, Donald, is
[85:44] set flow parameter,
[85:46] and I'll put the link in.
[85:49] Okay.
[85:50] Is there anyone that have a question? I'm scrolling up. I don't see anything that still needs answer, and I was answering a couple while I was going.
[85:58] Can you expect feature to override download folder? Oh, that's what you were just asking.
[86:02] Yes and no. It's already there. It's just not a checkbox.
[86:08] What cluster environment, what back end database easy to do. Okay. People are talking to each other.
[86:14] That's good. Okay.
[86:16] Anyone, any questions? Because we're running out of time, and we really need to wrap up. But
[86:22] You you can show show show us how
[86:25] you can activate the API in order to get the the listing.
[86:29] Yep. We can do that.
[86:32] And that will be and in the meantime, everyone else,
[86:36] last question. So as you know, by the way, if you don't know, couple of things will come here.
[86:42] If you haven't realized that, web client team editor is now on the admin UI. You can save your web client. It's a UI that allows you to change your,
[86:54] how the web client looks like, and then save it and upload it. We don't need to compile outside of ST anymore. And while it's loading, I'll also go and open the API.
[87:08] So
[87:09] from here, if you haven't seen that, please go and play with that. If especially if you need to change logos and colors and so on, it's now all plug and play,
[87:17] which is very nice. Okay. API.
[87:21] Sites.
[87:22] There are sites.
[87:46] K. On the post operations, I think that's where we hit it.
[87:52] Oops.
[87:54] Let's try it out.
[87:57] Here it is.
[88:04] And then you specify which of the two folders you want to do,
[88:09] and that's it.
[88:13] So
[88:21] test it.
[88:23] It's on 10.
[88:30] And
[88:32] then you you will need to build to the
[88:35] the the I think
[88:40] all I'll need to pass will be the name.
[88:43] I I don't have the exact names. I don't need to pull them, but, basically, let's see.
[88:48] Annie and the site was called. It was my site.
[88:54] Let's see.
[89:18] Oh, it requires the protocol.
[89:20] Okay.
[89:25] It's protocol now.
[89:55] Okay.
[89:57] Let me play with that a little bit. I I still have the wrong syntax. But, basically, that's all it that's all you need to do.
[90:05] Basically,
[90:06] it's under operations. You specify the
[90:10] that you want to list
[90:12] and which of the two folders because you can list both of them.
[90:16] And then all I need is to figure out the JSON over here.
[90:20] What about
[90:24] during the so the
[90:26] next user group
[90:28] Yeah. So that's what I'm going to be August 22,
[90:32] and it will be an ask any on ST best practices. So
[90:37] you know how those works. You ask whatever questions you want. So maybe this is an answer that you can start with. I will have it for them, but can you send me a quick email? I didn't catch who the question came from. I'm sorry. Send me an email. I'll get you the JSON before the twenty second, but we'll also start on the twenty second with that.
[91:00] Perfect. Was it Rosvan?
[91:03] Maybe. Rosvan, was that your question?
[91:10] For the listing, you mean? Yeah. No. Because someone. Oh, okay. Yes. Yeah. Send me an send me an email.
[91:18] I'll work out, probably the JSON about three seconds after we're done with the call. But I I know how to get it also. If I if I know where to look, it's fine. So if Oh, okay. If it's that button, I know how to look for it. No. No. Not not with the button, with the API.
[91:33] Yes. But if it's there, I can look what the API is called. You can. Okay.
[91:39] Good. Okay.
[91:40] Okay. Last question. That run a little longer than I anticipated,
[91:44] but live servers and all that fun.
[91:47] Any questions?
[91:48] Anything in as
[91:50] Nicole mentioned, we'll have another user group three weeks from now,
[91:54] which will be the standard ask Annie.
[91:58] Anyone that was on this call and had played with what we'll send, so we will send you the presentation, we will send you the recording, and I will also send the download of the two accounts
[92:11] and all of the other parameters, so you can put it onto your desk server and play with them.
[92:16] Anyone that have any questions
[92:20] after that This is your last chance. Welcome to show up on the twenty second. Yes. In the meantime,
[92:27] last questions before
[92:29] we wrap up this call. It's past seven,
[92:33] and I know it's frustrating.
[92:35] So I prefer
[92:37] not,
[92:39] unless it's one word answer.
[92:44] Okay. So let's stop
[92:46] now.
[92:47] And I thank you all for joining. It was a pretty big crowd
[92:53] today, so this is very great. And what was very good also is that you had been sharing
[93:00] your experience and answered
[93:03] one another. So
[93:05] this is, this is great.
[93:07] So thank you very much, Annie, for this great presentation.
[93:13] And, thank you all
[93:15] customers
[93:16] for joining and for having been, very active
[93:21] during, this session. Again, this is your forum, so this is, something that, we want
[93:29] for you.
[93:31] Thank you very much.
[93:33] Have a happy summer.
[93:36] Thank you, Nicole. Thank you, Annie. Thanks. Thank you, everyone. Bye. Thanks, Annie. Bye bye. Thank
[93:42] you.
[93:44] Thank you.
[93:49] Okay. Mark, Bill still here by any chance?
[93:54] Nope. Okay.