FileMaker 19 For Beginners Daily Live Training
FileMaker 19 For Beginners Daily Live Training
FileMaker 19 For Beginners Daily Live Training
Hello everyone welcome to an awesome monday my name is richard carlton i'm the creator of fmtraining. tv where we create great filemaker training on demand training courses for the filemaker platform we also do daily live streams monday through friday at this time slot happens to be one o'clock for me but for everyone else around the world welcome i really appreciate that you're here today is day one of some really exciting stuff remember i have a rule i never want to be the smartest guy in the room because that's a real bad right so today i have many people much smarter than i am they are here to help us out so we're going to be talking about filemaker in the enterprise specifically and synchronization it's not just the enterprise but really syncing between two copies of filemaker and it's kind of applicable in the enterprise a little bit clarisse for 30 years has really never positioned itself as an enterprise product or platform and what's happened
recently is that we don't hear marketing talking to us so much as we don't hear that clara saying hey we're enterprise now what ends up happening is that uh when uh the ceo at claris talks uh to folks especially in the media he starts using these words like enterprise software and so clarisse is apparently attempting to reposition itself as an enterprise product just keep in mind that that is something that they have not historically done and of course that brings up the question what is enterprise because most of us don't really build enterprise applications we build applications the biggest filemaker solutions most of the time have two or three hundred users but what if you needed thousands of users right well you know simultaneous users then really you have to have more than one server you want those people to be sharing the same data and that's really what today is all about it's about bringing filemaker to the enterprise or the enterprise to filemaker
whichever and so a basic filemaker solution by itself you know you really need to think about designing efficiently designing what we call lean design in our course curriculum i'm going to jump over here real quick to our live training schedule but what you're going to see is if i go up under here a lot of you who are familiar we have some new people here but a lot of you are familiar with nick hunter right and nick hunter has a free video course uh where we talk about uh lean design and so this is uh probably 20 hours of training here that's just i'm pretty on the high end of learning how to make your filemaker solutions work very very well right and so uh so the lean design is a big part of this but you also need some great third-party companies like 360 work basically the plan for today is to start the conversation about sync i'm going to be playing air traffic
control a little bit with this with jesse or helping him out a little bit um he and uh so let's talk about the little uh news real quick filemaker is coming up with their uh engage conference or their dev con conference they call it engage but no one knows what engages so it's still kind of defcon for those of you who wanted a devcon deal we are bringing back the two basically the the double bundle that we have so you can get two copies of our complete training which includes filemaker pro 19. includes all the stuff that we have you can access that at fm at https but is this bitly it's this fm double we'll make sure we put that into discord twitch and youtube as well this is a really great bundle you get everything for two people uh pretty exciting um so today's conversation this is jesse barnum this is the awesome guy that he is he is pretty uh neat guy he's been
doing devcon presentations for many many years we kind of bug him periodically and and present on critical technology i frequently will spend time as you folks know this i do live streams we will talk about 360 words we talk about you know nikkor at goya another mission critical company 360 works has been ultra mission critical for claris um i don't know that claris would ever tell you this but i mean like i said 360 work saved back at the time filemaker's asked back in the day uh was around filemaker 85910 time frame um digital document management never existed in filemaker until jesse brought it to the platform um and so you know once again this is another situation where claire's wants to push to the enterprise but they can't really get there unless you have a really highly efficient filemaker solution one two and you have some replication or synchronization software to synchronize the server so we'll get into that conversation just briefly
i also want to point out the fact that we were trying to decide what artwork we were going to use with jesse since he is now in the enterprise and i told him make sure that he has a for pips because he is a captain so we are going to be doing uh live streaming once again for with jesse for the next four days this is all on the same topic jesse has a propensity to just like run through this at 500 miles an hour he practices doing this presentation at speed the problem is today is more of a conversation to bring everyone up to date everyone who's not a certified ninja filemaker developer if i show you our diagram right here uh we're gonna try to include everyone if you're bunny right bunny are you there maverick de goose bunny are you there come in i don't see bunny uh so bunny may not be here yet but we have brand new developers one of which
is bunny he is in the netherlands i think is that where you're at but intermediate developers have seen your ninjas jesse if you had to rank jesse's skill set he's over here breaking off the side of the paper over here right so he tends to talk like we're all over here but i have to slow him down and with this for the entire audience so what i want to do is make sure that everyone here ask questions ask questions feel free to slow us down if we're going too fast now i want to see if jesse is there he's unmuted are you there jesse barnum everyone this is jesse he's the ceo of 360 works really awesome guy very talented individual he's talented in amazon and building products like i'm talented and putting training together in filemaker so um we together make a very dangerous team in fact this is the guy when we watch the rolling credits is hiking the fifty thousand dollars to me
right in the football we're playing fifty thousand dollars it's actually fifty thousand dollar football stack up there was a better story that happened in that downtown what was that that was remember the guy walking around with his xylophone trying to get every clear out of the vendor area and you got him in the headlock yeah oh yeah okay yeah am i am i allowed to mention that uh well yeah that was the year before so they yeah i'm kind of a marginally aggressive individual so yeah i told them so they have this poor hotel okay so one story so so they're they so what they're doing is that they're at devcon if they ever do a def con again which is a really interesting question if we're taking bets on whether that ever happens again but if it does i hope it's in nashville yeah being really nice and so uh so they though they have like everyone's having lunch and then they walk through the
guy with like some poor hotel employee gives us this like xylophone chimes with the ding dong thing on our cup or whatever and he's walking through going ding ding ding ding ding ding and that's your like subtle clue that you should head for a session and quit talking to your friends about filemaker and go watch a session on filemaker any and i'm trying to like talk to people and sell product or talk about jesse or i'm doing a live broadcast i remember what it was and and he's going and i told him i said you keep doing that i'm gonna you up right and he came right by did it again i went over and grabbed and put him in a headlock and i guess he dropped the dinger i don't remember what happened i've kind of like fuzzed that out of my mind but everybody was smiling he was smiling you were smiling i just couldn't tell if he was like a smile of fear or a
smile of like he's in on the joke i couldn't quite tell you yeah let's get on with this thing so chris so christian i keep calling you christian because we got christian olsen he's floating around here too somewhere um talk about the enterprise what is the enterprise to you jesse and what is this meersync product you want to walk us through that a little bit verbally yeah so you know i mean an enterprise is a company where you know typically large companies they're very risk-averse uh you get to be a certain size and your priorities change uh and rather than focusing necessarily on being like a nimblest and the fastest you're focused more on being the la the most reliable and uh and scalability and richard you brought up the the points um you know enterprise and scalability i think what you were talking about uh like an introduction yeah and uh it made me think about something that i hadn't planned on mentioning but but but
it really kind of draws a point of mind is that i think a lot of people in small businesses may not understand what scalability means they hear scalability they think scalability means fast you know if a customer says i want something that's scalable i think a lot of us are used to thinking oh well they wanted to be fast um and those two terms don't mean the same thing and we'll get into this more you know as we talk about some of the things that nursing can do but scale so let's say that you have two systems and system a takes uh five seconds to run a particular operation system b takes one second to run that same operation well you could accurately say that system b is faster than system a a takes five seconds b takes one second so b wins on speed now let's talk about scalability let's say that you now have a thousand users accessing those those same servers and uh system
a uh still takes five seconds each time a user does it when there's a thousand users doing it at the same time system b each user who clicks it takes a thousand seconds to get that thing done system b is faster than a but it is definitely not more scalable than a and so uh so scalability is basically the question of how adversely affected does your speed become as you increase the number of users and that's where you really get into the enterprise is it's not so much about does this button take half a second or three quarters of a second it's more like if i've got 500 people all clicking this button at the same time what happens yeah that is a really great explanation i mean i can't do any better so the idea is that it's and we talk about this with filemaker the i the idea with scalability we talk about this and i just want to frame this and then we'll go forward
is that when you're brand new to the filemaker platform you're happy to get it to work at all that makes sense you're happy to get filemaker to work at all and then as you start to roll it out with a team of people on a filemaker server the cloud five people it's great 10 people's great there's a magical number in there where the performance starts to drag because you haven't thought about performance at all and so it typically happens somewhere between 20 and 30 people maybe a little higher than 30 but well the performance will go off a cliff and that is where it's not scalable so we're trying to help you kind of grow your user base to 50 or 100 or 200 people last year we took fm starting point which is this free crm we talk about the free crm all the time i'm gonna hit the button for the uh starting point right here so this is the free starting points an updated version
this has been tested on a single pretty decently sized amazon server at 200 to 300 people it was 261 and then we did some things to try to break it and so one filemaker server can do all that right does that make sense but um what if we have a thousand people or we have people around the world well today's conversation we're going to kind of kick this over and get this going so we're going to jump over to jesse uh hang on one second i'm jumping to you jesse there we are okay uh yeah all right so go ahead jesse so uh his arm slides up right now yeah you're a lot you're yeah clear absolutely all right cool um okay so i want to talk about uh so the product that we make is called mirror sync and before i really dive into these bullet points here i just want to give a really kind of just very broad overview uh i'm sure people on this call
i know some of you by name and i know some of you are using nursing heavily so i know you guys already know this uh but i'm not sure everybody on the call understands you know what sync is or or what mirror sync is or why you would care or why you would use it so um our most common reason that people buy mirror sync is because they want to be able to take filemaker go on their either their phone or their uh ipad and go places that either they have no connection or they have a slow connection or they have an inconsistent connection um and you know all three of those are kind of different one of our one of our biggest nursing customers they are they use it for emergency medical uh uh staff paramedics firefighters that kind of stuff and uh so they might have a really good signal in the back of their ambulance as they're driving in one area but
as they start to key data in about the guy that they're rushing the hospital they drop they drop the connection between you know cell towers and so if it was if it was a normal situation where there were guests to filemaker server that would kick them out of whatever they were doing they would have to re-log back in reload all of the schema and scripts and layouts and things like that and then try to find where they were and pick back up and hopefully not lose the connection again so clearly in that case you need the ability when you're you know riding in the back of an ambulance on the way the hospital you need a way to enter data that is not going to get interrupted if your cell connection drops um if you are doing uh you know i didn't even really realize how big this industry was until people started buying mirror sync but oil pipelines uh there's a field inspections of uh like
oil rigs pipelines water uh lines gas lines and things like that and they're out in the middle of nowhere uh and so obviously they're not going to have any internet connection but they still want to be able to use filemaker to take pictures to take readings enter measurements that kind of stuff ensure compliance a really funny one is one of our customers uses uh mirrorsync and filemaker for cell phone tower construction and so by definition wherever they are there is no sell signal otherwise they wouldn't be building a cell tower there um so they need to do land surveys uh that kind of stuff and they want to use filemaker to do it so mirror sync is a really good choice for those kinds of things even if you are not out in the wild or in a remote place or driving at 70 miles an hour in traffic it's often still annoying to connect to filemaker server as a guest it's just slow in many cases
uh you know it's a you're often limited by the amount of bandwidth available to you on either your cell connection or your wi-fi connection so another uh case where people are using nursing first stuff is rental car check-ins um they some of you know some filemaker developers that i'm friends with built a vertical market app so that uh you know when someone returns a rental car they have a filemaker go application that they use to scan the odometer and that kind of stuff now they could put in wi-fi hotspots and it would probably work but they don't want to they don't want to worry about that potential point of failure they don't want to worry about all of a sudden a dozen people not being able to do their jobs if the wi-fi gets disrupted for some reason so if you're able to work on your own copy of the database that is loaded on your ipad or your phone then you know it's going
to be fast you know that nothing is going to interrupt it and you know you can go anywhere with it so that's the kind of core mission of mirror sync joe martin's answering some questions scott asked does my mirrorsync work between server versions say between like an 18 server and a 19 server can you sync between the two uh yeah uh mirrorsync 6 which is our newest and best version supports filemaker server 17 and later okay so you can sync between anything in that range you could sync between 17 18 17 19 18 and 19 whatever you want it also does support filemaker cloud the new filemaker cloud too but uh there are some there's a really awful terrible bug that i've reported to filemaker that i keep following up with them on that the data api simply doesn't work there's really not a really great easy way with a trusted third party where you can actually use the data api really really well at least that's
like if you're going to make a page there with it it ha you have to use some um some questionable libraries with question questionable lineage to make that work and so uh pending the current time if you need a php slash gateway data api like when i when i press this button here to show you the upcoming live schedule this is basically the filemaker data api these are records vertical these are records in a filemaker solution cloud cloud 2 just for the moment doesn't do this like it needs to and and so we're basically for the most part if you're doing any sort of creative syncing of any of this stuff you need to use filemaker server and most people do that with amazon with an ec2 instance on windows right or something along those lines so but that's not and actually let me jump in for one second point out everybody in this call can actually help with this if you would like to try
and help with this go on to the filemaker community search in the bugs and issues list i've reported a bug to them and it's called uh oh gosh i wish i remember the title exactly but basically if you for data api filemaker cloud intermittent you'll find the the issue that i've outlined look for you know jesse barnum and list of posters and vote for it um and if you know half of the people in this call voted it up yeah it would probably get fixed well we'll be back tomorrow we'll bring the link so any hanging things we'll bring back tomorrow uh good all right cool so uh jesse where are we going to go from here so so what i just really covered was this first bullet point here okay i was wondering because i'm typing to other people so i'm not listening to everything you're saying so you haven't gotten down here yet all right keep moving chief yep so we just covered client to server
sync that is basically client to server sync basically you got ipads you got phones it does also work with laptops which we don't get a lot of use for but it's actually really cool to be able to work on filemaker on your laptop and then sync it to a server when you have wi-fi um but that's bullet point number one let's talk about bullet point number two bullet point number two is you've got this big system built in filemaker server but you need it to uh you need to also have it running in some other system maybe it's my sequel maybe it's dynamo maybe it's salesforce mirrorsync can do all those things it can do oracle sql server red amazon redshift any jdbc compatible database it'll support all those things so we have a another very large client of ours they are they do non-emergency medical transportation and they have approximately 2 000 drivers it's kind of like uber for medical appointments and they have 2
000 drivers that go around and pick people up and take them to the doctor's appointments and it's their specialized training and equipment and oxygen tanks and stuff like that so it's not just quite as simple as uber but the point is is that they had a web interface that was built in filemaker and if you've ever tried to build a web interface in filemaker you know it works but not when you get 2000 people accessing it so what they did what we helped them do was sync that filemaker database they've got about 200 to 300 internal people working on filemaker using filemaker pro connected directly to filemaker server and then these 2000 external people that have a web interface to get to you know to pick the people up at their appointments so we help them build the synchronization that would sync the filemaker server database to a mysql database and then we built the web interface to talk to mysql uh and that has completely
solved their problems they would have half hour to an hour outages before where server needed to be restarted backups need to be reloaded it was actually hanging filemaker up from the volume it was called threads to deadlock uh and so that problem has been 100 solved by taking that load off of poor filemix server which is really not designed from the ground up to support 2000 web users and then moving that to my sequel which is designed from the ground up for that exact kind of thing um a similar kind of situation yeah similar situation i'll keep it short because it's not that different but we did a system for a large public school system and we had to have 60 000 people log into their database simultaneously and when i say simultaneously i mean like at five o'clock p. m in zero seconds on friday because everybody wanted to check and see which programs their kids had gotten accepted into and that is the time
they posted those results so basically within like a 15-minute block of time you had 30 to 60 000 people all trying to hit the website at the same time to check the site and see what their what programs their kids got accepted to and we knew that was going to kill filemaker we actually knew that's going to kill my sequel too because another company had done it in my sequel the year before and that had melted down actually three years running and had not been able to handle it so we did it with dynamodb which runs amazon. com and that's it's not even like breathing hard to handle that kind of load and we use mirrorsync to sync this filemaker database to the dynamo database and and they actually did a load test on it with 40 000 simultaneous people and they said going back to my point about scalability they said uh that the web application ran the exact same speed with one simultaneous
user and 40 000 simultaneous users uh so that's just kind of it's it's making all of this power and the scalability mirror sync is putting that within the reach of all of us and the filemaker community we don't necessarily need to switch to a new tool we just need to extend our existing tool with mirror sync to use these other capabilities absolutely absolutely yep yep yep and that that would be a good another webinar someday we could do talk about dynamodb because it's you know that's like unnatural things you do with the database it's really i'd love to dynamo is crazy i i was reading that on prime day three years ago and i'm sure it's more now three years ago on prime day they measured dynamodb as sustained just for amazon. com traffic they measured a sustained 13. 9 million requests per second 13. 9 million requests per second so on amazon. com from people shopping and dynamo handled that so whatever purposes we can
think of dynamo is going to handle it right the trick is is though that when you have to when you go into that conversation everything you think you know about databases like oh it's a relationship and it's a key yeah you have to like okay just check all that at the door because we're starting from scratch it's like re it is reimagining it and of course it has it filemaker does things it won't do right but but but uh yeah so it's the right tool for the right job and and any more about filemakers about integrating with the correct third-party tool and then of course we get to kind of our the last bullet point here which is the whole thing about multiple filemaker servers which you can use them two different ways you can put them all together so they're clustering kind of backed up ready to go or you can spread them around the world or do both right so yeah yeah and that's
going to be the focus of of this kind of four day thing is this third bullet point here um we might talk about the first one i'll leave that up to richard if he wants to talk more or do a demo or something like that but the i think the the heart of this is going to be this third bullet point so i've got a couple i've got just a few slides here to kind of graphically show what i'm talking about okay so the first one is clustering clustering means so the green box in this diagram is a particular um amazon region and i'm going to talk more everything that miracin does it can do with or without amazon but i'm going to be talking about amazon throughout this because amazon has a lot of tools that just are out of the box to do these things you could do this on your own server on your own data center in your own office you could do it on
azure you could do with google you could do with any service provider you want i'm most familiar with amazon and so i'll use amazon terminology as i'm going throughout this to talk about this but the green box here represents a single amazon region it could be you know the northern california region could be the eastern uh the the northern virginia region which is close to me uh you know sweden let me interrupt so let me we talked about this in the last week or two but basically a region is basically like a town or a small area and what amazon does is they set up different data centers in different buildings around that town for example right that makes sense so yeah they've got 22 around the world yeah they've got 20 regions around the world and within and within each region is multiple buildings with multiple data centers where that data is sinking and replicating on their own throughout that that well yeah so
the idea is that if uh one of the buildings gets hit by an asteroid or something like that or burns to the ground really you you have that they provide their own level of redundancy right so yeah they call it an availability zone so you know for instance their ohio region that the region is ohio and it's kind of designed to serve kind of the northeast united states and then they'll have three availability zones which is three separate data centers that are kind of like triangularly in this kind of north southwest and southeast suburbs of uh i think it's cleveland i don't remember the exact city um but but yeah they'll have an availability zones all have direct pipes to each other so they have you know under a millisecond ping times between those availability zones uh and it's probably a good idea so the green bar the green box i have here to be clear does not represent an availability zone it represents a
region so this would be the ohio region or the northern california region and then your servers could be in the same availability zone or better yet different availability zones so like richard said if a backhoe like destroys their internet connection and it actually that's not realistic because they have multiple but if something if if a fire destroyed their availability zone they would still be your other server running in the other data center on the other side of town but so the clustering concept the goal of clustering is to simply increase scalability and reliability remember scalability is not equal speed speed is not going to be affected by clustering but scalability will be so if you have two filemaker servers that are syncing with each other and they're sharing data with each other and then you have a bunch of users connecting to those servers if you split it so that those users are now half of them are going onto one server and half of
them are going on to the other server now you've put instead of 300 users on server a you've got 100 you've got 150 on a and 150 on b so you're decreasing the amount of load on each server which is increasing your scalability it's also increasing your reliability because if server a goes down or server b goes down the load balancer was that orange box in the middle will shut shuttle all the traffic over to one the load balancer is the part that knows which of those servers are up and running so the users don't connect to filemaker server a and they don't connect the filemaker server b they connect to just filemaker server which is configured in the dns settings to point to the load balancer the load balancer looks feels and sounds quacks just like filemaker server but it's actually passing all the traffic through some of it to server a and some of it to server b and it makes sure that
if a user was connected to a then all of that user's traffic always goes to a and if a different user was connected to b all of their traffic always goes to server b so the user can't really tell that there are multiple filemaker servers there's not two different host names or two different addresses for them to connect to they just it looks to them like it's they can't tell the difference between whether they're connecting to a filemaker server or to a load balancer that's kind of secretly splitting that traffic to two or three or eight or however many servers you want to set up and then you've got mirror sync which is you know the black arrow between those two servers making sure that they stay in sync so that you know if one user adds a record to server a within a few seconds it'll also appear on server b so that the other half of the users who are connected to server b will
see that new record so it's a it's an inexpensive fairly simple setup to do the load balancer costs like 15 dollars a month from amazon the filemaker server i'm going to jump ahead in my slides for one second and show you this look at my highlighted text here this is copied and pasted directly from the license text in in filemaker each filemaker user licensing contract includes three licenses of filemaker server so you're not having to pay for additional filemaker server licenses to do any of the configurations i'm going to be showing you that uh showing you today okay so from a cost standpoint your costs are 15 an hour for the load balancer plus one extra piece of hardware plus a license for embarrassing okay so i'm going to uh just jump in here because we do have comments coming in so once again welcome to uh stephen dolinsky i think is here lexi just said this for what it's worth the last part of
your third bullet is my current use case scenario clients using third-party hosting based in california we are now adding a server with the same provider in hong kong for asia's client user beca because access for them across the uh to the us from hong kong is unreliable and unacceptably slow that's the whole point of this enterprise conversation right so alexia listen lexie could give this presentation i know how she is she knows all this already but the short version we're going to show how easy this is and the tools that we use to make this work so for for those of you wondering we are going to be showing i got to see the auto load balancer earlier and it just made me all hot and and i was excited i was really hot and bothered it was like this is very nice so uh yeah so so you'll get to get but over the next but one you know we can't cram it into one hour
so we'll go for another 20 minutes or so once again this is going to be recorded we will uh you know clean it up a little bit as we go but yeah so what's next jesse uh so the next so that's clustering let's talk about warm backups form backups are not super sexy but they're very easy to set up and uh and they can be invaluable so a warm backup notice i'm going to jump back to the previous slide notice my arrow in the clustering sync is double ended it's a bi-directional sync everything you write to server a gets written server b everything you write server b gets written to server a server a and b always match and you can write to both the warm backup scenario is very similar but you remove the load balancer and you make the sync one way everybody connects to filemaker server one and then mirrorsync is just running in the background making sure that there is an extra
server that nobody connects to that exactly matches server one the benefits of this are pretty obvious if server one goes down if anything happens to it instead of hours to try to figure out what was going on and recover it and recover backups and maybe get off-site backups or whatever you just simply switch your dns server or your router or whatever your elastic ip address to server b and now all the stuff that was on server a you know if you're running your sync every 30 seconds then server b is going to have everything within 0 to 30 seconds of whenever server a went down it's much more effective than um you know filemaker backups that you run every hour or two or ten it's it's much more live uh there really is not much of a performance hit when that server is running when when the sync is running in the background so it allows you to basically just have an extra server that stays very up
to date with your primary server it doesn't have the some of the complexities of the load balancer sorry clustering thing that we talked about before okay yeah we got a problem here major malfunction nathan schneider just uh let the secret out uh nathan schneider says on discord i guess jesse didn't get the memo very important memo that richard's clients don't do backups i tell you and i rail on that topic i rail because there are people who just they they argue with me i got we we got fired by a customer because they refused to do backups and we told them that it was going to catch fire and melt to the ground actually nick hunter told him this nice people but they told him this and said it's gonna day it's gonna come with this gonna happen you're not gonna be and this is not gonna work anymore he goes we don't like nathan he threatened or uh nick we don't like nick because he threatened
us i'm like he didn't threaten he just told you what's gonna happen when he catches fire right so um they were like offended that oh you're threatening us like uh okay never mind it's like it's like arguing with uh mother nature and physics right and you can argue with physics to a point but then you get burned you get burned to the ground right physics doesn't really care if you argue with it yeah yeah somebody doesn't mind if you agree with it or not yeah well you know i rented that once in a while so yeah so it's uh anyways yeah thanks uh ken on these guys all right so okay so okay just just so that so this is this one but i mean wouldn't you just use the other one go back one slide when you use the previous one because then it would like double it it would do the same thing for you right kind of yeah so the warm back up uh i
would say in general pros and cons wise i think there's more pros in this scenario than this scenario really the advantage of well in the land clustering maybe i should i think this land clustering it gives you uh first of all it splits your load in half so it increases your scalability whereas the warm backup does not increase your scalability the clustering uh gives you kind of a live failover in case anything goes wrong whereas the warm backup usually would be something that an administrator would need to go and flip a switch to switch him over so i think in for the most part there's advantages to the clustering that you don't have with the warm background right the advantage of the warm backup is that psychologically it's much closer to what we're used to as traditional filemaker developers so uh you know the idea of kind of this third-party proxy you know magically shuttling people to different servers can be very bizarre
dude just stop just stop just stop i'm going to veto this right so claire's claire's tried this claire's tried to do this right and yeah the standby server for those you're wondering about that and um i'm not sure what i should say about that i'm not supposed to like you know be overly i i the thing is with this is like it's a mental thing so it's like you're saying if we're too old like that's like uh i'm old and i'm in a wheelchair and i'm in my walker i don't want to use a new one cars it's too fancy for me right right right what i don't understand this makes it listen they tried this right in order for you to switch so these people would all so so this server would blow up and i don't know lexi was here for this or not this server would blow up and then what would happen these people all go hey our stuff is offline and we're
unemployed and so then you have to find the it guy a little guy stick figure over here right and he has to go over here and come over here and do command line c slash backslash slash and do like four or five lines of backs and then if he if you can find the guy one he's not playing some game online or two you can get him to log on then he can do a bunch of command line stuff and switch the line so they go over this way whereas this previous slide would have done it for you automatically right so yeah um it's like an automatic hostage negotiation with the i. t guy assuming that your it people are friendly and happy right right i think that the warm backup is a good transition i think that this is a very low risk quick win that you can set up immediately you can show your client hey look this is working then you can kind of go to our
clustering scenario where you say how about if we make the sink bi-directional and we introduce a load balancer so i would say do this as a baby step do the warm backup as a baby step and then progress to the clustering all right if you say so chief we just i think we just get me just skip to the end and get to the one you want right so which is the whole point of what we're doing today right so i'm not i'm not trying to pedal i'm not trying to pedal dog doo doo here yeah here we go baby let's go so this now let me point out uh before i really get to talking about this the land clustering option that i talked about before it's really important to understand that is for people that are in a single geographic region you know if all of your users are in one office or one state or one city or one pretty much relatively small area then the
clustering option works really well but it doesn't work that well when your users are in different continents or countries uh from each other so then rather than doing like a random distribution where you know kind of a round robin you get a server you get that server you get the server you get that server it becomes a more geographic distribution basically everybody in this geographic area goes to this server everybody in that geographic area goes to that server and that way everybody has a reasonably fast connection and that's that's what we've got set up as a demo that's we're going to be showing you guys um and the idea here notice that there's no load balancer i'm going to jump back for two slides our orange box in this case is a load balancer so people don't connect directly to filemaker server they connect to the load balancer the load balancer kind of magically talks to filemaker server for them in this case instead of a
load balancer you have a dns lookup and so i i've got some after these slides if i have time i'm going to jump to my browser i'm going to show you guys how i've configured the dns in order to enable this using a feature of amazon called latency based routing and latency based routing is all it simply means is that you will when you say what is the ip address of the server called fmsp. 360works. com amazon won't just spit out an answer it'll say let me see where you are now let me see where all the available servers are and now let me give you the ip address of whichever server is a closest to you and b up and running and so um yeah so you will always go to the closest and fastest server that is up and running the up and running is a is an important part because if in this example if the united states server goes down amazon will detect that
and the next time that a united states person says where is my closest server amazon won't send them to the dead server in the u. s it will send them to the to the live server in europe that might be slow but at least it's running right and and that's the good part because in the event of a failure what happens is people get routed to the next closest server which might be not obviously as fast but it's better than zero right so yeah which is useful so yeah so if people are bobbing up and down alexis agreeing with us about the warm feature et cetera et cetera yeah so um so i mean so what we got is about say we have about 15 minutes uh we're gonna go about ten after then we're gonna roll it for today but i do want to at some point today we're not going to walk all the way through to the end of have you know setting this up
i would kind of like to show it though right the end product so people know that you know that is kind of working right even if we just log on to it so yeah okay so before i get into that uh i'll since i mentioned the wrap 53 thing i'll i'll spend five minutes or less and just show that okay um so my route 53 console here is so route 53 is amazon's name for dns and it's kind of a play on words because dns runs on port 53 uh and so that's that's where amazon came up with the name and so what i've done is i have created uh four health checks over my dashboard here i'm in the health check section and i've got four copies of filemaker server running which richards richard staff worked kind of to set up for me um and we've got one in northern california one in northern virginia one in ireland and one in mumbai so that's one two three
and four and so what i've done is i've configured amazon to check uh this this server by going to this url and making sure that url works hang on here one more second hang on because i'm scroung i'm kind of zooming in so people can see this right so all right so there you go so that's url okay good so that is basically making a call to the data api saying what version of filemaker server are you i could come up with something a little bit more comprehensive to make sure the database did that kind of stuff but for demo purposes this is sufficient it just at least tells us that filemaker server is running and the data api is running um and so in fact here just for fun let's go to ireland and let's turn off their data api and this will take a little bit because the checks happen every 30 seconds and there needs to be a certain number of checks that fail before it
will really consider it to dead but if we check back later then i think what we'll notice is that filemaker server three will show us unhealthy let's pause that well there's pause for a second so just so everyone's up to speed on this once again i'm not talking the dalinskis and folgers and the tk55678s um for everyone else so you can log on to amazon when we say amazon this is not the shopping part of amazon this is the backend data center technical stuff and you can run your own virtual server here in fact jesse can you show us the current home screen for them because they revised that on a year or two again right they changed the uh to this uh so basically for those of you who've never seen this before just as a basic sort of thing uh yeah back yeah all the way out to the super that one okay so scroll the top so this is when you log on to amazon
and you have your own server these are all the really cool things that it can do down here and i know it's kind of chunky and it's kind of hard to see but i mean if you log on you can see you know every one of these if you're an i. t person is really interesting okay hang on jesse quit moving because when you move then it's chunking here so so you got easy so much yeah i know there's ec2 where we build our virtual filemaker server right we have the s3 is where we have our big you know hard drive in the sky for storage right then there's s3 glacier which is super cheap but kind of goofy uh slower back storage if you have like you know many many many terabytes to store on an ongoing basis i mean all sorts of stuff are in here crazy and then you saw the at the top was route 53 so this is basically a a computer developer
slash programmers you know most funnest location ever because there's always something new to learn in here because they're always adding new things so if you go back up a little bit jesse for me if you go back up uh assuming he's there hello jesse there we go yeah so you see yeah you see the route 53 so this is a separate service separate than the verdict virtual servers right so route53 is our checking the health check so if you want to go back that it should be updated now yeah so let's take a look here oh man oh there it goes unhealthy and if i go here and i go to monitoring here's here's all those checks that worked and then here's where the checks started to fail yeah and so what'll happen is if somebody in europe were to try to go to the server right now amazon would exclude ireland from the check and it would send them to whichever the server was closest
whichever service closest either mumbai northern virginia or west coast and i don't know which one would be but amazon would figure that out and send you to the right place right and i mean if you're really serious about this you'd have a server in germany and a server in ireland and you could put them all over the place right so there's not really a limit here um cool all right so that part of it and then the last last ingredient here is so these are the health checks this is uh here i've got to open this other tab right here these are the health checks which i've turned this back on so before long it'll come back and say healthy and then once you've configured the health checks then you configure the actual dns entries and so i've created an entry called fmspdemo. 360works. com and um and you know i'm not gonna i'm just gonna talk really quickly about this and uh we can watch the
video if we want but basically you set up a c name uh which is kind of an alias to some other server i like to set a really short time to live so that i can change it anytime i want and have it be cached for a long time then which real server it's going to so the fmspdemo. 360works. com uh for the us east will go to this is the actual host name here of the real server and then this routing policy the the norm case is simple simple just means this server goes to that address but latency means i know this server is running in this region and this is the health check that goes with it and so find you know which of these four things are in the region closest to where the user is that's how you that's how you can set up latency based dnx okay and that's the key to making this kind of work smoothly for users no matter
where they are all right so i'm going to repeat these questions and let you handle them uh uh jesse so this is from muhammad um and he says hello richard on fmp if you uh on filemaker i guess filemaker platform if hosted on amazon what are the hidden costs as a customer usually have to pay for extra bandwidth consumption right i mean the answer is bandwidth out you definitely pay for bandwidth in not so much right isn't that how that works yeah bandwidth and is free bandwidth out is nine cents per gigabyte you have to accumulate a very large amount of traffic to have a significant amount of uh uh fees from data out okay and then i mean if you're doing 20 gigs a month which is a whole lot that's a dollar 80 per month yeah um so muhammad also had another question here which potentially concerns me if i'm understand i'm going to discuss i'm going to tell you well the your redraw there
needs to catch up uh so he asked the question yeah we'd have to let the screen redraw it's lost its mind he asked the question also can we host fmp apps on a shared cloud storage and have server redundancy question mark i'm not sure i understand the question exactly um shared shared cloud storage sounds like s3 or something right right and no you cannot you can put a filemaker database on s3 you can absolutely do that and anybody who wants to can download that database and work on their own copy the database that works too uh what you can't do is you're not going to be able to merge that data together any simple way mirrorsync won't do it filemaker server is not even involved so you would only do shared storage if you want each user to have their own separate copy that's their own local copy that doesn't sync anything if you wanted that data to sync to something then you would need to have
filemaker server involved and you would need to have mirrorsync involved uh and then at that point it's not clear that there would be any advantage to even having the file filemaker databases located on s3 to stick with uh you could you could put them anywhere you could put in dropbox or s3 or just download it directly from filemaker server but but you would need mirrorsync and filemaker server if you wanted those offline copies to be able to share data with each other yeah i think i think i can't tell but sometimes these conversations are conversations about how do i use this without using filemaker server and the answer is you if you if you just want to put it in like storage like you're going to still store a piece of meat in the freezer or if you don't like meat whatever you want to put in the freezer but the point is is that it's not really easily edible in the freezer but that's kind of
what s3 is or this cloud storage is if you want to have people actually accessing it and actively using it it has to be on filemaker server and so the conversation is about not one filemaker server but three filemaker servers or four filemaker servers and and if you buy a license say like you buy a 10 pack or 20 pack or 30 pack of filemaker you're going to get three servers that you can use right most people just use one but you actually get three set up in different regions so i we're back this conversation how do i share with the world without paying for it i think is i'm guessing that's kind of the the question right so this i mean we're in an area here where not only do you have to buy filemaker server you have to buy mirrorsync and mirrorsync is not free but i think it's worth its weight in gold right um absolutely without a question so so uh we're coming
up at five o'clock i know richard you wanted to do a quick demo did we want to have people connect to the server or today not today we're gonna invite everyone in uh over the next couple days at some point we're gonna we've got more awesomeness we haven't even gone through the filemaker file or any of that kind of stuff or kind of what we've got cooked up but why can you open why don't you open why don't you open it on your end and i can open one on my end that'll be kind of funny so i need to go at a host so how can i add a host people are going to be like hey richard we're going to do this with you show host what'd you say the host name was for that that fmspdemo. 360works. com fmsp demo fmsp demo dot 360 word fmsp notice we're going to save okay got the ssl okay i'm going to open this up now when i'm going to
open this up we're not going to share the password everyone just yet um but that will be coming so then now jesse opened it up on your it's the title my end it it flashed during the start but it's in northern virginia okay mine says northern california right here so you see that everybody i'm in atlanta georgia okay so he's here so he's on the east coast on the west side that's why sometimes the screen is a little chunky now let's go to contacts real quick each of us go to contacts and what i'm going to do is i'm going to uh go to i'm going to create it's kind of chunking a little bit but we're trying this hang on everyone stand by uh i'm going to uh what we're going to do is i'm going to create a or do you want to create a record jason you want to put a record on your end and then do you have a picture no like create
one for alexis i don't have a picture of alexi though so i don't know what you're going to do there i've gone either handy okay okay now i got record six now the automatic syncing is going on it's replicating so we're waiting for that to happen oh record seven is here now so that is two different copies of fm starting point enterprise edition the enterprise one that we have um and it synced and we haven't even got the conversation whether we're both working on alexi at the same time that's a conversation for another day that we are going to you know we got three more days of this but we want to show this now what if i yeah what's important what's it what's the important summarize what just happened what because we're all used to the idea that okay richard changes something and i said i changed something and richard sees it so what's cool about what we just did why is this interesting the cool and interesting
thing is i changed it on a different server than him i changed my i changed i made my change on a server running in virginia richard made his change on the server running in california and then we both saw each other's changes because mirrorsync is running in the background bringing both of those servers up to date with each other so that's that's the kind of the essence of what we're trying to do here i just did it again that was about a second and a half delay that's what we're talking about now imagine this thing replicates now instead of just the east coast of the united states the west coast united states but there are the servers also in mumbai india and also in ireland and then we might even put one together for australia if we want to throw one together we'll have to talk about that so stephen delinski has a question if you have a dev server and use dmt which is data
migration tool or deploy or auto i'm sorry we can't say the word auto in this broadcast what is the workflow with databases in different locations to ensure that they are synced uh with the latest version stephen dielinski i am sorry to say that you have to come back tomorrow for that actually that's not true we are going to talk about that that's a big part of it steven's like oh i just wanted everything in one hour most people are not as smart as steven dolinsky here right so i got to bring everyone along so the question is if you change a file in location a how do we replicate that everywhere right we are going to do that it's going to be awesome it's very doable it's very simple simple right you just you want to walk do a verbal real quick on that don't do it because we have to have a reason people come back tomorrow right so well it's uh basically you change one
field in the nursing so the first before i even answer the question i think the first i would start with a question of my own which is if you do have like a testing server and maybe a a deployment server what is your goal with the data do you want the data to be the same in both of those or do you want the testing server to have a testing copy the data and a and this and the deployment server to have a deployment copy the database if you want the data to be different you know if you want testing data testing server and live data on the production server then mirror sync doesn't really even enter into the equation you would use one of the tools you mentioned i would highly recommend our product 360 deploy which does that very well um yeah and and so you wouldn't need mirrorsync if you didn't want the data to be the same on both of those servers but let's
say that you had uh like a staging server and a production server and you wanted those to be identical to each other then uh you would do that by going into you would get your everything set to go on the staging server you would have a sync running so the staging server has the same data as the production server and then when you're ready to deploy it to the production server you would just go into the mirror sync table and there's a field with a calculation in there you would change that calculation mirrorsync would see that the calculation changed and on the next sync it would take a full copy of that database from the staging server and move it with all of its data onto the deployment server so yeah so it's a question of whether you'd use deploy or um i don't know i mean i'd be in yeah yeah because because really because because deploy is where you're going to get that data migration
tool for those of you who are new we have a whole set of training so here comes as we wrap up the broadcast here for those of alex you haven't seen this before this is kind of funny so this is where we have to put the the shameless plug in here so you're everyone ready for this all right ready all right this broadcast is brought to you by fmtraining. tv bringing you the greatest in filemaker on demand video training visit fm trading tv and right and then we have bundles here and if you're looking for a double double bundle you can do that if you're looking for jessie's uh awesomeness where is the so if you want the jesse 15 off mirror sink deal that expires on june 26 it turns to a pumpkin uh that's the deal you want um highly recommend it so uh but yeah i mean we for a lot of the awesome filemaker people here we do a lot of training a lot
of it's kind of more beginning intermediate some super advanced stuff this uh i think uh as we go along here from day to day we're going to keep cracking open the level difficulty let jesse get into the code talk about what happens if we both edit alexey at the same time and then she hits in between what happens to alexi does she die a horrible death how is that handled right those are all great questions um and to be honest with you we have this crm called fm starting point we've been giving away and we actually went through a process to because we use it a lot for training but it's really been lean designed by nick hunter there's you know nick hunter and it's got a great interface but it's been really optimized for high performance right we know it'll do 260 300 400 people on a single filemaker server so we took it we adjusted it so it works great with mirrorsync and so
that is kind of like the default demo go to crm because we also know it'll deploy around the world it's already working around the world right now we just only have so much time to show you so uh so yeah so if you folks are around once again if you can't make tomorrow it will be recorded but if you're here in person you can ask questions which is really awesome alexi goes paying full price for mirsynk is way cheaper than roll your own yeah i had an engineer and to lexi's comment i was i was seeing if she see this is why alexi was so great god i missed seeing her around so i had an insurer come to me and he was like i'm going to roll my own sink i'm not going to play that jesse stuff or whatever it was it goes inc or whatever the price think was and he goes i'll rule my own and i'm like okay i'm like okay but good
luck with that and so like two months later comes back i'll never do that again that was the worst it was so hard and i'm talking about a guy who's really quite talented but his ego got ahead of the equation so um yeah so uh yeah so definitely if you can use jesse barnum's brain which is much better than most of ours do it right and then stephen dalinski fmp natively isn't acid yeah so yeah i didn't want to get into acid once again steven because yeah the um acid is like the super redundancy availability kind of capability i think if you have vigorous backups and you have a uh and you're using like mirrorsync to like either replicate locally or around the world um i think that the question is the question whether fmsp is acid no he's just making a statement steve was just saying yeah it isn't natively acid so tools like this get us further down the road which is absolutely i'm just
agreeing so yeah yeah i mean right so that's the thing right so i mean the idea is that having multiple servers and then one server can back up for another server it doesn't really solve the transactional kind of thing but if you're a talented developer you can do kind of transactional kind of stuff with filemaker well the fundamental problem with trying to do distributed transactional databases the idea that uh you know if you commit it on one server you know that it's been committed on all the servers around the world that sounds like a great idea but the problem is it thwarts the whole idea of why you're doing the first place which is to improve performance and scalability because what that means is that if you're if you're going to a server that's nice and close and convenient to you and you commit a record now that server can't return a successful result to you until it's talked to the other three servers all around
the world and make sure that people committed it too so you wind up working at whatever the slowest speed is of the connection between all the different servers uh and so you know mirrorsync and was and you'll see that in other database relational systems where they'll talk about synchronous transactions meaning that uh synchronous mainly all happen at the same time meaning that it happens that whatever the slowest transaction is that completes i think in general we want this to be asynchronous we want it to be non-transactional because we want to be able to know that we're working at a fast speed to a local server instead of waiting for the latency all around the world we just yeah we just need a plan to triage the uh the coalition the double alexi collision thing so yes which we'll talk about tomorrow we're going to talk about that tomorrow so everyone i want to appreciate you coming to fm training. tv once again don't forget to keep
an eye on this if you want to sign up and get a reminder click on that down there you'll be talking to your our filemaker server and it will load you into a database and it'll send you a reminder which is useful you can watch it on twitch on youtube or get the discord invite which is up here and uh and then you can participate and lexi and stephen and folks over here um definitely wanna if you wanna check that out feel free um because it's kind of a persistent conversation it's ongoing um we're kind of building a community here of people who are trying to learn so um because that's why i do training that's pretty much what i do all the time so anyway uh uh dulinsky had one last question can uh meersync potentially do be an audit log um we don't really sell it that way it does actually keep an internal audit log and it will record a log file record into
a text file every time that it writes a change from one server to another it's not really it it's not really written in a way that's designed to be like easily pulled back into filemaker or queried it's more for a human being to be able to go back later and say why did this record change from this to that the other there's a big limitation of it for audit logging purposes which is the way we do deletion scanning and deletion detection and that's kind of maybe a little bit too long of an answer for right now but the really short version of it is mirrorsync doesn't know who deleted a record or when they deleted it it only knows that a record was deleted uh and so a lot of why you want to do audit logging is to know who and when not just whether so it has limitations for purposes of audit logging all right everyone i appreciate it we'll be back tomorrow one o'clock
at the same time same place same channel one thanks to ken nathan schneider dave one dave two lexi stephen uh dennis michael all the people from around the world literally a worldwide audience here once again this is recorded you'll be able to catch the recording a little bit cleaned up it'll be on youtube etc um but in the meantime we'll catch you tomorrow and for now jesse and i and the folks at 360 works see you later see you guys thanks richard bye so so and the guys just stepped up the whole way calm cool collected the quarterback great read good patience more importantly great job up front protecting this quarterback to give you a chance and that's all you can ask for trying to rally down 10. 9 25 to go here in the fourth short motion by amendola from the left brady takes the shotgun step stands in throws it left ramada reaches up and snacks a high throw and lands inside the tent rolling with an eye bracelet behind him but danny makes the grip
Daily Open Q&A with FileMaker Experts. Questions about Installation, Upgrading, Purchasing or How to make FileMaker work for you in your company to increase ROI.
Get up to speed with the FileMaker Video Training Course!
Top Rated Course by FileMaker Expert, Richard Carlton.
http://fmtraining.tv/fmpro19.php
Experience Richard's dynamic and exciting teaching format, while learning both basic, intermediate, and advanced FileMaker development skills. With 30 years of FileMaker experience and a long time speaker at FileMaker's Developer Conference, Richard will teach you all the ins and outs of building FileMaker Solutions.
The course is 60 hours of video content! Transform your business with the FileMaker Platform
Richard has been involved with the FileMaker platform since 1990 and has grown RCC into one of the largest top tier FileMaker consultancies worldwide.
Richard works closely with RCC's staff: a team of 30 FileMaker developers and supporting web designers. He has offices in California, Nevada, and Texas.
Richard has been a frequent speaker at the FileMaker Developers Conference on a variety of topics involving FileMaker for Startups and Entrepreneurs, and client-server integration.
Richard is the Product Manager for FM Starting Point, the popular and most downloaded free FileMaker CRM Starter Solution.
Looking for FM Starting Point free software download: http://www.fmstartingpoint.com
2122 9th St., Suite 102
Los Osos, CA 93402
(805) 946-6552
FileMaker Pro is a simply powerful software used to create custom apps that work seamlessly across iPad, iPhone, Windows, Mac, and the web
Free FileMaker Training Videos Channel https://www.youtube.com/user/FileMakerVideos
FileMaker 19 Video Training Course Introduction-FileMaker 19 News-Learn FileMaker From The Experts
FileMaker 19 New Release - Top 10 Need To Know Items in FileMaker 19 - FileMaker Top Ten News
https://youtube.com/watch?v=kRDXfpMrjvM
FileMaker Purchasing Options-FileMaker Training-Best Way To Purchase FileMaker 19-FileMaker 19 Video
https://www.youtube.com/watch?v=LlZqf_8N1Ho
https://www.youtube.com/watch?v=LlZqf_8N1Hou0026list=PLjTvUZtwtgBTCbN3Sr7yn6weIyIofDoMLu0026index=3
A database management system (DBMS) is a computer software application that interacts with the user, other applications, and the database itself to capture and analyze data
Video introduction to iOS App Training https://www.youtube.com/watch?v=cVxQe_yAshw
Free FileMaker videos check out ...http://www.filemakervideos.com
Download the FileMaker Pro and FileMaker GO for mobile devices training videos at http://www.fmtraining.tv
Download FileMaker Go video training at https://fmtraining.tv/fmgo19.php
Download FileMaker Full Video Training Bundle at https://fmtraining.tv/subscription.php
Learn how to use FileMaker to create an app with the FileMaker Training Series
Comment, Like and Share All of Our Videos.
Feel Free to Embed any of Our Videos on Your Blog or Website.
Follow Us on Your Favorite Social Media
https://www.facebook.com/FileMakerVideos
https://twitter.com/filemakervideos
Join us for live training on Twitch daily at 1pm Pacific
https://www.twitch.tv/fmtraining
https://www.youtube.com/user/FileMakerVideos
https://www.filemakervideos.com/
https://fmstartingpoint.com/downloadform3.php
Download FM Starting Point
https://filemakerfree.com/signup.html
FileMaker 17 manual For Novises kindle
https://rcconsulting.com/courses.html
https://www.guidetofilemaker.com/
Blogger sites
https://fmtrainingtv.blogspot.com/
WordPress sites