I'm not familiar with Servers, other than I did have one at one time, I used it like a file server and with a small database.
But my question is unrelated to that usage.
Take a company that has 5 servers hosting a single but complex program to 1000 different interactive users.
The users have to log-in to access this program, and they can work on each others specific files, but not all of them.
My question is basically this.
How can the company only have 2 of those servers running, unless more than let's say 300 are logged on, so they bring the 3rd server on-line to prevent a slowdown. Perhaps server #5 is used for log-in data, but can also be kicked in if the loads that day are super heavy.
In layman's terms, how is this done? Assume I know nothing about servers.
Servers, Multiple
Forum rules
Post a new topic to begin a chat.
Any topic is acceptable, and topic drift is permissible.
Post a new topic to begin a chat.
Any topic is acceptable, and topic drift is permissible.
Re: Servers, Multiple
I'll start this with a disclaimer. I know some fundamental things about servers and have worked as first line administrator for such things. That gave me a very superficial knowledge of how things work, and since I retired I forgot most of it. But, that won't stop me from giving you my best guess to answer your question. 
First of all I will correct you and say, as far as I know you are operating a server in your office. It's called a Network Accessible Storage (NAS). You may be the only person using that NAS, but it doesn't have to be that way. It can be set up for many users to use it, remotely and/or locally.
Keep in mind that a server is built pretty much like your desktop computer in that it has all the same hardware. A pure server usually does not need graphics or audio and is accessed by using a remote terminal, or what we commonly call a client. We can imagine the server of your question being stand alone and running software that is interactive for any and all the clients. Managing how that interaction takes place is something the server resident program should be designed to do. Essentially that is what is going on inside your NAS.
Multiple clients, in your example 1000 of them, can be connected to the server via something akin to multi-pole (KVM?) switches in the analog world. Imagine something like a router that has multiple Ethernet input ports (we will ignore WiFi for this discussion). Each input port can be connected to a single computer. All those computers connected to the router would be called a Local Area Network, or LAN. Your router is also connected to a modem which in turn connects to "The Internet" via network switches over at your ISP's office. That side of the router is called a Wide Area Network, or WAN. Now imagine your router being connected to that interactive server instead of to the WAN. That would be one way to connect multiple clients to a single server. In this case the router acts as a multi-pole switch.
Now, instead of the router, imagine a computer taking it's place. That computer could be what is known as a terminal server. It does all the same things that the router would do, but it does it better AND can handle, say, 200 clients. If you have five terminal servers, each with 200 clients being managed, then you have attained 1000 client/users being connected to a single server. Schematically it looks like a pyramid with the server resting on the top and all the clients forming a tree with branches for the rest of the pyramid. It might look like a family tree in fact. LOL
Unfortunately, if all 1000 of those clients tried to connect to the server at the same time, that single server may not have enough processing power to accommodate everybody. Hackers attempting a DDoS (Directed Denial of Service) attack count on this overloaded server crashing to do their dirty work. A company with one server and 1000 clients cannot tolerate such a crash, of course. Thus they limit the amount of connect time each client has. The amount of time depends on the number of clients connecting at any given moment. A good design in your example would allow all five terminal servers to be active and serve all 1000 clients simultaneously. Things may start to lag at that level of traffic so that one or more of those terminal servers (200 clients each) might be turned off for some period of time. And, just to complicate the image even further, all this turning off and on of terminal servers is done by yet another computer, probably a domain controller.
You, the client, only see the server and the documents you are working on. However, your workstation could be connected to a terminal server (network switch) which in turn is connected to a domain controller (another switch) which in turn is connected to that server you are working on. if all those 1000 clients happen to be remotely located then you can toss in a Domain Name Server (DNS) between your workstation and the terminal server. You need the DNS server to find the IP address of those terminal server switches.
in other words ...
First of all I will correct you and say, as far as I know you are operating a server in your office. It's called a Network Accessible Storage (NAS). You may be the only person using that NAS, but it doesn't have to be that way. It can be set up for many users to use it, remotely and/or locally.
Keep in mind that a server is built pretty much like your desktop computer in that it has all the same hardware. A pure server usually does not need graphics or audio and is accessed by using a remote terminal, or what we commonly call a client. We can imagine the server of your question being stand alone and running software that is interactive for any and all the clients. Managing how that interaction takes place is something the server resident program should be designed to do. Essentially that is what is going on inside your NAS.
Multiple clients, in your example 1000 of them, can be connected to the server via something akin to multi-pole (KVM?) switches in the analog world. Imagine something like a router that has multiple Ethernet input ports (we will ignore WiFi for this discussion). Each input port can be connected to a single computer. All those computers connected to the router would be called a Local Area Network, or LAN. Your router is also connected to a modem which in turn connects to "The Internet" via network switches over at your ISP's office. That side of the router is called a Wide Area Network, or WAN. Now imagine your router being connected to that interactive server instead of to the WAN. That would be one way to connect multiple clients to a single server. In this case the router acts as a multi-pole switch.
Now, instead of the router, imagine a computer taking it's place. That computer could be what is known as a terminal server. It does all the same things that the router would do, but it does it better AND can handle, say, 200 clients. If you have five terminal servers, each with 200 clients being managed, then you have attained 1000 client/users being connected to a single server. Schematically it looks like a pyramid with the server resting on the top and all the clients forming a tree with branches for the rest of the pyramid. It might look like a family tree in fact. LOL
Unfortunately, if all 1000 of those clients tried to connect to the server at the same time, that single server may not have enough processing power to accommodate everybody. Hackers attempting a DDoS (Directed Denial of Service) attack count on this overloaded server crashing to do their dirty work. A company with one server and 1000 clients cannot tolerate such a crash, of course. Thus they limit the amount of connect time each client has. The amount of time depends on the number of clients connecting at any given moment. A good design in your example would allow all five terminal servers to be active and serve all 1000 clients simultaneously. Things may start to lag at that level of traffic so that one or more of those terminal servers (200 clients each) might be turned off for some period of time. And, just to complicate the image even further, all this turning off and on of terminal servers is done by yet another computer, probably a domain controller.
You, the client, only see the server and the documents you are working on. However, your workstation could be connected to a terminal server (network switch) which in turn is connected to a domain controller (another switch) which in turn is connected to that server you are working on. if all those 1000 clients happen to be remotely located then you can toss in a Domain Name Server (DNS) between your workstation and the terminal server. You need the DNS server to find the IP address of those terminal server switches.
in other words ...
- Workstation client ==> DNS server (for remote connections) ==> Terminal Server ==> Domain Controller ==> Server
Re: Servers, Multiple
No, not confusing at all. I've also been doing some reading up on how multiple servers can handle a million folks.
I guess where I get confused is how can they turn on and off servers without it messing things up.
Or, how can when the load is high, they can turn on one or more servers to help handle the load.
Does each server hold all the same data as the other servers. I wouldn't think so, or turning one off would cause data loss to those who were connected to that one. Unless they all mirror themselves together. But I don't think that is it either.
Back in the days of dial-up, I sorta understood how that system worked. They had a dial-up modem for every single person connecting to their ISP. It was more or less like the telephone company handled all the telephone calls to a single phone number, and a controller kept track of which Modem was Live, and when someone disconnected, then that Modem would be added back into the pool. In essence, if you had 50 modems and only 30 clients connected, those last 20 modems may never get used at all. Well, unless you physically removed a modem and replaced it with one.
Now, all these modems connected to the computers in the system. Plus they had a couple of huge storage devices.
When you called in to log-onto the system, it would run a log-in program, and if your name and password were found on the storage device where than info was held, it would then connect you to the server pool. You never knew exactly which server you would be connected to, just whichever one was the least busy I suppose.
Now, if you connected to a server for the purpose of checking your account status, billing, or other office type things, it connected you to a different server, than if you wanted to connect to check your e-mail, or to go on-line on the Web.
It's just that servers are so powerful, lots of RAM, and I'm thinking probably lots of storage space as well.
Back when I had all of my computers working, as they died it was no big deal, because my data was always saved to a pair of external hard drives.
I could also, although I forgot how now, probably SSH or something. Rather than switching to that computer using my KVM, I could connect to it from my main computer and use a program on it that I would normally connect to that computer using the KVM to run that program. I found that some more intensive programs ran faster, although I can explain why, if I used the main computer to connect to it, rather than jumping directly to that computer using the KVM. Seems like it was done over SSH, but I could be wrong, it's been over ten years since I ever did it that way anymore.
Seems like there were some things I couldn't do from the main computer through another computer, so I just went back to using the KVM. Not that I plan on doing anything like that ever again.
My main thing was just trying to understand how a company with 10,000 people on-line at the same time, could kick servers on when when that grows to 12,000 folks, and how they can turn off servers when it drops down to 8,000. Apparently the Data is NOT on the Servers, but drawn from databases or program servers. It almost has to be for it to be interactive, sorta. Not talking about multi-player games here like the kids play and they are all on the same game at the same time.
I'm more or less trying to understand how one person can be working on a data file at the same time as another person, perhaps doing something else on that same file, without it causing both to become corrupt. Which I know was a big problem back in the early days of many people being able to work on the same thing. If one person wanted to save their work, they had to wait for an active file to be saved. Also, they could not be altering the same area of the file, confusing.
I know on my Farm Town game, more than one person could be hired to work on the same farm belonging to someone else and we could see each other there. But I also know they save every single change instantly. So you go to do something, and someone beats you to it, it can't be done again, hi hi. Or like the trick we used to do, we would load the farm, then go off-line to lets say harvest the crops that were ready on the other players farm we were hired to do. This way we got the credit for doing the whole farm. But the so did someone else who did the farm at the same time we did. So we both got credit for doing it, but to the person who owns the farm, they didn't get double the harvest, hi hi. I do know they fixed that so if you do go offline to do a farm, and someone does it just ahead of you, you no longer get the credit for doing it, because it was already done, hi hi. But I'm sure their programs are much more complicated than the simple server set-up I'm concerned about learning about.
OK, I've rambled long enough. Of all the searches I've done on-line, none of them has yet explained how servers can be brought on-line when needed. They mention it but not how it does that without corrupting the data.
I guess where I get confused is how can they turn on and off servers without it messing things up.
Or, how can when the load is high, they can turn on one or more servers to help handle the load.
Does each server hold all the same data as the other servers. I wouldn't think so, or turning one off would cause data loss to those who were connected to that one. Unless they all mirror themselves together. But I don't think that is it either.
Back in the days of dial-up, I sorta understood how that system worked. They had a dial-up modem for every single person connecting to their ISP. It was more or less like the telephone company handled all the telephone calls to a single phone number, and a controller kept track of which Modem was Live, and when someone disconnected, then that Modem would be added back into the pool. In essence, if you had 50 modems and only 30 clients connected, those last 20 modems may never get used at all. Well, unless you physically removed a modem and replaced it with one.
Now, all these modems connected to the computers in the system. Plus they had a couple of huge storage devices.
When you called in to log-onto the system, it would run a log-in program, and if your name and password were found on the storage device where than info was held, it would then connect you to the server pool. You never knew exactly which server you would be connected to, just whichever one was the least busy I suppose.
Now, if you connected to a server for the purpose of checking your account status, billing, or other office type things, it connected you to a different server, than if you wanted to connect to check your e-mail, or to go on-line on the Web.
It's just that servers are so powerful, lots of RAM, and I'm thinking probably lots of storage space as well.
Back when I had all of my computers working, as they died it was no big deal, because my data was always saved to a pair of external hard drives.
I could also, although I forgot how now, probably SSH or something. Rather than switching to that computer using my KVM, I could connect to it from my main computer and use a program on it that I would normally connect to that computer using the KVM to run that program. I found that some more intensive programs ran faster, although I can explain why, if I used the main computer to connect to it, rather than jumping directly to that computer using the KVM. Seems like it was done over SSH, but I could be wrong, it's been over ten years since I ever did it that way anymore.
Seems like there were some things I couldn't do from the main computer through another computer, so I just went back to using the KVM. Not that I plan on doing anything like that ever again.
My main thing was just trying to understand how a company with 10,000 people on-line at the same time, could kick servers on when when that grows to 12,000 folks, and how they can turn off servers when it drops down to 8,000. Apparently the Data is NOT on the Servers, but drawn from databases or program servers. It almost has to be for it to be interactive, sorta. Not talking about multi-player games here like the kids play and they are all on the same game at the same time.
I'm more or less trying to understand how one person can be working on a data file at the same time as another person, perhaps doing something else on that same file, without it causing both to become corrupt. Which I know was a big problem back in the early days of many people being able to work on the same thing. If one person wanted to save their work, they had to wait for an active file to be saved. Also, they could not be altering the same area of the file, confusing.
I know on my Farm Town game, more than one person could be hired to work on the same farm belonging to someone else and we could see each other there. But I also know they save every single change instantly. So you go to do something, and someone beats you to it, it can't be done again, hi hi. Or like the trick we used to do, we would load the farm, then go off-line to lets say harvest the crops that were ready on the other players farm we were hired to do. This way we got the credit for doing the whole farm. But the so did someone else who did the farm at the same time we did. So we both got credit for doing it, but to the person who owns the farm, they didn't get double the harvest, hi hi. I do know they fixed that so if you do go offline to do a farm, and someone does it just ahead of you, you no longer get the credit for doing it, because it was already done, hi hi. But I'm sure their programs are much more complicated than the simple server set-up I'm concerned about learning about.
OK, I've rambled long enough. Of all the searches I've done on-line, none of them has yet explained how servers can be brought on-line when needed. They mention it but not how it does that without corrupting the data.
Re: Servers, Multiple
You asked for a simple explanation, and to be honest that is about all I could give you. There is a lot more to it that what I described. For example, we speak of "a server" but what exactly is a server? It can be a single tower like the one I have sitting on the floor next to me, or even a small box like your NAS. Those things are for small home networks. When it comes to enterprise systems and global networks, those small boxes are useless. The processes and services that make them work, switching and control, are still needed but on a much larger scale.
In the case of a server that 10,000 people use it's probably an entire rack full of motherboards that compose "the server"; possibly an entire room full of racks such as Google would need for its search engine. A classic server will not be used to run any user programs. That would be a waste of resources. Servers manage system services and processes as well as the flow of data. The storage of data and the actual manipulation of data is done on some other server. Think video cards. That nVidia card could be considered a video server which does nothing but generate frames of graphics and then sends those frames to the main processor. The main processor acting like another server will decide what to do with the graphics, i.e, store it or display it on your monitor. Likewise with pure data. Server storage space can be on some other computer in the form of a database. It can also be on hard drives or up in The Cloud. When the server sees data obtained from the database it follows scripted instructions to decide what to do with it. So, you see, a server functions to manage devices and serve data.
As I noted previously there is a type of server that manages all those thousands of connections as a device called a terminal management server. Each server, of course has limited capacity. Therefore all 10k connections are handled in batches of 100 (per our example) by several computers designed for the purpose. The management software knows which modems are trying to get to the server and which ones are idle. Thus it only forwards server queries from active modems. Down at the nitty gritty (kernel) level there are processes and services internal to the (Linux?) operating system. You've seen the list when you use the "htop" command in Linux. If the right service is disabled, then the hardware stays powered up but goes into an idle state. That effectively shuts down 100 possible access channels. Multiply this scenario by 100, i.e. have 100 terminal managers each responsible for 100 connections and that results in 10,000 managed connections. The terminal management server typically has the ability to monitor traffic going through it and thus report back to the controller to decide what gets shut down, or not.
Now, direct your attention to the microprocessors neither you or I can afford to buy these days. I know of one with 64 cores and costs as much as a new pick up truck. That means you actually have 64 microprocessors stacked on top of one another. Each one can do two things, or handle two threads of data so that this processor looks like 128 fully functional processing units. Not all those cores are used up when you play your farm game, for example. The innards of the processor plus the instructions from the farm game program determine how many cores actually get used. If you are modeling weather on a global scale in real time, you might actually need all 128 threads to see that Doppler radar on your monitor. But harvesting virtual potatoes? Nahh. We are talking one 64 core CPU here, but that giant server with 10,000,000 clients has a whole slew of these CPU chips stacked up in a rack in a server room. That is what you need to process orders if you are Amazon, for example.
Obviously you don't want to run 1000 CPU's with 64 cores all at the same time if you can avoid it. That takes a hella lot of power from the electric company. Thus there are processes and services built into the operating system which shuts down unneeded CPU's. Windows 11 does this, but it is also common in large server farms. When you are done computing for the day you can shut off the power to your computer, or just let it go to sleep. The same things happen either way. The kernel never shuts down entirely in Windows 11, unless you force it to do so temporarily with a restart command. The reason the kernel is kept alive, but not active, is to provide super fast start up sequences. The kernel remembers it's last state and keeps it in memory while asleep, even while the power is shut off. Thus when you turn on the computer again, the kernel is working instantly, or as soon as it can read it's status from memory. A room containing 128 servers might only have a dozen CPU's going at any one time. The others are processor dead, but with quick recall when needed. Thus the server never shuts down entirely, but the processors do go to sleep. The operating system would have monitoring capability in this case so that it knows when to power down or call up more processors.
Continuing on to the gaming scenario things get slightly more complicated, but not any more difficult to understand. The first thing you need to know is that what you download onto your computer to play the Farm Game is simply the rendering engine. The only thing that program can do, typically, is create and display the farm you previously constructed. The engine knows where to put things, but what things need to be put into place? The answer to that question is the inventory of items likely to be sitting on a cloud server somewhere. Thus your rendering engine queries the cloud database and knows what to put in your farm scene. If you change something on your farm, the cloud database gets modified to reflect the changes. That same scenario works for all people who happen to log into your farm. Keep in mind, however, that those neighbors are not actually looking at the farm on your computer. They simply queried the database storing your inventory and the rendering engine on their computer made a duplicate copy. This is why cloud computing is so awesome. Any number of people can visit your virtual farm, read the cloud data, and voila! They have a copy of what you are doing. Of course the database needs to stay current for all this to work smoothly.
All the people active on your farm are in sync - could be yet another server doing the sync job. You all read the same data from your farm's database in more or less real time. Thus, when somebody makes a change to the farm environment, the data base is updated instantly, and the new information is synced to all the other players' rendition of your farm. If it is possible to work the farm off line, then of course nobody else can see what you have done until you log back into the game server and everything is synchronized again. Should there be conflicts because somebody else made changes too, it's the synchronizing server's job to figure it out according to rules set up by the programmer.
The above scenario is typical for games but it is also what makes any collaborative software work as well as it does.
In the case of a server that 10,000 people use it's probably an entire rack full of motherboards that compose "the server"; possibly an entire room full of racks such as Google would need for its search engine. A classic server will not be used to run any user programs. That would be a waste of resources. Servers manage system services and processes as well as the flow of data. The storage of data and the actual manipulation of data is done on some other server. Think video cards. That nVidia card could be considered a video server which does nothing but generate frames of graphics and then sends those frames to the main processor. The main processor acting like another server will decide what to do with the graphics, i.e, store it or display it on your monitor. Likewise with pure data. Server storage space can be on some other computer in the form of a database. It can also be on hard drives or up in The Cloud. When the server sees data obtained from the database it follows scripted instructions to decide what to do with it. So, you see, a server functions to manage devices and serve data.
As I noted previously there is a type of server that manages all those thousands of connections as a device called a terminal management server. Each server, of course has limited capacity. Therefore all 10k connections are handled in batches of 100 (per our example) by several computers designed for the purpose. The management software knows which modems are trying to get to the server and which ones are idle. Thus it only forwards server queries from active modems. Down at the nitty gritty (kernel) level there are processes and services internal to the (Linux?) operating system. You've seen the list when you use the "htop" command in Linux. If the right service is disabled, then the hardware stays powered up but goes into an idle state. That effectively shuts down 100 possible access channels. Multiply this scenario by 100, i.e. have 100 terminal managers each responsible for 100 connections and that results in 10,000 managed connections. The terminal management server typically has the ability to monitor traffic going through it and thus report back to the controller to decide what gets shut down, or not.
Now, direct your attention to the microprocessors neither you or I can afford to buy these days. I know of one with 64 cores and costs as much as a new pick up truck. That means you actually have 64 microprocessors stacked on top of one another. Each one can do two things, or handle two threads of data so that this processor looks like 128 fully functional processing units. Not all those cores are used up when you play your farm game, for example. The innards of the processor plus the instructions from the farm game program determine how many cores actually get used. If you are modeling weather on a global scale in real time, you might actually need all 128 threads to see that Doppler radar on your monitor. But harvesting virtual potatoes? Nahh. We are talking one 64 core CPU here, but that giant server with 10,000,000 clients has a whole slew of these CPU chips stacked up in a rack in a server room. That is what you need to process orders if you are Amazon, for example.
Obviously you don't want to run 1000 CPU's with 64 cores all at the same time if you can avoid it. That takes a hella lot of power from the electric company. Thus there are processes and services built into the operating system which shuts down unneeded CPU's. Windows 11 does this, but it is also common in large server farms. When you are done computing for the day you can shut off the power to your computer, or just let it go to sleep. The same things happen either way. The kernel never shuts down entirely in Windows 11, unless you force it to do so temporarily with a restart command. The reason the kernel is kept alive, but not active, is to provide super fast start up sequences. The kernel remembers it's last state and keeps it in memory while asleep, even while the power is shut off. Thus when you turn on the computer again, the kernel is working instantly, or as soon as it can read it's status from memory. A room containing 128 servers might only have a dozen CPU's going at any one time. The others are processor dead, but with quick recall when needed. Thus the server never shuts down entirely, but the processors do go to sleep. The operating system would have monitoring capability in this case so that it knows when to power down or call up more processors.
Continuing on to the gaming scenario things get slightly more complicated, but not any more difficult to understand. The first thing you need to know is that what you download onto your computer to play the Farm Game is simply the rendering engine. The only thing that program can do, typically, is create and display the farm you previously constructed. The engine knows where to put things, but what things need to be put into place? The answer to that question is the inventory of items likely to be sitting on a cloud server somewhere. Thus your rendering engine queries the cloud database and knows what to put in your farm scene. If you change something on your farm, the cloud database gets modified to reflect the changes. That same scenario works for all people who happen to log into your farm. Keep in mind, however, that those neighbors are not actually looking at the farm on your computer. They simply queried the database storing your inventory and the rendering engine on their computer made a duplicate copy. This is why cloud computing is so awesome. Any number of people can visit your virtual farm, read the cloud data, and voila! They have a copy of what you are doing. Of course the database needs to stay current for all this to work smoothly.
All the people active on your farm are in sync - could be yet another server doing the sync job. You all read the same data from your farm's database in more or less real time. Thus, when somebody makes a change to the farm environment, the data base is updated instantly, and the new information is synced to all the other players' rendition of your farm. If it is possible to work the farm off line, then of course nobody else can see what you have done until you log back into the game server and everything is synchronized again. Should there be conflicts because somebody else made changes too, it's the synchronizing server's job to figure it out according to rules set up by the programmer.
The above scenario is typical for games but it is also what makes any collaborative software work as well as it does.
Re: Servers, Multiple
Thanks for the wonderful explanation Yogi!
As I was reading what you said, my mind drifted back to the Lisa System we had, ancient by today's standards of course.
But since we did have six work stations connected to, I knew a very little bit about how it kept from messing up.
It was fairly simple, two people could not be working on the same order updates at the same time.
Lisa would show a blank order form on an order takers screen, it also checked once a name or address was typed in, if they already existed in the system, and if they did, much of the form was filled in for us. Once we were done taking an order, we did have to save it simply by hitting an F-key for that purpose.
My dad could be doing accounting work, the billing department could be printing out invoices, order takers could be taking orders, and designers could mark they are working on an order, and mark it as filled. When the filled button was pressed, which was another F-key, it would appear in the proper routing table for the delivery drivers. When they loaded it on a truck, they also pushed an F-key that it was loaded, and when they got back, they punched another F-key for delivered, and they also had a place to put a note for the billing department to see, just in case a customer called in about their package.
Despite the fact we had computers, dad still did all of the actual accounting in his ledger books, by hand, and only looked at what the computer was showing him.
Then, once we got the WANG VS 350, about 40 more terminals were added along with hundreds of peripheral devices located in each of our greenhouse zones. And as I said before, we used so little on the computer, even with all that we added, Wang could still run programs on our computer to help it pay for itself, which it did.
Because we did not yet automate things like ventilators and fans, there was a monitor at the main office of each greenhouse that would instruct the workers which greenhouse needs to open or close a ventilator and by how much. It also handled turning on and off the lights for crops raised under black cloth. The greenhouse team entered every plant that was planted, when, and where, so we knew at a glance when a crop was ready to be potted up, moved, or ready for sale, etc.
And to think, all of that stuff in the greenhouses, at one time, used to be handled by little cards held to a long wire with clothespins. Every grower had their own wire full of clothespins for their rotation schedules. When a crop was near ready to sell, those cards were pulled and sent to the greenhouse distribution office, so they knew what was coming, how much, and if they could wholesale out the excess they couldn't sell directly in our own sales houses to customers.
One thing the Wang System didn't do was stop pilfering by family members, hi hi. aka Product Shrinkage, hi hi.
Now I know on both the Lisa System, and on the big Wang System, the programs we ran were on the computer, and only displayed on the workstations. Other than accounting, most of the programs could be brought up at any terminal.
Accounting was limited to only three of the workstations, my dad's terminal, uncle CJ's terminal, and it could be brought up in the Billing Department on the corner computer, which was normally only used for machine maintenance and servicing. Like running backups at night, and for the Wang Guy to gain access to the machine.
But I knew from looking at the machines functions. And without knowing what was inside the machine itself. Most of the Programs ran on a different part of the machine, while the database was like a file on the hard drive, and there were several databases also. On the Wang, every connection was logged as to who signed on at what time, and for how long. Plus which programs they ran and for how long. Boring to read though. It might show Joan as connecting from workstation 5 at 8:15am, brought up a new blank order form at 8:35am, and saved the order form as Daily #108 at 8:44am. Then another Log would show Designer Angela marked #108 in progress at 8:52am, finished and ready for delivery at 9:13am. Then #108 would show up on the drivers routing board for that route. You can figure out the rest from there easily enough.
There was a limit to how many workstations and devices could be active on the WANG VS 350. But the number of devices could be up in the hundreds, because they were only scanned for a reading like once every 15 seconds or more or less.
I know the WANG VS 550 could handle more than double the amount of workstations if we ever needed them, and I don't think there was a limit to the number of scanned devices, just that the more devices we added, the longer the amount of time before each one was scanned again. It was just a couple of seconds until we added so many in the greenhouses, and then it slowed down to probably more like around 10 seconds. And besides the Logs, there were alarms if like the temp in a greenhouse was above or below a pre-selected range, which then activated the notice to open or close a ventilator, or turn on or off a group of fans, or crank up the heat, hi hi.
Oh, well, BORING, hi hi.
I did learn a single server acting as a controller, controls how many servers are on and working at any given moment. Most are handled automatically, some have to be manually switched on, especially if they were dedicated to something else that had to stop to use it in the bank of servers for a heavy load process.
As I was reading what you said, my mind drifted back to the Lisa System we had, ancient by today's standards of course.
But since we did have six work stations connected to, I knew a very little bit about how it kept from messing up.
It was fairly simple, two people could not be working on the same order updates at the same time.
Lisa would show a blank order form on an order takers screen, it also checked once a name or address was typed in, if they already existed in the system, and if they did, much of the form was filled in for us. Once we were done taking an order, we did have to save it simply by hitting an F-key for that purpose.
My dad could be doing accounting work, the billing department could be printing out invoices, order takers could be taking orders, and designers could mark they are working on an order, and mark it as filled. When the filled button was pressed, which was another F-key, it would appear in the proper routing table for the delivery drivers. When they loaded it on a truck, they also pushed an F-key that it was loaded, and when they got back, they punched another F-key for delivered, and they also had a place to put a note for the billing department to see, just in case a customer called in about their package.
Despite the fact we had computers, dad still did all of the actual accounting in his ledger books, by hand, and only looked at what the computer was showing him.
Then, once we got the WANG VS 350, about 40 more terminals were added along with hundreds of peripheral devices located in each of our greenhouse zones. And as I said before, we used so little on the computer, even with all that we added, Wang could still run programs on our computer to help it pay for itself, which it did.
Because we did not yet automate things like ventilators and fans, there was a monitor at the main office of each greenhouse that would instruct the workers which greenhouse needs to open or close a ventilator and by how much. It also handled turning on and off the lights for crops raised under black cloth. The greenhouse team entered every plant that was planted, when, and where, so we knew at a glance when a crop was ready to be potted up, moved, or ready for sale, etc.
And to think, all of that stuff in the greenhouses, at one time, used to be handled by little cards held to a long wire with clothespins. Every grower had their own wire full of clothespins for their rotation schedules. When a crop was near ready to sell, those cards were pulled and sent to the greenhouse distribution office, so they knew what was coming, how much, and if they could wholesale out the excess they couldn't sell directly in our own sales houses to customers.
One thing the Wang System didn't do was stop pilfering by family members, hi hi. aka Product Shrinkage, hi hi.
Now I know on both the Lisa System, and on the big Wang System, the programs we ran were on the computer, and only displayed on the workstations. Other than accounting, most of the programs could be brought up at any terminal.
Accounting was limited to only three of the workstations, my dad's terminal, uncle CJ's terminal, and it could be brought up in the Billing Department on the corner computer, which was normally only used for machine maintenance and servicing. Like running backups at night, and for the Wang Guy to gain access to the machine.
But I knew from looking at the machines functions. And without knowing what was inside the machine itself. Most of the Programs ran on a different part of the machine, while the database was like a file on the hard drive, and there were several databases also. On the Wang, every connection was logged as to who signed on at what time, and for how long. Plus which programs they ran and for how long. Boring to read though. It might show Joan as connecting from workstation 5 at 8:15am, brought up a new blank order form at 8:35am, and saved the order form as Daily #108 at 8:44am. Then another Log would show Designer Angela marked #108 in progress at 8:52am, finished and ready for delivery at 9:13am. Then #108 would show up on the drivers routing board for that route. You can figure out the rest from there easily enough.
There was a limit to how many workstations and devices could be active on the WANG VS 350. But the number of devices could be up in the hundreds, because they were only scanned for a reading like once every 15 seconds or more or less.
I know the WANG VS 550 could handle more than double the amount of workstations if we ever needed them, and I don't think there was a limit to the number of scanned devices, just that the more devices we added, the longer the amount of time before each one was scanned again. It was just a couple of seconds until we added so many in the greenhouses, and then it slowed down to probably more like around 10 seconds. And besides the Logs, there were alarms if like the temp in a greenhouse was above or below a pre-selected range, which then activated the notice to open or close a ventilator, or turn on or off a group of fans, or crank up the heat, hi hi.
Oh, well, BORING, hi hi.
I did learn a single server acting as a controller, controls how many servers are on and working at any given moment. Most are handled automatically, some have to be manually switched on, especially if they were dedicated to something else that had to stop to use it in the bank of servers for a heavy load process.
Re: Servers, Multiple
My very lengthy explanation of networking is necessarily broad and not the only way things can be done. Network architecture varies depending on the application and the size, not to mention the expertise of the network administrator. All the above can be done on one machine, such as your NAS. It can also be spread over seven continents as is Google, Amazon, Microsoft, and the others. Your description of the Lisa machines and the Wang system is familiar and interesting to read about how some of the theory I know can be applied to the real world.
Taking a server off line isn't as exotic as it seems. It takes more or less planning how to do it depending on the number of clients that will be affected by the missing server. In large networks, as my babbling tells, it can be all automated with no human intervention. If I had it all to do over again with Motorola, I would have tried to break into the corporate network administration scene. I managed a factory full of terminals at one point, but that only amounted to 300 clients or so. Corporate probably had a thousand times that for clients and a thousand times more stress keeping it all running. Believe it or not managing big and complicated is my kind of stress. LOL
Taking a server off line isn't as exotic as it seems. It takes more or less planning how to do it depending on the number of clients that will be affected by the missing server. In large networks, as my babbling tells, it can be all automated with no human intervention. If I had it all to do over again with Motorola, I would have tried to break into the corporate network administration scene. I managed a factory full of terminals at one point, but that only amounted to 300 clients or so. Corporate probably had a thousand times that for clients and a thousand times more stress keeping it all running. Believe it or not managing big and complicated is my kind of stress. LOL
Re: Servers, Multiple
I was reading about a guy last night who took 60 Raspberry PI computer boards and built 3 cabinets of 20 in each.
He's using two older MAC computers as the controllers, each with a different purpose.
Each of those computer boards has a SIM card in them, but he is using a rack of Solid State Drives for the data storage.
But like every other article I've read, he shows what equipment he has, and partly about how it is all connected, but never how it all works.
If I wasn't so old, and if I had the time, I would probably take a class on computer sciences or something, just to have the knowledge of how it all works. Not that I would ever have a need to use it, and from disuse would forget it all.
But I've always been the curious type and like to know how they did things, if you know what I mean.
He's using two older MAC computers as the controllers, each with a different purpose.
Each of those computer boards has a SIM card in them, but he is using a rack of Solid State Drives for the data storage.
But like every other article I've read, he shows what equipment he has, and partly about how it is all connected, but never how it all works.
If I wasn't so old, and if I had the time, I would probably take a class on computer sciences or something, just to have the knowledge of how it all works. Not that I would ever have a need to use it, and from disuse would forget it all.
But I've always been the curious type and like to know how they did things, if you know what I mean.
Re: Servers, Multiple
I do indeed know what you mean when you tell me that you have a need to know how things work. Your stories about your Tuesday trips to local manufacturing facilities says it all. But, I also know you try to understand things on a micro level as well. There is in fact great benefit in knowing how things work even though you have no involvement with the subject. That knowledge teaches you techniques that your creative mind calls upon when you need to think out of the box. That is exactly why a college education is highly valuable. You learn things for which you have no immediate use, but your thinking process is affected forever by it all.
Those articles that never explain the details of how things work are necessarily written that way. The author is usually explaining a concept and not giving practical application instruction on how to apply that concept. The assumption is that the reader already knows enough to find the article worth reading in the first place. Going into details would be boring and somewhat of an insult to the intended audience. Your age should not be the reason you can't build your own network at home and learn all those fuzzy concepts that intrigue you. I know you are very capable of learning new things. Your health and your finances are likely the limiting factors. Your body might be a problem, but there certainly is nothing wrong with your mind.
Those articles that never explain the details of how things work are necessarily written that way. The author is usually explaining a concept and not giving practical application instruction on how to apply that concept. The assumption is that the reader already knows enough to find the article worth reading in the first place. Going into details would be boring and somewhat of an insult to the intended audience. Your age should not be the reason you can't build your own network at home and learn all those fuzzy concepts that intrigue you. I know you are very capable of learning new things. Your health and your finances are likely the limiting factors. Your body might be a problem, but there certainly is nothing wrong with your mind.
Last edited by yogi on 22 Oct 2022, 16:14, edited 1 time in total.
Re: Servers, Multiple
Because I got by with murder, I actually enjoyed the college classes I took over the span of about 18 years.
A lot of those were special seminars required to keep up with various licenses, mostly horticultural in the beginning.
I learned just enough chemistry over a couple of years to really be dangerous, hi hi.
I don't know Yogi, I truly forget a lot of things lately, and a couple of them were important.
I mentioned something to my wife, and for the life of her, she can't remember the phrase I used.
We've both been wracking our brains for a week trying to remember what I said, again it was important, hi hi.
I truly did enjoy seeing how other companies did things, and I'm sure it helped with my thinking outside the box.
Every once in a while, something will pop into my head from my restaurant days, on how we did a menial task fast and efficiently, when we didn't have the semi-automatic tools to do them with.
I did play with a server on my own many years ago when I installed the EduBuntu server.
At first I thought that was the way to go, but then realized it was better just to use a stand-alone computer and run external backups of all my data. And I'm sure glad I did keep so many backups, when my external Raid Array had the controller card go bad. I've stayed away from Raid ever since. Sorta. My NAS before yours was a Raid Array of two disks. But I still kept everything redundantly on external drives, just in case, hi hi.
A lot of those were special seminars required to keep up with various licenses, mostly horticultural in the beginning.
I learned just enough chemistry over a couple of years to really be dangerous, hi hi.
I don't know Yogi, I truly forget a lot of things lately, and a couple of them were important.
I mentioned something to my wife, and for the life of her, she can't remember the phrase I used.
We've both been wracking our brains for a week trying to remember what I said, again it was important, hi hi.
I truly did enjoy seeing how other companies did things, and I'm sure it helped with my thinking outside the box.
Every once in a while, something will pop into my head from my restaurant days, on how we did a menial task fast and efficiently, when we didn't have the semi-automatic tools to do them with.
I did play with a server on my own many years ago when I installed the EduBuntu server.
At first I thought that was the way to go, but then realized it was better just to use a stand-alone computer and run external backups of all my data. And I'm sure glad I did keep so many backups, when my external Raid Array had the controller card go bad. I've stayed away from Raid ever since. Sorta. My NAS before yours was a Raid Array of two disks. But I still kept everything redundantly on external drives, just in case, hi hi.
Re: Servers, Multiple
It may not have occurred to you, but, you are in fact running several servers on that Debian box you like to use. Well, you are if you are running Samba which can be a server as well as a client for file transfers. Your IP addresses on your LAN are delivered by a DHCP server installed in two places the last time we talked about it. And, printing is done in Linux via yet another server. My point here is that servers do not have to be physical boxes of electronics nor cinder blocked rooms full of motherboards and a/c units. A server can be as compact as a single program running on your desktop computer.
You can learn a lot about servers by studying something such as Samba. I did that when most Linux distributions changed their file managers to not be able to see Windows shares. To be honest I got very confused when I tried reading up on the subject, but I did walk away from it with at least a vague understanding of why I had problems with the Samba Server ... I should have been using the client, not the server. LOL Unfortunately the explanations I dug up were all very technical and assumed I knew a lot ahead of time. That's my pet peeve with Linux in general. They assume the user is well versed in areas they have no concept about.
As they say, memory is the second thing to go for us old timers. I've not personally forgot important things entirely, but I have had to spend more than the usual amount of time recalling what used to be familiar. My wife forgets things that are surprising. Not too long ago she came into this computer room and looked at the router. She asked what it was and when did I get it. I was slightly stunned because it is about ten years old and I had it visible and working in our previous house. She claims she doesn't remember ever seeing it before.
You can learn a lot about servers by studying something such as Samba. I did that when most Linux distributions changed their file managers to not be able to see Windows shares. To be honest I got very confused when I tried reading up on the subject, but I did walk away from it with at least a vague understanding of why I had problems with the Samba Server ... I should have been using the client, not the server. LOL Unfortunately the explanations I dug up were all very technical and assumed I knew a lot ahead of time. That's my pet peeve with Linux in general. They assume the user is well versed in areas they have no concept about.
As they say, memory is the second thing to go for us old timers. I've not personally forgot important things entirely, but I have had to spend more than the usual amount of time recalling what used to be familiar. My wife forgets things that are surprising. Not too long ago she came into this computer room and looked at the router. She asked what it was and when did I get it. I was slightly stunned because it is about ten years old and I had it visible and working in our previous house. She claims she doesn't remember ever seeing it before.
Re: Servers, Multiple
Yeppers, everything inside the computer is usable because of servers, internal and external.
For a while I was using an old computer as a print server, and it helped a lot.
There are basically two types of people who buy computers, and they are not much different than those who buy cars either.
Windows laid the groundwork and had the programs and hit the ground running, which gave them the edge to dominate the home computer market.
Of the two types of people who buy cars, one group goes only about the color of the car. The other type want's to know what size engine is in it, and what is the gear ratio of the rear end, etc.
Of the two types of people who buy computers, one group buys only the one that works with all the Windows programs they have bought, and the one computer they learned how to use, which was Windows.
Now Apple got their computers into many schools, so the kids who learned on them, often buy Apple/MAC computers, at least until they are retrained in the workplace to use Windows computers.
Then there is a whole other group of people, like those who want to know what equipment is in the car before they buy it. How can I modify it, get more horsepower, take it to the track, etc. Now these folks are the ones who buy Linux based computers, because they can tweak and tune it to their hearts desire, and it runs like a jackrabbit compared to the Doze.
But like me, I just use it because it works for me, and is free, and I can fiddle with the controls, even though most of the time I don't unless I know what will happen.
You don't buy a Chevy Starter motor for a Ford. OK, you also don't buy Windows software to run on a LInux Box, problem solved.
I often wonder what Debi's father would have thought if he were still here and looked into Debi's office.
We did give him a Windows 95 computer to play a few games on.
But looking into Debi's office now, you see tons of LED's flashing, the modem, the router, then NAS, the backup external drives, and that big old gaming computer box with all of its flashing lights her son gave her.
I'll bet to someone in their 80s, it would seem like they are looking into an alien spacecraft, hi hi.
I had that same feeling after one of my epilepsy attacks while working on video and gaming machines.
I expected to see those big old glass gandy dancer radio tubes when I pulled the back off the cabinet, and instead there were these little flat boards with what looked like plastic spiders all over them. It was scary for me that day! Thought I slipped through a Time Warp or something, except for the fact I recognized my office and everything in it, but no longer knew what any of that equipment was for or how to use it. And this happened to me nearly every five years.
For a while I was using an old computer as a print server, and it helped a lot.
There are basically two types of people who buy computers, and they are not much different than those who buy cars either.
Windows laid the groundwork and had the programs and hit the ground running, which gave them the edge to dominate the home computer market.
Of the two types of people who buy cars, one group goes only about the color of the car. The other type want's to know what size engine is in it, and what is the gear ratio of the rear end, etc.
Of the two types of people who buy computers, one group buys only the one that works with all the Windows programs they have bought, and the one computer they learned how to use, which was Windows.
Now Apple got their computers into many schools, so the kids who learned on them, often buy Apple/MAC computers, at least until they are retrained in the workplace to use Windows computers.
Then there is a whole other group of people, like those who want to know what equipment is in the car before they buy it. How can I modify it, get more horsepower, take it to the track, etc. Now these folks are the ones who buy Linux based computers, because they can tweak and tune it to their hearts desire, and it runs like a jackrabbit compared to the Doze.
But like me, I just use it because it works for me, and is free, and I can fiddle with the controls, even though most of the time I don't unless I know what will happen.
You don't buy a Chevy Starter motor for a Ford. OK, you also don't buy Windows software to run on a LInux Box, problem solved.
I often wonder what Debi's father would have thought if he were still here and looked into Debi's office.
We did give him a Windows 95 computer to play a few games on.
But looking into Debi's office now, you see tons of LED's flashing, the modem, the router, then NAS, the backup external drives, and that big old gaming computer box with all of its flashing lights her son gave her.
I'll bet to someone in their 80s, it would seem like they are looking into an alien spacecraft, hi hi.
I had that same feeling after one of my epilepsy attacks while working on video and gaming machines.
I expected to see those big old glass gandy dancer radio tubes when I pulled the back off the cabinet, and instead there were these little flat boards with what looked like plastic spiders all over them. It was scary for me that day! Thought I slipped through a Time Warp or something, except for the fact I recognized my office and everything in it, but no longer knew what any of that equipment was for or how to use it. And this happened to me nearly every five years.
Re: Servers, Multiple
Computers, cars, or whatever, are obtained to satisfy a need. We all have different needs and expectations so that there are literally millions of choices when we go shopping. My friend in the Czech Republic tells the story of how there was a Velvet Revolution when they kicked those godless commies out of officer and installed a democratic republic in its place. She says it's not much different in most respects, but when she goes shopping there are now millions of choices instead of only the state made dozen or so clones. That all says your needs vary in proportion to the supply.
It's a shame that your memory is so fragile due to epilepsy. On the other hand you seem to have picked up what you forgot as a result of the attacks. I don't know much about that disorder but to me memory problems are the worst kind to be plagued with. They seem to be so debilitating and make you feel helpless as well. I could say that would be an argument in favor of using Windows because it just works for everybody. But I know your likely response so I'll just sit here and answer some of the other posts.
It's a shame that your memory is so fragile due to epilepsy. On the other hand you seem to have picked up what you forgot as a result of the attacks. I don't know much about that disorder but to me memory problems are the worst kind to be plagued with. They seem to be so debilitating and make you feel helpless as well. I could say that would be an argument in favor of using Windows because it just works for everybody. But I know your likely response so I'll just sit here and answer some of the other posts.
Re: Servers, Multiple
Most folks who buy cars today, don't know very much about them. Many don't even know how to change a flat tire, or in some cases, how to check the oil or transmission fluid on automatics. When they try, they are the ones looking for 710 oil for their car, because that's what is written on the Cap, hi hi. Like the two guys standing facing each other arguing about the number printed on the ground in front of them. One says it's a six, the other says no it's a nine, hi hi.
What I seem to forget after an attack is something that was learned recently and/or required book learning or training.
But then some things I remember from doing them hands on for many years. Although I do forget the names of many things I once new, like all the timber names for lumber depending on their location in the house. Rafters, Studs I remember, but Band Boards, Joist Types, etc. that I've long forgotten.
As far as Windows goes, I knew how to use 3.0, 3.11, and XP, but 7, 8, 10, forget it. It's not the same Windows anymore.
Probably one of the reasons I love Debian with the MATE desktop, is it is like Windows early years. I hate the Cinnamon desktop because it is like a fancy cell phone, and all cluttered up.
What I seem to forget after an attack is something that was learned recently and/or required book learning or training.
But then some things I remember from doing them hands on for many years. Although I do forget the names of many things I once new, like all the timber names for lumber depending on their location in the house. Rafters, Studs I remember, but Band Boards, Joist Types, etc. that I've long forgotten.
As far as Windows goes, I knew how to use 3.0, 3.11, and XP, but 7, 8, 10, forget it. It's not the same Windows anymore.
Probably one of the reasons I love Debian with the MATE desktop, is it is like Windows early years. I hate the Cinnamon desktop because it is like a fancy cell phone, and all cluttered up.
Re: Servers, Multiple
I understand where you are coming from regarding people who don't know anything about cars and computers yet go about buying them Willy Nilly. They can do that because most cars and most computers ... just work. It's not necessary to know the oil even needs to be changed because that car will work regardless of what you know about it. Which brings me to a story I know will make you smile.
I've posted about the wedding I attended recently up near Chicago. My youngest granddaughter, Grace, was the one being married. She and her significant other live in Maryland these days, but both grew up in the same Chicago suburb. Thus that was the logical spot for the wedding. They set off from Maryland two days before the wedding and almost made it to their destination. The car's engine shut down and could not be started again. They were close enough to Grace's dad's house so that he was able to rescue them and have the car towed to the local mechanic. It was tense for a few hours but the mechanic finally called to say everything is working again. Apparently the engine shut down ... because it was low on oil and overheated. Yes, Grace, you need to check the oil from time to time and have it changed.
I didn't realize cars shut down in those situations. I figured the engines just melted and fused together when they didn't have enough oil. I think they have a Mazda with a computer running the whole show. Good thing the computer knew what to do, eh?
I've posted about the wedding I attended recently up near Chicago. My youngest granddaughter, Grace, was the one being married. She and her significant other live in Maryland these days, but both grew up in the same Chicago suburb. Thus that was the logical spot for the wedding. They set off from Maryland two days before the wedding and almost made it to their destination. The car's engine shut down and could not be started again. They were close enough to Grace's dad's house so that he was able to rescue them and have the car towed to the local mechanic. It was tense for a few hours but the mechanic finally called to say everything is working again. Apparently the engine shut down ... because it was low on oil and overheated. Yes, Grace, you need to check the oil from time to time and have it changed.
I didn't realize cars shut down in those situations. I figured the engines just melted and fused together when they didn't have enough oil. I think they have a Mazda with a computer running the whole show. Good thing the computer knew what to do, eh?
Last edited by yogi on 25 Oct 2022, 17:16, edited 1 time in total.
Re: Servers, Multiple
Not the old cars like I drive, but all these newer cars do have all kinds of sensors in them, which can keep them from running.
My 1997 replacement Blazer does have an oil sensor but it is not there to shut down the engine, just to display the oil pressure and flash a red light when it is low. However, this replacement Blazer has an aftermarket Auto-Start system. A couple of times the car would start then shut off right away. I checked the oil and it was 2 quarts down, added the oil and the Auto-start worked again. But then later it stopped working again, and the oil was just fine, full and clean. So there was something else causing it to not keep running. Only this time it would try three times before not trying again.
I got a grease job and oil change new filters for both air and oil, plus the cabin filter too this time. Then it started OK for about 2 weeks, then went back to not starting again.
This Blazer does not have the engine codes on the secondary display panel like my original Blazer did, so I have to take it to a shop that has a reader they plug into a port under the dash. But it is not showing any bad sensors, and they don't mess with aftermarket add-ons to a car. But then it surprises me and the car will start using the auto-start, like when I accidentally bump into something and the button gets pressed in my pocket. Naturally, then it will start and run until the engine is up to temp and shut off. I used to do that a lot with my old battery to keep it charged up, since I only use my car maybe once every other week, hi hi.
I once had a used Cadillac, and that thing had all kinds of things to go wrong. Heated seat and steering wheel. Once the car was started of course, plus it had instant heat before the engine was warm, but it still took about 15 seconds before you felt the heat from the heater. Unfortunately, that car was a real lemon and cost me a fortune in repairs. Glad to be rid of it, even though it cost me like 7 grand in repairs, I traded it in for only 2 grand toward a new van for my work.
My 1997 replacement Blazer does have an oil sensor but it is not there to shut down the engine, just to display the oil pressure and flash a red light when it is low. However, this replacement Blazer has an aftermarket Auto-Start system. A couple of times the car would start then shut off right away. I checked the oil and it was 2 quarts down, added the oil and the Auto-start worked again. But then later it stopped working again, and the oil was just fine, full and clean. So there was something else causing it to not keep running. Only this time it would try three times before not trying again.
I got a grease job and oil change new filters for both air and oil, plus the cabin filter too this time. Then it started OK for about 2 weeks, then went back to not starting again.
This Blazer does not have the engine codes on the secondary display panel like my original Blazer did, so I have to take it to a shop that has a reader they plug into a port under the dash. But it is not showing any bad sensors, and they don't mess with aftermarket add-ons to a car. But then it surprises me and the car will start using the auto-start, like when I accidentally bump into something and the button gets pressed in my pocket. Naturally, then it will start and run until the engine is up to temp and shut off. I used to do that a lot with my old battery to keep it charged up, since I only use my car maybe once every other week, hi hi.
I once had a used Cadillac, and that thing had all kinds of things to go wrong. Heated seat and steering wheel. Once the car was started of course, plus it had instant heat before the engine was warm, but it still took about 15 seconds before you felt the heat from the heater. Unfortunately, that car was a real lemon and cost me a fortune in repairs. Glad to be rid of it, even though it cost me like 7 grand in repairs, I traded it in for only 2 grand toward a new van for my work.
Re: Servers, Multiple
I've heard stories over the years regarding the poor quality in those Cadillac automobiles. They are supposed to be premium products, but the only thing that is premium about them is the MSRP. The first Saturn I bought had all the luxuries you can put into a Saturn because it was a demo model. It was well over ten years old when I traded it in for another less adorned Saturn but one with only 36,000 miles on it. Been driving a Saturn for at least a decade and am pretty well pleased with how they are built. I've looked under the hood a time or two, but never so much as cleaned the battery cables. I used to do my own maintenance and tune ups, but that was before computers and fuel injection came onto the scene. Oh, I do put air in the tires when the sensors tell me one is low on pressure. But, that's about it for maintenance.
Wife's 2019 Toyota might just as well be a Boeing 747 given all the electronics behind the steering wheel. They must be nice cars to drive if you can get used to all the "safety" and convenience features. I do like the rear view camera and the fact that you can integrate your smartphone into the car's dash instrumentation. I guess they do that for hands free telephoning, or something. In any case I'd not even know where to start on that car if it required maintenance. Well, maybe I can put air in those tires too. I recognize those cute little air valves on the tire rim.
Wife's 2019 Toyota might just as well be a Boeing 747 given all the electronics behind the steering wheel. They must be nice cars to drive if you can get used to all the "safety" and convenience features. I do like the rear view camera and the fact that you can integrate your smartphone into the car's dash instrumentation. I guess they do that for hands free telephoning, or something. In any case I'd not even know where to start on that car if it required maintenance. Well, maybe I can put air in those tires too. I recognize those cute little air valves on the tire rim.
Re: Servers, Multiple
Mow you know why I spent a year looking for another 1997 Blazer after I totaled mine do to an ABS System Failure disabling my brakes. I did hit the emergency brake, but rear wheels don't do much for stopping you at highway speeds.
I got burned bad on that Cadillac I bought. Same way with the Dodge Charger I bought for my wife. I bought it against my better judgement, but since she was working at the dealership where we bought it, I thought they wouldn't burn us. But they did and really bad too.
I've not met too many old timers who like how the transmissions work in these new cars, not even if they are rear-wheel-drive cars. It must be all electronic and with serious delays to everything. And then there is that new weird transmission that works like a variable ratio lathe. Crazy to put something like that in a car.
I got burned bad on that Cadillac I bought. Same way with the Dodge Charger I bought for my wife. I bought it against my better judgement, but since she was working at the dealership where we bought it, I thought they wouldn't burn us. But they did and really bad too.
I've not met too many old timers who like how the transmissions work in these new cars, not even if they are rear-wheel-drive cars. It must be all electronic and with serious delays to everything. And then there is that new weird transmission that works like a variable ratio lathe. Crazy to put something like that in a car.
Re: Servers, Multiple
The performance of automobiles went downhill the first year they started with antipollution devices being added. It's been a battle ever since because they are trying to burn the fuel efficiently and sacrificing performance to do it. There also were laws about increasing fleet MPG ratings which made the designers go into a frenzy to reduce the weight of a vehicle in every way possible. That's when they switched to the donut for a spare tire, for example. The lighter cars got better mileage, but the costs went up instead of down. I wonder why that was necessary. So, obviously, the standards that must be followed today in the automobile industry are not the same as what they were when your Blazer was a new model. We probably are getting better mileage and cleaner exhausts, but at what cost?
Re: Servers, Multiple
In the 1960's when they ADDED an air-pump to blow fresh air into the exhaust pipes to FOOL the emissions sensors during the inspection. It actually made the cars burn more fuel, in order to run that useless air-pump.
My brother proved this easily. His car did not have an air-pump, and it didn't pass the emissions inspection. So he went home, mounted a scuba tank in the trunk and ran a pipe into the exhaust pipe. Drove back to the inspection station, turned on the air from his scuba tank, and now it passed with flying colors.
I know the emissions test no longer works that way, but it did back then.
It was just another Royal HOAX pulled on the people by the government!
I also paid some big bucks to have my van converted to run on Propane. Sounded great at the time, as propane was cheap to buy and did get me more miles per gallon than gasoline did. However, my van had no power to speak of. On the highway when I hit a hill, and there are a lot of them in Missouri, sometimes with the engine floored, it would still slow down by the time I got to the top of each hill. Even my little 4-cylinder cars had more power than that did using Propane.
I had it taken off and my old system installed back the way it was and sold the propane set-up to someone else. That was an expensive thing to find out that wasn't worth it.
Then a year or so later, another guy using Propane power for a delivery van also added a small touch of oxygen to the mixture and he had plenty of power, but his engine ran about 20 degrees hotter than it should. Plus I think he burned out his valves from doing that too.
There are different types of Hydrogen Fuel Cell vehicles in testing. Those that use an internal combustion engine are horrible, not really emission friendly either. But one I read about was really great, the problem was, it cost too much to build the power system, even if it was done in mass production, but it would have been emissions free.
I still think in the future, most cars will have miniature nuclear power plants in them, just big enough to power the electric motors of the car. And unlike LIthium batteries, they won't go up in flames and smoke, hi hi.
My brother proved this easily. His car did not have an air-pump, and it didn't pass the emissions inspection. So he went home, mounted a scuba tank in the trunk and ran a pipe into the exhaust pipe. Drove back to the inspection station, turned on the air from his scuba tank, and now it passed with flying colors.
I know the emissions test no longer works that way, but it did back then.
It was just another Royal HOAX pulled on the people by the government!
I also paid some big bucks to have my van converted to run on Propane. Sounded great at the time, as propane was cheap to buy and did get me more miles per gallon than gasoline did. However, my van had no power to speak of. On the highway when I hit a hill, and there are a lot of them in Missouri, sometimes with the engine floored, it would still slow down by the time I got to the top of each hill. Even my little 4-cylinder cars had more power than that did using Propane.
I had it taken off and my old system installed back the way it was and sold the propane set-up to someone else. That was an expensive thing to find out that wasn't worth it.
Then a year or so later, another guy using Propane power for a delivery van also added a small touch of oxygen to the mixture and he had plenty of power, but his engine ran about 20 degrees hotter than it should. Plus I think he burned out his valves from doing that too.
There are different types of Hydrogen Fuel Cell vehicles in testing. Those that use an internal combustion engine are horrible, not really emission friendly either. But one I read about was really great, the problem was, it cost too much to build the power system, even if it was done in mass production, but it would have been emissions free.
I still think in the future, most cars will have miniature nuclear power plants in them, just big enough to power the electric motors of the car. And unlike LIthium batteries, they won't go up in flames and smoke, hi hi.
Re: Servers, Multiple
I recall those air pumps we were obligated to have on our cars. I don't see how that was a hoax, however. The exhaust being measured was the same exhaust gases that every gasoline powered engine spews out, but in this case it was force mixed with air from that blower. Of course the output would be cleaner because you are adding more air. And it doesn't matter if the air being added is from a pump in the engine compartment or a scuba tank in your trunk. The end result is the same, i.e., diluted exhaust gases. I will agree that the total amount of pollution from the engine didn't change, but that's not what was being tested. Back in those days it did not matter how the exhaust gasses were generated as long as the proportions at the end of the exhaust pipe were within the specified limits. That's not the case these days.
Of course anything could happen, but I doubt that nuclear powered automobiles is the future. Right now the nuclear power plants are generating too much radioactive waste because they are running out of places to dump it safely. If you increase the waste byproducts by putting a nuclear powered engines in every car, the entire planet will become radioactive in a matter of months. I think there is a better chance of seeing automobiles being replaced by a different type of technology than there is a chance of them going nuclear.
Of course anything could happen, but I doubt that nuclear powered automobiles is the future. Right now the nuclear power plants are generating too much radioactive waste because they are running out of places to dump it safely. If you increase the waste byproducts by putting a nuclear powered engines in every car, the entire planet will become radioactive in a matter of months. I think there is a better chance of seeing automobiles being replaced by a different type of technology than there is a chance of them going nuclear.