User is able to understand what is good and what is not.
If you are a going to move responsibility to the user side, this is calling bad software design.
User didn't must be software developer, CS engineer or something else to use your program.
BTW, CAT is showing few times less RAM that actually used by JVM in the environment - 250 MB in CAT vs >1GB in task manager.
Is it reliable? For sure no. but again, for this issue for some reason user must be responsible because you think so.
No, you have an Option to slow the CAT fifo, it's fully configurable. Probably you don't know about this function.
The period for recall algo is nothing to the number of API calls. The number of open markets in single exchange much more important here.
And again - it's a 100% responsibility to count the number of API calls from CAT side to avoid IP block from the exchange.
User physically can't count these things, because of it much deeper inside CAT.
And it software developer could approximately count number of calls that need to do some algo and predict network load.
What you will say to the generic user who didn't know what is http, rest, json etc?
Go to CS university and learn all this stuff before ask to you?
CAT handle all the errors
Not truth.
yes, the only difference is that i know every single line of code. You don't know nothing.
BUT you have the arrogance to talk about HOW my program works, what is able to do and what is not.
And i don't want to loose my time to find PM from customers who tells me that they run CAT for weeks without stops.
Thinking that your users an a idiots - not the best practice.
Ah, code. Even without knowledge of java, do you think it needs much time to use jd-gui or jad to have sources? Want to do a public code review?
I don't tell anything arrogance about software. I didn't tell that it good or bad.
I show you problems, in hope that they will be fixed. But, unfortunately, I see you are too much confident in CAT.
You tell that it's 24/7, I showed you 1 example why it not - silent restart is not possible.
Without any arguments from your side, you still say that it 24/7. No sense to argue with you more.
And about your programm and how it work. Sorry, but it's a you tell that you sell a automatic bot.
You tell that you sell 24/7 program.
You tell that you sell program that can run on the 100 markets.
But where the modern algo for trading if it automatic bot?
Where is automatic silent recovery from crash without user invention if it 24/7?
Where is good UI, design of program that can effectivelly run on 100 markets?
All this things only in theory, not in practice.
I read all the comments in the thread and only couple of comments even tell success results.
But no comments that long run was able without problem. Long I mean weeks or more. Couple of people reported someting like a week. Not so confident result.
But no comments that 100+ markets run was able without problem.
Etc.
If the running one ping-pong order for a 1 week on 1 market it's a real goal of users - not a problem.
They got results with CAT. But I simply don't believe that anyone who needs specific work like 100 markets are really happy with CAT and its problems.
Now, if you have suggestions or error log to give me, i'm here.
Otherwise, discussion is close.
In PM I gave you log with error description. And instead of help I receive from you stupid (really, no sense and stupid) grep search that are totally not the place of the problem.
Sure I analyzed logs myself, found the problem and try to use CAT that will avoid coming to the problems. But it's a not a software side fix.
It's a like not a pressing 'that button', or application will crash again.
BTW, honestly, now, when I see some problem with application, UI, algo, etc, I just prefer to rearrange my work or app settings instead of asking your help and fix.
Think about it.
