| Pages:
1
..
11
12
13
14
15
..
19 |
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by Daniel UK1  | | Quote: Originally posted by Daniel UK1 | Error, systems being developed on CL, systems being generated and then when i click on then, the graph is deleted and i just see 0... i have tested
different version of GSB, i have removed and loaded price files again, i have changed dates... the strange thing is that dates on the graph thats
deleted, shows 1899 12 30 .... i am using US date formats and you can see price file in the screenshot... i dont know what to do... can not develop
any systems until hat is causing this is fixed, kindly help// Thanks
|
Btw its not only on CL where it has happened |
Sorry I missed this.
Save opt settings.
As a test, reset your opt settings to default. (done by right click them)
then change settings and test again.
If that fixes it (likely) go back to an older copy of your settings that worked.
The same process can be done for app settings, but likely fauly is in opt settings.
then send me opt settings that work and fail so I can diagnose further.
Thanks received (1):
+1 Daniel UK1 at 2019-10-28 12:38:27
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by OUrocketman | Peter,
Hope you and your family are well, and thank you for all the work you put into your products!
I have noticed what could be a bug. Whenever I run the manager 55.60 with Nth mode set to no trade some of the systems come back with results shown
as having nth mode set to all and some come back as having nth mode set to no trade. Important to mention here that I have the macros run on opt
complete flag set to false.
Can you confirm if there is programming in GSB that's not transparent to the user that makes results in the manager show up as nth mode set to all
even though they may be being built with nth mode set to no trade, or does something seem to be going awry somewhere?
Also, not really a bug, but trying to keep posts to a minimum, is it a known thing that NQ is not as easy to build systems for as ES? Starting to do
some systematic work on improving NQ, but first would like to verify I'm not inventing the wheel in this regard.
Thank you for your time!  |
bug fixed in 55.80. Under testing now
Thanks received (1):
+1 OUrocketman at 2019-10-28 12:26:27
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by Daniel UK1 | | Quote: Originally posted by Daniel UK1 | Error, systems being developed on CL, systems being generated and then when i click on then, the graph is deleted and i just see 0... i have tested
different version of GSB, i have removed and loaded price files again, i have changed dates... the strange thing is that dates on the graph thats
deleted, shows 1899 12 30 .... i am using US date formats and you can see price file in the screenshot... i dont know what to do... can not develop
any systems until hat is causing this is fixed, kindly help// Thanks
|
Btw its not only on CL where it has happened |
Sorry I missed this.
Save opt settings.
As a test, reset your opt settings to default. (done by right click them)
then change settings and test again.
If that fixes it (likely) go back to an older copy of your settings that worked.
The same process can be done for app settings, but likely fauly is in opt settings.
then send me opt settings that work and fail so I can diagnose further. |
Thanks, so i first loaded another instance of GSB manager, loaded the used opt setting, then changed to default, then loaded my own opt setting again,
result... same issue... i then brought up another instance of GSB, used default... then manually changed all settings to my own... then it
worked... however i by mistake saved it as the opt setting that did not work from start, so i cant send you the error version... however... for sure
it was due to opt settings and not app settings... what caused it from start i have no idea though... thank for the help
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
I had a freak situation with a dll upgrade on my cloud that caused sever miss-match. If the version of the worker is >56.18 and < 56.32 there is a
problem.
Only my own cloud had this, but of course the cloud was free and so users got this issue.
You cant use the system. 58.18 and versions 56.32 on-wards are fine. Please except my humble apologies, as its wasted a high amount of my time, +
annoyance for some users. The problem has only occurred over the last approx 2 days.
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
OK, but its ok now? no matter what version worker is used?
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
correct, all fixed.
Peter
|
|
|
YanArnd
Junior Member
Posts: 3
Registered: 28-12-2019
Member Is Offline
|
|
Hello Peter,
My GSB Manager (56.41) currently does not work with newly downloaded price data. With those, it does never reach the "Searching..." status.
What I did:
download price data from TradeStation into /GSB/Data/Price Data
format file name accordingly (Contract.1.Minute.SomeOtherInfo.txt)
check the content and formatting of the file (same structure as in those files that work)
create contract matching the file name, save it
create price data using the contract, the file and a session; save it (with and without moving into a separate per-contract folder)
reset opt settings
reset app settings
reset price cache
choose price data created before as Opt. Prize Data
run Manager (click green button)
What should happen:
Manager should either refuse price data immediately or go into status "Searching..." after a few minutes
What happened instead:
Green button gets grayed out
Manager becomes unresponsive
a ZIP containing the price data file is created in /GSB/Data/Temp
no status change is displayed (status field stays empty)
after about 15 minutes an error message occurs (one of those "Windows Ding-Dong pop-ups" in the lower right corner which disappear after a few
seconds) - more on that below - and
status switches to "Initializing..." plus a mm:ss clock
that clock continues counting, Manager is fully responsive again, but nothing else happens (no status change, no workers starting, no results
from cloud workers).
What still works:
running Manager with default price data (which came as part of the GSB installation)
running Manager with previously imported data files (for which strategies were produced in former runs)
running GSB Standalone with the new price file
Took a look into /GSB/Exceptions and found the following errors (file modification times correspond with the attempts of running Manager with new
price data) - each Manager run attempt resulted in one out of those:
| Quote: | | System.AggregateException: One or more errors occurred. ---> System.Net.WebException: The underlying connection was closed: An unexpected error
occurred on a receive. |
| Quote: | | System.AggregateException: One or more errors occurred. ---> System.Net.WebException: Unable to connect to the remote server |
| Quote: | System.AggregateException: One or more errors occurred. ---> System.Net.WebException: The remote server returned an error: 227 Entering Passive Mode
(216,230,225,114,244,189)
. ---> System.Net.Sockets.SocketException: A connection attempt failed because the connected party did not properly respond after a period of time, or
established connection failed because connected host has failed to respond 216.230.225.114:62653 |
To me, it looks like my Manager attempts to upload my new data file to your server (216.230.225.114 without the 62653 port points to trademaid.info),
from where it probably will be distributed to the (cloud) workers - but that server currently does not handle the connection properly. I had that
problem once before already in my evaluation phase (56.04) a few weeks ago.
If my assumptions are correct, I would like to suggest that
the manager should fail early (reduce timeout?) and explicitly (and error message should stay on the screen,
in case user is AFK in the moment of failure)
admin should get informed of the issue (ping the server every n minutes from another server, send email to yourself in case of timeout or
connection close)
in case of a planned down time, a forum info would be appreciated (but I remember I saw something like that already in the forum)
This issue is kind of annoying, because it costs a lot of personal time for "tweaking this, resetting that, then see if it helped - oh no, looks
like it didn't - rinse and repeat".
Thank you in advance for looking into the problem. I will report when I notice that things are working as expected again.
---------------------
Edit 2020-01-19:
added above that GSB Standalone always works
Another observation: Some new price files worked and some don't. Tried NG and RB (with new contract data cloned from their *1330 versions) and
they did not work. Tried RR (Rough Rice, just to see what happens if something completely different is fed into Manager), and it worked just fine.
Strange...
|
|
|
Carl
Member
Posts: 342
Registered: 10-5-2017
Member Is Offline
Mood: No Mood
|
|
Hi YanArnd,
Frustrating when things don't work the way you want to.
Yesterday I started a new GSB run (GSB Manager, version 56.41) on new Tradestation data.
Everything works fine here. No errors or stalling whatsoever,
What happens when you open your opt settings and apt settings file in GSB Standalone?
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
Since about a week back, it seems something changed using the cloud ....
1- My Managers much more often than before , reaching a 1 running 1 stopped scenario... ie no workers working... i then need to restart..
2 - WF sent to cloud, example when doing 250 WF to cloud i had several scenarios last week that has never happened before, where for example 150 is
done, waiting 100, 0 running, and its like this for days, despite the fact that RM is running and all other managers is closed down and no
applications... So the managers that desperately needs workers should use all possible from RM, since it seems impossible for it to use any from the
cloud for days..
3- Real example, A managers that is running, gets 1 workers from cloud, i restart, and it gets 21... this happens a lot, that often its better to
start and stop if you get appointed to few workers from cloud... something seems off..
All above examples has been happening using latest managers released, and since about a week back issue 2 is new, so something seems to have changed
since about a week back that causing this.
Regards
|
|
|
YanArnd
Junior Member
Posts: 3
Registered: 28-12-2019
Member Is Offline
|
|
Hi Carl,
I'm glad everything works for you and thanks for asking about GSB Standalone - I forgot to mention that in GSB Standalone everything works fine,
always. (I guess that is because no external server communication takes place there.)
As of you, most of the time everything worked fine for me with new TS data in Manager, too. It seems to depend on whether Peter's server (that
216.230.225.114:62653 as of the error message) is able to handle the traffic (upload of new price data files for distribution to other cloud workers)
in the moment you run your Manager with the new data for the first time.
--------------------
Edit 2020-01-19:
| Quote: | | It seems to depend on whether Peter's server (that 216.230.225.114:62653 as of the error message) is able to handle the traffic (upload of new price
data files for distribution to other cloud workers) in the moment you run your Manager with the new data for the first time. |
It probably isn't that simple. Some files seem to work, some don't. RR worked, NG, RB and TY did not. All tried within one hour.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
HI Yanard
Your experience seems frustrating and we will get it resolved.
please do a support ticket, and we will look into the exceptions.
Exceptions are logged into a data base, and common exceptions are looked into on a regular basis.
Peter
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by Daniel UK1 | Since about a week back, it seems something changed using the cloud ....
1- My Managers much more often than before , reaching a 1 running 1 stopped scenario... ie no workers working... i then need to restart..
2 - WF sent to cloud, example when doing 250 WF to cloud i had several scenarios last week that has never happened before, where for example 150 is
done, waiting 100, 0 running, and its like this for days, despite the fact that RM is running and all other managers is closed down and no
applications... So the managers that desperately needs workers should use all possible from RM, since it seems impossible for it to use any from the
cloud for days..
3- Real example, A managers that is running, gets 1 workers from cloud, i restart, and it gets 21... this happens a lot, that often its better to
start and stop if you get appointed to few workers from cloud... something seems off..
All above examples has been happening using latest managers released, and since about a week back issue 2 is new, so something seems to have changed
since about a week back that causing this.
Regards |
I will look into this for you. WF to the cloud had a major re-write of code. The one user who had lots of problems said it was now much better. Ive
had no issues since the rebuild as well. Is the problem solid or intermittent?
|
|
|
RandyT
Member
Posts: 123
Registered: 5-12-2019
Location: Colorado, USA
Member Is Offline
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by Daniel UK1 | Since about a week back, it seems something changed using the cloud ....
1- My Managers much more often than before , reaching a 1 running 1 stopped scenario... ie no workers working... i then need to restart..
2 - WF sent to cloud, example when doing 250 WF to cloud i had several scenarios last week that has never happened before, where for example 150 is
done, waiting 100, 0 running, and its like this for days, despite the fact that RM is running and all other managers is closed down and no
applications... So the managers that desperately needs workers should use all possible from RM, since it seems impossible for it to use any from the
cloud for days..
3- Real example, A managers that is running, gets 1 workers from cloud, i restart, and it gets 21... this happens a lot, that often its better to
start and stop if you get appointed to few workers from cloud... something seems off..
All above examples has been happening using latest managers released, and since about a week back issue 2 is new, so something seems to have changed
since about a week back that causing this.
Regards |
I will look into this for you. WF to the cloud had a major re-write of code. The one user who had lots of problems said it was not much better. Ive
had no issues since the rebuild as well. Is the problem solid or intermittent? |
Peter, I think you meant to write "it is NOW much better".
I'm not seeing those problems with 56.45. I do have 3 total machines so not sure if that makes any difference.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Thanks Randy, I corrected my post.
|
|
|
Daniel UK1
Member
Posts: 470
Registered: 4-6-2019
Member Is Offline
|
|
| Quote: Originally posted by admin | | Quote: Originally posted by Daniel UK1 | Since about a week back, it seems something changed using the cloud ....
1- My Managers much more often than before , reaching a 1 running 1 stopped scenario... ie no workers working... i then need to restart..
2 - WF sent to cloud, example when doing 250 WF to cloud i had several scenarios last week that has never happened before, where for example 150 is
done, waiting 100, 0 running, and its like this for days, despite the fact that RM is running and all other managers is closed down and no
applications... So the managers that desperately needs workers should use all possible from RM, since it seems impossible for it to use any from the
cloud for days..
3- Real example, A managers that is running, gets 1 workers from cloud, i restart, and it gets 21... this happens a lot, that often its better to
start and stop if you get appointed to few workers from cloud... something seems off..
All above examples has been happening using latest managers released, and since about a week back issue 2 is new, so something seems to have changed
since about a week back that causing this.
Regards |
I will look into this for you. WF to the cloud had a major re-write of code. The one user who had lots of problems said it was now much better. Ive
had no issues since the rebuild as well. Is the problem solid or intermittent? |
Hi Peter,
Thank you, 1 and 3 seems persistent, and 2 has only happened since last week. Perhaps then its just me, will observe going ahead and let you know if
it persists..
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
Daniel, good to do a support ticket.
also before you start a manger, from dos prompt run ping xeon3.zwag.biz -t and watch that the ping times are stable.
Thanks received (1):
+1 Daniel UK1 at 2020-01-20 15:51:06
|
|
|
YanArnd
Junior Member
Posts: 3
Registered: 28-12-2019
Member Is Offline
|
|
Support ticket upload failed
Hi Peter,
Unfortunately the ticket system seems to have similar issues; tried 3 times, but I was not able to send a support ticket successfully. Compressing
support data works, then upload up until 100% works, too, then it stays for 2 minutes at 100% and finally an error pop-up informs that ticket upload
failed. Error message from /GSB/Exceptions reads
| Quote: | | System.Net.WebException: The underlying connection was closed: An unexpected error occurred on a receive. |
Can you please let me know which files you need for looking into the (original) issue (see my post above)? I would then put them onto a cloud drive
URL where you can download them.
Thanks.
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by YanArnd |
Hi Peter,
Unfortunately the ticket system seems to have similar issues; tried 3 times, but I was not able to send a support ticket successfully. Compressing
support data works, then upload up until 100% works, too, then it stays for 2 minutes at 100% and finally an error pop-up informs that ticket upload
failed. Error message from /GSB/Exceptions reads
| Quote: | | System.Net.WebException: The underlying connection was closed: An unexpected error occurred on a receive. |
Can you please let me know which files you need for looking into the (original) issue (see my post above)? I would then put them onto a cloud drive
URL where you can download them.
Thanks. |
can you let me look via teamviewer.com? email me the two codes.
|
|
|
moresi522
Junior Member
Posts: 50
Registered: 31-1-2020
Member Is Offline
|
|
bug in statistic calculation?
Watching
"Genetic System Builder, used to get better out of sample results for Tradestation & Multicharts"
link here at 14:00
I see if you select for example only 10 systems and rightclick->statistics-> Calc & Save... only that systemes are involved in calculation, but, as
you can see in the attacched images, if I select only a specific number of systems all the system are involved in calculation.
Am I doing something wrong or there is a bug ?
Thanks
 
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by moresi522 | Watching
"Genetic System Builder, used to get better out of sample results for Tradestation & Multicharts"
link here at 14:00
I see if you select for example only 10 systems and rightclick->statistics-> Calc & Save... only that systemes are involved in calculation, but, as
you can see in the attacched images, if I select only a specific number of systems all the system are involved in calculation.
Am I doing something wrong or there is a bug ?
Thanks
|
This not a bug, but deliberate.
If you want to do stats on say 10, put them say in favourites A, then do stats in favourites A
Thanks received (1):
+1 moresi522 at 2020-02-21 05:17:21
|
|
|
moresi522
Junior Member
Posts: 50
Registered: 31-1-2020
Member Is Offline
|
|
| Quote: Originally posted by admin |
This not a bug, but deliberate.
If you want to do stats on say 10, put them say in favourites A, then do stats in favourites A |
Thank you Peter
|
|
|
erlendsolberg
Junior Member
Posts: 11
Registered: 4-10-2019
Member Is Offline
|
|
Hello
There seems to be a bug in 'LibGSB.GSB_Scripts_2020_02_13'. GSB scripts for Ninjatrader will not compile.
The same error message every time:
The type name 'PrevCloseD' does not exist in the type 'NinjaTrader.NinjaScript.AddOns.LibGSB'
Using GSB 57.27 and 'LibGSB.GSB_Scripts_2020_02_13'
Regards,
Erlend
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by erlendsolberg | Hello
There seems to be a bug in 'LibGSB.GSB_Scripts_2020_02_13'. GSB scripts for Ninjatrader will not compile.
The same error message every time:
The type name 'PrevCloseD' does not exist in the type 'NinjaTrader.NinjaScript.AddOns.LibGSB'
Using GSB 57.27 and 'LibGSB.GSB_Scripts_2020_02_13'
Regards,
Erlend
|
Thank you for finding this. I will get it fixed next week.
|
|
|
moresi522
Junior Member
Posts: 50
Registered: 31-1-2020
Member Is Offline
|
|
Hi,
2 problems with EWFO.
First one:
Anytime I lunch EWFO I get "A new version..." (see image attached msg.png)
If I click Yes It downloads something and I get the same message, if click No I can notice my version is newer (see msg2.png)
Second one.
I generate a ewfo file (see msg3.png)
then I go to EWFO click on Open, from C:\GSB\Data\ewfoFiles I choose the directory but I get an error:
"An element with the same key has already been added" (see msg4.png)
I restart my pc, delete all files in C:\GSB\Data\ewfoFiles and re-generate the directory
the 7z file is the directory generated from GSB
   
Attachment: Login to view the details
|
|
|
admin
Super Administrator
Posts: 5060
Registered: 7-4-2017
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by moresi522 | Hi,
2 problems with EWFO.
First one:
Anytime I lunch EWFO I get "A new version..." (see image attached msg.png)
If I click Yes It downloads something and I get the same message, if click No I can notice my version is newer (see msg2.png)
Second one.
I generate a ewfo file (see msg3.png)
then I go to EWFO click on Open, from C:\GSB\Data\ewfoFiles I choose the directory but I get an error:
"An element with the same key has already been added" (see msg4.png)
I restart my pc, delete all files in C:\GSB\Data\ewfoFiles and re-generate the directory
the 7z file is the directory generated from GSB
|
Your error should be fixed in the updated build.
EWFO cant update as its not run in administrator mode.
This is needed for updates only. It doesnt have permision from the operating systems to write to cL\programfiles\ewfo...
The alternative is re-install into c:\ewfo\
Thanks received (1):
+1 moresi522 at 2020-03-09 04:24:13
|
|
|
| Pages:
1
..
11
12
13
14
15
..
19 |
![[*]](images/davis/default_icon.gif){kind=icon}
