Errors, bugs, questions - page 748
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
The application has been accepted. Several bugs have been fixed in the tester, leading to a discrepancy between the optimisation and single test results.
Therefore, it is up to you (as the author of the EA) to provide evidence that the new build has not been entered (or exited) correctly. The sooner you do this, the better for everyone. It is much harder for us than for you, the author.
I've run optimization with a dummy parameter (I've set 1000 passes).
As a result, Cloud proudly reported the following:
locals 0 tasks(0%), remote 0 tasks(0%), cloud 512 tasks(100%) ?? ?
Total passes 484 (successful 457 passes)
Of these:
15 times result 5605.09 (as if agents from previous build worked, matches single test of previous builds - 642, 619)
442 times 7175.27 (matches with the single test of build 665)
A lot of messages "cannot synchronize history", "cannot initialize expert", "no memory", "expert rejected by MQL5 Cloud Network due to endless loop".
build 655
remote agents refuse to take part in optimization for some symbols.
For example: for EURUSD - all OK!, for EURJPY - optimisation only on local agent.
The history of the tools involved during the testing period, as far as I understand, has not changed.
There was a long-standing problem with "broken" containers that behaved as if they were intact during data sync (even the cloud agent found one such container on my computer, which helped localise the problem). Such containers are now killed and replaced on resync. Therefore, one could argue that the history may have changed.
"Wait for the foam to settle and demand a refill". Let's wait a couple of days.
build 655
remote agents refuse to take part in optimization for some symbols.
For example: for EURUSD - all ok!, for EURJPY - optimisation only on local agent.
Do you have access to remote agent logs?
If yes, please provide them to the service desk.
I would like to ask if the issue will be resolved to skip weekends in the future, as per application#318662.
I put the June 18 line on the chart. But it is 6 working days and 36 bars before that date.
I personally want to put the Fibo time zones in the future and see the date.But it will be wrong.
Or the futures expiry date will be June 18.But it will not be where the line is.It will be where the rectangle ends...
I would like to know if the issue will be resolved in order to skip weekends in the future.On application#318662.
Let's look at the picture of the other terminal... Well, it is realistic to estimate the time development of this or that variant, when I can see the exact location of the date...
At the top of the chart there is June 18. And there are really 36 bars to it... And where is our 18th...
There was a longstanding problem with "broken" containers that behaved as if they were intact during data sync (even the cloud agent found one such container on my computer, which helped localise the problem). Such containers are now killed and replaced on resync. Therefore, one could argue that the story may have changed.
"Wait for the foam to settle and demand a refill". Let's wait a couple of days.
In this context, "container" is what?
Do I understand correctly that if, after the containers are replaced, run the tester of build 642, it should show the same result as build 665?
In this context, "container" is what?
Am I correct in assuming that if, after the containers are replaced, you run the tester of build 642, it should show the same result as build 665?
A container is a repository for a portion of historical data.
Yes, you understand correctly. The problem was on the test agents, not the tester itself.
The container is the repository of a portion of the historical data.
Yes, you understand correctly. The problem was on the test agents, not the tester itself.
Something's not working:
I delete history for 2011 in Tester (Tester\bases\metaquotes-demo\history\...\2011.hcs) and Metatrader (bases\metaquotes-demo\history\...\2011.hcs)
I run the test (several passes) in build 655 - history is downloaded, deleted files appear, I get the result 7175.27.
I change the tester to build 642, run the test. The history for the year 2011 does not change. I get a result of 5605.09 !?!
Build 642 logs 11752200 ticks, 665 logs 11752140.