Spontaneous corruption of database MC12

Questions about MultiCharts and user contributed studies.
wilkinsw
Posts: 662
Joined: 21 Apr 2013
Has thanked: 154 times
Been thanked: 104 times

Spontaneous corruption of database MC12

Postby wilkinsw » 20 Jun 2019

Hi,

MC, just to let you that for the first time ever I've suffered a spontaneous corruption of my database.

For some reason, for all instruments, data stopped being stored at various times today and yesterday: If I copy and paste a chart it will show end times that are several hours before the chart that was copied and I can no longer open a chart with the most up to date data.

No idea how to reproduce; but is probably the most significant error I've come across in the several years of using MC. No popups warnings etc... just new charts with wrong data.

Please have a think as to how this could've happened.

Thanks.

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Re: Spontaneous corruption of database MC12

Postby TJ » 20 Jun 2019

For the current MC version,
all the operating data are stored in RAM.
The data is flushed to your harddisk when you shut down MC.

As the first debug step, I would run the memory test to check the memory integrity.

sptrader
Posts: 742
Joined: 09 Apr 2010
Location: Texas
Has thanked: 483 times
Been thanked: 274 times
Contact:

Re: Spontaneous corruption of database MC12

Postby sptrader » 20 Jun 2019

Another important thing besides what TJ said- make sure that after exiting MC, check "processes" (ctrl-alt-del) and make sure that all of MC's processes are closed before turning off or rebooting your computer. It can take several minutes, depending upon CPU speed, HD, RAM etc.
Last edited by sptrader on 07 Jul 2020, edited 1 time in total.

hughesfleming
Posts: 275
Joined: 22 Apr 2014
Has thanked: 70 times
Been thanked: 72 times

Re: Spontaneous corruption of database MC12

Postby hughesfleming » 03 Jul 2019

Must be the season. Just lost mine as well. Very frustrating.

regards,

Alex

User avatar
Smoky
Posts: 507
Joined: 03 Dec 2010
Location: Thailand
Has thanked: 97 times
Been thanked: 115 times

Re: Spontaneous corruption of database MC12

Postby Smoky » 03 Jul 2019

be carefull, when use on slow PC sometime MC make long time to close All, don't shutdown your PC quickly after closing MC...

wilkinsw
Posts: 662
Joined: 21 Apr 2013
Has thanked: 154 times
Been thanked: 104 times

Re: Spontaneous corruption of database MC12

Postby wilkinsw » 09 Jul 2019

It was a strange case. It didn't happen during a restart. I noticed it when opening a new chart. Existing charts had the correct data; so it couldn't have been a datafeed disconnect. Could only solve by deleting the effected instruments, reimporting from a backup, then restarting.

Maybe TJ is right when suggesting it could be corrupt memory. So maybe a hardware issue.

hughesfleming
Posts: 275
Joined: 22 Apr 2014
Has thanked: 70 times
Been thanked: 72 times

Re: Spontaneous corruption of database MC12

Postby hughesfleming » 28 Jun 2020

I have another database / IQFeed tick data problem. On Saturday something broke and it is affecting the @ES#C contract and not the September. Back tests break with bar magnifier set to tick. Should I bring this up with IQFeed as well? Fortunately the @ES# is working normally. This is what it looks like. What could have caused this?

regards,

Alex
broken.JPG
(231.87 KiB) Not downloaded yet

User avatar
TJ
Posts: 7740
Joined: 29 Aug 2006
Location: Global Citizen
Has thanked: 1033 times
Been thanked: 2221 times

Re: Spontaneous corruption of database MC12

Postby TJ » 28 Jun 2020

Have you checked the scaling? . . . Make sure it is attached to the dataseries.

wilkinsw
Posts: 662
Joined: 21 Apr 2013
Has thanked: 154 times
Been thanked: 104 times

Re: Spontaneous corruption of database MC12

Postby wilkinsw » 28 Jun 2020

Plot your tick data and report back. It looks like you’ve got tick data with very high values for some reason. Maybe something to do with the rollover, not sure.

Then delete the symbol. Press ctrl+b (i do that but not sure if it’s obligatory). Then readd the symbol.

Again plot the tick data and report back.

hughesfleming
Posts: 275
Joined: 22 Apr 2014
Has thanked: 70 times
Been thanked: 72 times

Re: Spontaneous corruption of database MC12

Postby hughesfleming » 28 Jun 2020

Thanks @TJ and @ wilkinsw,


I just loaded the strategy on a backup machine and that is working. I should be able to see what is different between the two. The odd thing is that it was working fine on Friday and strangely broken on Saturday. I will check the scaling and see if they match and if not, I will just reinstall from backups.

Kind regards,

Alex

Sallymeow
Posts: 10
Joined: 28 Nov 2018
Has thanked: 2 times

Re: Spontaneous corruption of database MC12

Postby Sallymeow » 07 Jul 2020

I have a similar problem just like the case of hughesfleming. Look at the picture attached. The entry points and exit points are outside the bars buying and selling at unrealistic prices. Very odd thing! It happen when I use the bar magnifier (1min resolution) to backtest the strategy. Still finding solution to solve it.
Attachments
Using bar magnifier.png
Problem with bar magnifier
(155.92 KiB) Not downloaded yet

hughesfleming
Posts: 275
Joined: 22 Apr 2014
Has thanked: 70 times
Been thanked: 72 times

Re: Spontaneous corruption of database MC12

Postby hughesfleming » 07 Jul 2020

@Sallymeow,

I tried a couple different things to fix the problem which didn't really help. I eventually deleted everything including the database and reinstalled. It is working again normally now but this is probably not the answer you were hoping for.

regards,

Alex


Return to “MultiCharts”