I was using ODK Survey to conduct a small survey. While sending data from
the field I am getting a duplicate record on the ODK Aggregate server. I am
using ODK Aggregate 1.4.4 and MySql for the database on a local server
which have a public IP address.
I tried to see the source code of Survey and every time a record is resent
a new instance ID is generated so that meta-instanceID should be different.
But the duplicate records that I am getting have the same meta-instanceID
and meta-rowID. I called the data collectors and they told me that they are
sending a record only once. Please see the attached file which contains the
duplicate data that I exported from ODK Aggregate.
The generation of new instanceIDs with each submission was added at rev
122. Prior to that, the submission instanceID would be the rowID, and could
therefore collide when submitting records to ODK Aggregate.
···
On Fri, Oct 16, 2015 at 6:06 AM, Pomi Daniel wrote:
Hello ODK Developers,
I was using ODK Survey to conduct a small survey. While sending data from
the field I am getting a duplicate record on the ODK Aggregate server. I am
using ODK Aggregate 1.4.4 and MySql for the database on a local server
which have a public IP address.
I tried to see the source code of Survey and every time a record is resent
a new instance ID is generated so that meta-instanceID should be different.
But the duplicate records that I am getting have the same meta-instanceID
and meta-rowID. I called the data collectors and they told me that they are
sending a record only once. Please see the attached file which contains the
duplicate data that I exported from ODK Aggregate.
Thank you very much Mitch for your response. The thing that I wasn't able
to understood was if I submit one record from ODK Survey rev 126 to ODK
Aggregate I get two or more records with the same instanceID on the server
for that single record. Why is that happening?
···
On Friday, October 16, 2015 at 8:51:02 PM UTC+3, Mitch wrote:
>
> The generation of new instanceIDs with each submission was added at rev
> 122. Prior to that, the submission instanceID would be the rowID, and could
> therefore collide when submitting records to ODK Aggregate.
>
> https://github.com/opendatakit/opendatakit/wiki/Tools-Suite-Release-Notes
>
>
>
> On Fri, Oct 16, 2015 at 6:06 AM, Pomi Daniel <pom...@gmail.com > wrote:
>
>> Hello ODK Developers,
>>
>>
>> I was using ODK Survey to conduct a small survey. While sending data from
>> the field I am getting a duplicate record on the ODK Aggregate server. I am
>> using ODK Aggregate 1.4.4 and MySql for the database on a local server
>> which have a public IP address.
>>
>>
>> I tried to see the source code of Survey and every time a record is
>> resent a new instance ID is generated so that meta-instanceID should be
>> different. But the duplicate records that I am getting have the same
>> meta-instanceID and meta-rowID. I called the data collectors and they told
>> me that they are sending a record only once. Please see the attached file
>> which contains the duplicate data that I exported from ODK Aggregate.
>>
>>
>> What could be the possible cause and solution?
>>
>>
>> Many thanks,
>>
>>
>> Pomi
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "ODK Developers" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to opendatakit-developers+unsubscribe@googlegroups.com
>> .
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
>
> --
> Mitch Sundt
> Software Engineer
> University of Washington
> mitche...@gmail.com
>
The 'instanceId' of an ODK Survey row is sent to the ODK Aggregate 1.x
legacy form in the field.
Every time you submit an ODK Survey row, a new random UUID will be sent
to the ODK Aggregate 1.x legacy form as the field. That
field is used to de-duplicate and reconcile submissions.
By always using a new random UUID for that field, this enables editing and
revising earlier ODK Survey rows and submitting the updates you have made
to them -- thereby having each edit to that one row appear as a new
submission in the the ODK Aggregate 1.x legacy form.
If the instance-upload screen's logic is working correctly, you should only
see the rows that have changes since they were last updated, so you
shouldn't get too many false updates. If you are getting a lot of
submitted rows with identical values in all the columns (except for the
instanceID), let me know; that is a bug.
You can find the most recent update for a by grouping the
submission records by and looking for the one with the most recent meta-data-marked-as-complete date. Alternatively, you could use the
field (which reports the timestamp of the save
on the device).
···
On Fri, Oct 16, 2015 at 1:11 PM, Pomi Daniel wrote:
Thank you very much Mitch for your response. The thing that I wasn't able
to understood was if I submit one record from ODK Survey rev 126 to ODK
Aggregate I get two or more records with the same instanceID on the server
for that single record. Why is that happening?
On Friday, October 16, 2015 at 8:51:02 PM UTC+3, Mitch wrote:
The generation of new instanceIDs with each submission was added at rev
122. Prior to that, the submission instanceID would be the rowID, and could
therefore collide when submitting records to ODK Aggregate.
On Fri, Oct 16, 2015 at 6:06 AM, Pomi Daniel pom...@gmail.com wrote:
Hello ODK Developers,
I was using ODK Survey to conduct a small survey. While sending data
from the field I am getting a duplicate record on the ODK Aggregate server.
I am using ODK Aggregate 1.4.4 and MySql for the database on a local server
which have a public IP address.
I tried to see the source code of Survey and every time a record is
resent a new instance ID is generated so that meta-instanceID should be
different. But the duplicate records that I am getting have the same
meta-instanceID and meta-rowID. I called the data collectors and they told
me that they are sending a record only once. Please see the attached file
which contains the duplicate data that I exported from ODK Aggregate.