We're still experiencing the same issue as well with Dspace 6.3 and XMLUI. 
The workaround that I've been using so far is to create a custom segment in 
Google Analytics to remove any traffic where the Landing Page is "(not 
set)". That seems to bring the traffic down to where we would expect it to 
be prior to the upgrade. I've attached an image of the filter in case it's 
helpful for anyone. 

I've been digging into this recently, and I think that I might know what's 
behind it. I believe that Dspace 6 was the first version to include Event 
tracking in Google Analytics (Neha Singh, I think that this might help 
answer your question above as well). The idea is that every time a document 
is downloaded, an event is recorded in GA. If I look under Behavior -> 
Events -> Top Events, it shows a massive amount of download events (61 000 
for Sept. 2019 alone), all of which have Homepage = (not set). I originally 
wondered if this was a bug with the way that Dspace registers events, but 
when I checked the Dspace Tomcat logs, I discovered that it does appear to 
be actual download traffic. 

To test this, I picked one document which had 1847 download events listed 
in Analytics over two days in September, and found an almost identical 
number of bitstream downloads in the logs for those two days. The downloads 
came from 610 unique IPs, and there were no other downloads in September 
after the two day flurry. Geographically, 46% of the download events for 
that item came from Ashburn, Virginia (which apparently suggests that it 
likely came from an Amazon data center: 
https://www.statusbureau.com/analytics-bot-traffic-ashburn/), and the rest 
were fairly evenly spread throughout the rest of the world. 90% of the 
download event traffic was direct (not referral, social, or search). 
Interestingly, the item I chose isn't even a recent upload, it was added to 
our repository in 2015.

>From all of this, my best guess is that the extra traffic is due to direct 
downloads from bots archiving academic content on the web. The workaround I 
mentioned above works fine to remove all of the download event traffic from 
your Google Analytics results, but it would certainly be preferable to only 
filter out that bot traffic and leave only the download events triggered by 
real people. I did double-check and the option to exclude hits from bots is 
already turned on for this view. I suppose for now the best option is to 
filter out traffic from Ashburn as the article above suggests. If anyone 
comes across a better solution, I would love to hear it!

Bryson


On Tuesday, October 1, 2019 at 7:40:51 PM UTC-6, Bhavesh Patel wrote:
>
> Yes, the same issue faced by us also in DSpace 6 version.
>
> Bhavesh
>
> On Wed, 2 Oct 2019, 12:36 am Anne Lawrence, <[email protected] <javascript:>> 
> wrote:
>
>> We are using XMLUI.
>>
>> On Tuesday, October 1, 2019 at 2:58:43 PM UTC-4, Anne Lawrence wrote:
>>>
>>> We just migrated from DSpace 5.8 to DSpace 6.3 on September 11 and have 
>>> noticed the same huge spike in sessions since then. 
>>>
>>> Comparing August 16-25, 2019 to September 16-25, 2019:
>>>
>>> Sessions: 12,000 vs 197,000
>>> New visitors: 9,000 vs 162,000
>>>
>>> Source / Medium    Sessions
>>> google / organic    4,949
>>> (direct) / (none)    2,791
>>> scholar.google.com / referral    1,670
>>> search.vt.edu / referral    297
>>>
>>> vs
>>>
>>> Source / Medium    Sessions
>>> (direct) / (none)    73,233
>>> vtechworks.lib.vt.edu / referral    62,301
>>> google / organic    43,079
>>> scholar.google.com / referral    6,248
>>>
>>> The only major change to our server configuration besides the upgrade is 
>>> implementing nginx.
>>>
>>> Can anyone explain what is happening?
>>>
>>> Thanks,
>>>
>>> Anne Lawrence
>>> Repository Application Administrator
>>> VTechWorks
>>> https://vtechworks.lib.vt.edu/
>>>
>> -- 
>> All messages to this mailing list should adhere to the DuraSpace Code of 
>> Conduct: https://duraspace.org/about/policies/code-of-conduct/
>> --- 
>> You received this message because you are subscribed to the Google Groups 
>> "DSpace Technical Support" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected] <javascript:>.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/dspace-tech/89530c9d-b598-4a87-b516-000aaf562e30%40googlegroups.com
>>  
>> <https://groups.google.com/d/msgid/dspace-tech/89530c9d-b598-4a87-b516-000aaf562e30%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>

-- 
All messages to this mailing list should adhere to the DuraSpace Code of 
Conduct: https://duraspace.org/about/policies/code-of-conduct/
--- 
You received this message because you are subscribed to the Google Groups 
"DSpace Technical Support" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/dspace-tech/689c3caa-9353-47e4-837a-72834530b755%40googlegroups.com.

Reply via email to