Disappearing peer reviews

[sasutton]

We are using OCS version 2.1.2.1. This is the first time we've used the OCS although we have used the OJS for a number of years. We just entered the process of accepting/declining submissions. Acceptances are working fine, but we have a serious problem with declines. Here is what is happening:

• Track Director is looking at the Review page for a submission where the completed peer reviews appear with their recommended decisions;
• Track Director selects "Decline submission" from the drop-down menu and clicks "Record Decision";
• The screen refreshes and the peer reviews have vanished (see attached .jpg screen image);
• IF the Director clicks to "Notify Author", he/she gets the correct decline email but trying to include reviews into the email fails;
• IF the Director nevertheless sends the decline email, it does not show up in the email history for the submission;
• The declined submission has been moved to the archive where it shows as "declined" in the listing;
• Clicking on the submission title in the archive takes the Director to the same screen as included in the attached .jpg;
• The peer reviews appear to be lost;

Obviously, I am desperate and quite distressed over the missing reviews on three declined submissions before I was notified of the problem.

Stuart

Missing-Reviews.pdf

Screenshot of submission after reviews lost.

(389.4 KiB) Downloaded 26 times

[jmacgreg]

Hi Stuart,

That sounds seriously worrying; I haven't heard of this happening before, and I can't replicate it. What is your submission materials method (Scheduled Conference Setup Step 2.1); and are any errors being reported to the server's error log?

Cheers,
James

[sasutton]

James, Set-up 2.1 Submission Materials is "Abstract and proposal together". I have attached a .jpg of 2.1. I'll have to ask the technical folks about the server's error log and get back with you. I am limping along processing declines by hand but am in real trouble with the four "declines" already processed since I have no access to the reviews attached to them. Is there some way to try and get to specific reviews for a submission in the data store?

Stuart (trying to take deep breathes)

[ramon]

Hello Stuart,

We've experienced a similar problem with our OCS 2.1.2.1 install.
The database inconsistency was caused by the change in the review process after submissions were already sent.
If this is your case, you'll have to manually fix the inconsistency in the database, and the PKP team will be more reliable in providing a more expert answer.

I would also suggest checking bugzilla for patches against this version, which may have been released to address version related problems.

[sasutton]

Ramon, that is quite likely the cause. We did have to change the process (not once but twice) after submission had already begun coming in. Let's see what James has to say. Thankfully, this round is nearly complete and I'll survive other than trying to find the peer reviews for 4 submission.

Stuart

[asmecher]

Hi Stuart,

I'd suggest working with the most recent release of OCS. In the past, the "submission process" setting was system-wide and changing it in a conference that was already running could result in "marooned" submissions -- i.e. the review information for that submission wasn't consistent with the setting. If you're able to muddle through the current set of submissions as is, that's great -- and once you update to a newer release, the setting will stay with the submissions themselves so that changes to the setting won't result in this kind of problem.

Regards,
Alec Smecher
Public Knowledge Project Team

[sasutton]

Thanks, Alec, I can muddle through with cutting and pasting reviews. I am, however, concerned that 7 submissions with declines have been disconnected from their reviews. Is there any way to manually determine which reviews go with these "marooned" submissions. I told the authors I would try.

Stuart

[asmecher]

Hi Stuart,

The easiest way to do this is probably to look in the database directly; look for the reviews in the review_assignments and paper_comments tables (by paper_id). The content should still be there.

Regards,
Alec Smecher
Public Knowledge Project Team