search

open-cogsci / omm-server

Open Monkey Mind server
Apache License 2.0
0 stars 1 forks source link

Queued at position counter is behaving strangely #117

Open nclaidiere opened 2 years ago

nclaidiere commented 2 years ago

On the OMM server there is a problem with queued at position. The queued at position counter is displaying various positions for the same number in the queue. It also stays the same when the participants finish experiments, so say they have finished expt 2, they stay queued at position 3, shouldn't that update?

bug-queued-at-pos-2 bug-queued-at-pos-1 bug-queued-at-pos-3

dschreij commented 2 years ago

The queue position actually does not represent the position of the participant in the experiment that is currently viewed, but rather represents the number of other experiments that the participant first is to finish (either with a higher priority or an earlier creation date) before he/she will start on the current experiment.So it is entirely possible that multiple participants have the same queue position.

nclaidiere commented 2 years ago

of other experiments that the participant first is to finish (either with a higher priority or an earlier creation date) before he/she will start on the current experiment

Yes, I understand that, but it's not what's happening. If you look at the 3 pictures above all the participants are in the same state on the first experiment, "first to run", then on the second experiment all the priorities are the same but Arielle is queued at position 2, the others at position 1 and this goes on in the third experiment.

dschreij commented 2 years ago

I see what you mean! This is only the case for Arielle here, am I correct? The queue positions of the other participants look ok. And it's entirely certain Arielle isn't also assigned to another experiment which the others aren't?

nclaidiere commented 2 years ago

This is only the case for Arielle here, am I correct?

No it's just Arielle in the snapshot, but it happens to several participants and I have tripled checked that they are all identical. Furthermore there are no other experiments with priorities 1, 2 and 3. So I have no idea why it is doing that.

dschreij commented 2 years ago

Just to rule out this possibility: are you the only one using the system at the moment? There may be another user that has an experiment to which Arielle is assigned and which has a higher priority. These queue positions are not only for your own positions, but for all experiments in the database, also those of other users.

nclaidiere commented 2 years ago

I am the only user, this happens on the monkey server but also locally, on my laptop.

dschreij commented 2 years ago

I looked into this and have trouble reproducing it. I can see from your screenshots that the issue exists, but when I try to get the same behavior on my local system, everything appears to function as it should. If I can't reproduce this locally, it is very difficult to pinpoint the problem. Can you maybe provide some exact steps (e.g. clicks) that I need to make to bump into this problem?

nclaidiere commented 2 years ago

Same here, I have no idea on how to trigger it, the best would have been to look on the server when the error was present but I think sebastiaan wiped that out.

kristian-lange commented 3 weeks ago

Hi @nclaidiere, Hi @smathot, since this issue is currently not reproducible I'd recommend to put it on hold until it reoccurs again. What do you think?

nclaidiere commented 3 weeks ago

Hi kristian, this problem was never solved and is still present, here is a screenshot of the position counter on the server: image The counter is supposed to be at 1 for all the monkeys... I think one problem is that it takes into account archived and finished experiments...