OFFICIAL "SERVER PROBLEMS" THREAD

[S485122]

Another small bug : the detailed reports "LL Results", will set the upper limit of the exponent range equal to the lower limit if one enters a lower limit. As soon as one modifies the range as shown in the initial view 999999937 defaults to blank. Entering 999999936 is OK though.

Jacob

[LaurV]

Originally Posted by Prime95

Something weird going on. <...> I need to understand why this is happening to create a true fix.

Unrelated but related (i.e. not server's fault, but proof not sent from computer), I repeatedly had to move the proof files around, as P95 refuses to send them, albeit there is no restriction for the network, etc. Some files just won't go out, but moving them out of the folder and back, and sometimes restarting P95, will solve the issue, and the files will be sent out after a while (like an hour or so). This "after a while" is what bothers me, because I don't have the time or patience to wait, or the memory good enough to remember to come back to it, after I moved to the daily tasks, so I won't find out till much later, next day or so, when the process repeats.

So, if that happens to other users too, then it may be an explanation of why you don't have all proof files, at least in some cases (exception is old versions which won't generate proofs, or computers which are traffic-restricted, but those are only few in percents).

At a point I was collecting over 30 proof files for PRP-CF (they are small and generated fast, but yet, that was few days of work for this slow computer!). My request, related to it, would be a button, or menu option, that says "send all proof files right now". Or similar.

Example from log file, maybe they can help finding the bug (note that in all cases the result/residue was sent hours before, in some cases days before, and the log error message persisted for long, and the proofs were not sent).

[Fri Jul 30 19:05:48 2021]
Unexpected error during p4449433.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB396213.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB437213.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB437399.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB437757.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB438633.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB439163.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB439979.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB440249.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB441291.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}
Unexpected error during pB441797.proof upload: {"error_status":401,"error_message":"Unauthorized","error_description":"JSON result not yet submitted by your user ID or proof file MD5 mismatch"}

[Zhangrc]

Another small bug:
https://www.mersenne.org/report_fact...75&tftobits=72 shows that exponents from 100000000 to 112999999 have all been TFed to at least 2^76. However, https://www.mersenne.ca/status/tf/0/0/3/10000 and https://www.mersenne.ca/status/tf/0/0/3/11000 shows that 2 exponents in 109M and 1 exponent in 111M have been TFed to 2^75.

[James Heinrich]

Originally Posted by S485122

Another small bug : the detailed reports "LL Results", will set the upper limit of the exponent range equal to the lower limit if one enters a lower limit. As soon as one modifies the range as shown in the initial view 999999937 defaults to blank. Entering 999999936 is OK though.

I don't quite understand what behaviour you're seeing that you think is incorrect. Please explain (with url and/or screenshot) what values you're entering in the form, what you get, and what you expected to get.

[James Heinrich]

Originally Posted by Zhangrc

https://www.mersenne.org/report_fact...75&tftobits=72 shows that exponents from 100000000 to 112999999 have all been TFed to at least 2^76. However, https://www.mersenne.ca/status/tf/0/0/3/10000 and https://www.mersenne.ca/status/tf/0/0/3/11000 shows that 2 exponents in 109M and 1 exponent in 111M have been TFed to 2^75.

The three exponents in question:
109882117
109882177
111959719
All of them on PrimeNet claim to be TF'd to 76, but there is no TF history recorded for 75-76 for those. I'm not sure how the server got out of sync; it's perhaps possible that a result was reported but somehow not recorded, but to be safe I think someone should re-TF these 75-76. Who volunteers?

[kriesel]

Originally Posted by James Heinrich

The three exponents in question:
109882117
109882177
111959719
All of them on PrimeNet claim to be TF'd to 76, but there is no TF history recorded for 75-76 for those. I'm not sure how the server got out of sync; it's perhaps possible that a result was reported but somehow not recorded, but to be safe I think someone should re-TF these 75-76. Who volunteers?

I'll do 111959719 to 76. Can't reserve it because it's assigned for PRP (no progress in ~2 months). Also needs a better P-1 which I'll do.

The other two are already PRP tested and certified composite.

[James Heinrich]

Originally Posted by kriesel

The other two are already PRP tested and certified composite.

Do we have other volunteers to do them anyways? If not I'll need to get George to fiddle the database back to 75 for those two.

[S485122]

Originally Posted by James Heinrich

I don't quite understand what behaviour you're seeing that you think is incorrect. Please explain (with url and/or screenshot) what values you're entering in the form, what you get, and what you expected to get.

Just try the following link it is explicit :
https://www.mersenne.org/ report_ll/ ?exp_lo=108968312&exp_hi=999999937&exp_date=2020-08-02&end_date=2021-08-01&exdchk=1&exbad=1&exfactor=1&dispdate=1&B1=
Once you click on "Gett LL data" it is changed to :
https://www.mersenne.org/ report_ll/ ?exp_lo=108968312&exp_hi=108968312&exp_date=2020-08-02&end_date=2021-08-01&exdchk=1&exbad=1&exfactor=1&dispdate=1&B1=
The upper bound of the range has been changed to be equal to the lower bound in the link and is blank on the form.

If you enter 999999936 as upper bound the query works as expected :
https://www.mersenne.org/ report_ll/ ?exp_lo=108968312&exp_hi=999999936&exp_date=2020-08-02&end_date=2021-08-01&exdchk=1&exbad=1&exfactor=1&dispdate=1&B1=

(Spaces added after each "/" of the explanation of the ling to avoid undesirable shortening.)

Thanks for your dedication !

Jacob

[Zhangrc]

Originally Posted by James Heinrich

The three exponents in question:
109882117
109882177
111959719

I will do them to 2^77, ok? I'll finish them by tomorrow.
So they are:
Factor=111959719,76,77
Factor=109882117,75,77
Factor=109882177,75,77

[James Heinrich]

Originally Posted by S485122

Just try the following link

Thanks for the example, very helpful. Please try it again now, it should work better.

[James Heinrich]

Originally Posted by Zhangrc

I will do them to 2^77, ok? I'll finish them by tomorrow.

No need to take them to 77, 76 is the target bitlevel. But thanks.