OFFICIAL "SERVER PROBLEMS" THREAD

[Prime95]

Was that in reference to PRP tests?

[James Heinrich]

Originally Posted by Prime95

Was that in reference to PRP tests?

Looking at my notes again, that is possible it meant to refer to PRP and not LL.

[1997rj7]

This is now working OK for me.

[Prime95]

@LaurV: Submit your results,json,txt file instead of screen output

@Everyone: Please submit JSON results whenever possible. It contains more data.

[garo]

Just looked at some really old prime.log files and it confirms my suspicion that Prime95 has been sending shift counts since at least 2007 and most likely a lot longer.

[James Heinrich]

We have discovered LaurV's problem, we're working on solving it, I'll have a tweak on the manual results page probably tomorrow. For everyone else everything should already probably be working as normal.

[Prime95]

@LaurV: Upgrade prime95 to get Gerbicz and JSON

[ATH]

https://www.mersenne.org/primenet/

is showing "PrimeNet Activity Summary 2020-05-13 12:00 UTC" instead of the correct date and time.

[LaurV]

Originally Posted by James Heinrich

If you want to re-run that exponent on the same system but with v29.8b6 we'll give you two candies when it completes.

That is a bit unfair, to say it softer. I ran the test, it matched the first time residue, the We4 result is reproducible (I didn't tweak with it), you just have to put the result into the input of the checksum routine of the version I use, to see that you get the same We4, however the server won´t accept it, because you (general you referring to the gods of PrimeNet) broke the compatibility with the old versions, and now I have, first, to upgrade to a SLOWER version (ETA 6 days and 9 hours, as opposite to the 5 days and 4 hours, which is A LOT SLOWER), then second, run it AGAIN to obtain the JSON file? Just to report a DC which I KNOW is right, and which there is some probability that the server won't accept it again, no matter what file I report, because it comes from the same user?

Be my guest...

I will shot three holes through my monitor before that...

Well... as you probably guessed from the percents I gave above, I installed the 29.8b6 and I started another run, 2.6% done, 6 days and about 9 hours to go... What the hack, it is curfew...

[Uncwilly]

Originally Posted by ATH

https://www.mersenne.org/primenet/

is showing "PrimeNet Activity Summary 2020-05-13 12:00 UTC" instead of the correct date and time.

Originally Posted by lycorn

PrimeNet Activity Summary Report appears to be stuck at 2020-05-13 12:00 UTC.

Correction: it is just the header. The figures seem to be updating properly.

Originally Posted by Viliam Furik

I can confirm, (here) it is stuck on 12:00, but on all top producer lists it's working properly.

That has been noted before, but no word from the powers that be.

[LaurV]

Originally Posted by LaurV

(ETA 6 days and 9 hours, as opposite to the 5 days and 4 hours, which is A LOT SLOWER)

Well, that was not fair either, after about 4 hours we are in about 4% and it says ETA 5 days, 1 hour. It seems that immediately after GC runs, P95 (newest) shows a longer ETA. Then the following lines, the ETA drops to "normal". It is not a lot slower, just marginally slower, including GC. (GC runs by default every 1M iterations, and screen output is also default to 100k).
We are good. Going to bed (00:32 here, or 12:32 AM for the people living behind the pool)