Indeed, something is still off, factorizations are being erased, obviously they shouldn't be.
I found the error where previously-factored numbers were being re-factored (and then erased), still working on the why it's getting an empty factorization.

edit: ok, I think I found the conflict and where I went wrong. I have rebuilt the 15000 or so k values with null factorizations, and added an alert if it seems about to clobber a factorization again. If you spot something amiss please do let me know right away.

edit2: I'm somewhat confused. The code works perfectly when I test it, but not when I run it. Possibly a permissions thing? Trying to investigate.

edit3: OK, I think I've found [i]how[/i] it was happening. [i]Why[/i] it's happening is a bit of a [url=https://www.mersenneforum.org/showthread.php?p=565068]mystery[/url], but I think I can work around it now.

Don't know if this is related, but there is some kind of fluke in the attached screenshot.

The first result (14087947) has a listed B1,B2 that doesn't reflect the actual B1,B2 we see when we click on the [URL="https://www.mersenne.ca/exponent/14087947"]link for 14087947[/URL].

[QUOTE=masser;565100]Don't know if this is related
The first result (14087947) has a listed B1,B2 that doesn't reflect the actual B1,B2 we see when we click on the [URL="https://www.mersenne.ca/exponent/14087947"]link for 14087947[/URL].[/QUOTE]I don't believe it's related. This particular exponent did have P-1 at the bounds show back in 2001, and then another one with good bounds in 2020-Aug, but the master bounds table for that exponent wasn't (for reasons unknown) updated with the new bounds. I'll have to investigate to find out why, whether it was a one-off data error or a systematic problem. I'll post back when I've found out which.

[QUOTE=masser;565100]The first result (14087947) has a listed B1,B2 that doesn't reflect the actual B1,B2 we see when we click on the [URL="https://www.mersenne.ca/exponent/14087947"]link for 14087947[/URL].[/QUOTE]Thanks for reporting this. In a relatively small number of cases (about 3k exponents) the master list of P-1 bounds did not match the best-bounds P-1 value. Many of these were very-close differences ([URL="https://www.mersenne.ca/exponent/15436093"]example[/URL]), and cases where the "better" P-1 actually had a lower B1 (but higher B2). I have walked through 20 years of P-1 history and made the adjustments where appropriate.

Another (minor) primenet data glitch, nightly update will be delayed until I get the corrected data (which may not be until tomorrow morning).

edit: glitch fixed, data has been synched.

One more (hopefully small) project to add to the queue)
[URL="https://www.mersenneforum.org/showpost.php?p=566269&postcount=3"]https://www.mersenneforum.org/showpost.php?p=566269&postcount=3 [/URL]

It seems that mersenne.ca hasn't been updated since December 18, 23:59:59. I reloaded the page as much and as many times as I could, but nothing.

I just noticed this about 5 minutes ago. I get emailed when something unexpected blocks the import, but for reasons-unknown some of my emails haven't been delivered for the last 5 days. Just some, not all, and not just one account, I don't really understand it. But short story is that I didn't know it wasn't working. Looking into it now.

edit: forgot to post back that I fixed it last night already.

Trouble with year end roll over?
 

mersenne.ca not updated since 12/31

Things should have run mostly correctly yesterday, but today Aaron has changed the PRP Certification results format in the nightly data sync. It's a minor change I need to make... but today I discovered bad sectors on the HDD that holds my web stuff. It's still usable (and the backup drive is healthy) but it's currently offline doing a scan. I will fix the import as soon as feasible.

edit:[quote]CHKDSK is verifying file data (stage 4 of 5)...
Windows replaced bad clusters in file 77111
234480 files processed.
File data verification completed.
CHKDSK is verifying free space (stage 5 of 5)...
77999833 free clusters processed.
Free space verification is complete.
[b]Adding 2 bad clusters to the Bad Clusters File.[/b]
CHKDSK discovered free space marked as allocated in the
master file table (MFT) bitmap.
Correcting errors in the Volume Bitmap.
Windows has made corrections to the file system.

5723036 MB total disk space.
5417763 MB in 151071 files.
84936 KB in 13970 indexes.
32 KB in bad sectors.
516211 KB in use by the system.
65536 KB occupied by the log file.
311999328 KB available on disk.[/quote]

I made a small fix to code yesterday that didn't quite work right at end-of-year rollover, and my fix was slightly flawed resulting in the false display of "data as of 2020-12-31". I've fixed the fix and (I think) all is as it should be now. I'm sure you'll let me know if it's not. :smile: