OFFICIAL "SERVER PROBLEMS" THREAD

[Madpoo]

Originally Posted by Runtime Error

Hey folks, hope this is the correct place to ask regarding www.mersenne.org, but I keep seeing this when I try to visit the website.



This has been happening on/off for a few days, but currently I cannot get it to work. I've tried a few different browsers and configurations, to no avail. Thank you.

I've been chasing this infinite redirect loop for a few days and finally got a chance to dig in to it.

I think it's better now. I had to totally redo the rules that handle going from mersenne.org to www.mersenne.org, and from http to https.

Ordinarily it wouldn't be so difficult, but there are exemptions to those rules. Like the old v4 clients that (incredibly) still exist out in the wild. They hit an old page at mersenne.org and don't like being told to redirect (and don't support https either).

Plus there are exemptions to the HTTPS rule like program downloads (which I think should be https, but people complained), certain pages used by misfit, and the new PRP proof URLs for uploading/downloading the large cert files.

Anyway... TLDR is that some of the rules had started jumping on each other and weren't playing nice. Should be all nice and streamlined now, assuming I didn't miss anything in the process.

[Mark Rose]

How many active v4 clients are there?

[James Heinrich]

Originally Posted by Mark Rose

How many active v4 clients are there?

And perhaps more important, how much useful work do they contribute?

[Chuck]

Originally Posted by Mark Rose

How many active v4 clients are there?

I wish there was a report available that showed how many of each version of Prime95 are in use.

[Runtime Error]

Thanks everyone for the replies. It worked for me just now. I appreciate it!

[S485122]

Originally Posted by James Heinrich

And perhaps more important, how much useful work do they contribute?

Your question implies another : should GIMPS be restricted to big producers that update frequently or should it retain its inclusion of whomever has some cycles to spare.

Jacob

[S485122]

Originally Posted by Chuck

I wish there was a report available that showed how many of each version of Prime95 are in use.

That is a wish I have expressed in my long post https://www.mersenneforum.org/showpo...&postcount=213. If the data about the version is not stored with LL results for exponents in the database, it would be the result lines as reported to the server that provide the data.

Jacob

[masser]

There appears to be a problem with the factors found report:

This page is returning nothing for me:
https://www.mersenne.org/report_fact...=21&max_len=99

[Prime95]

Originally Posted by masser

There appears to be a problem with the factors found report:

This page is returning nothing for me:
https://www.mersenne.org/report_fact...=21&max_len=99

Fixed, thanks.

[LaurV]

Happy Birthday Server !



(in spite of the fact that now it may be a backup server, and that most forum members forgot the reason, or never knew it... )

[kriesel]

Not that it will affect the wavefront, or 100Mdigit, or considerably higher, but:

There is a clash between the maximum bit level that manual tf will allow (85), and the GPU72 listed final bit level at mersenne.ca (86) for high exponents in the mersenne.org range. See for example https://www.mersenne.ca/exponent/993112609 and https://www.mersenne.org/manual_gpu_assignment/