[QUOTE=ATH;504515]When an exponent expire on the server, it does not send a message to the users client to delete the exponent.
The user can continue work on it for as long as he/she wants and his client will turn it in when it is done, it could be in a year.[/QUOTE]

There are a few interesting cases of machines that keep plugging away on expired assignments for a LOOONG time.

Consider these two (which I think are the same anonymous user). Keeps plugging away at some double-checks that are no longer necessary. They were assigned 8 years ago, expired about 5 years ago, and it's been over 6.5 years since the DC has been needed... yet it continues to report in on a near daily basis:
[URL="https://www.mersenne.org/M25413779"]M25413779[/URL]
[URL="https://www.mersenne.org/M25413671"]M25413671[/URL]

It helps that neither of those have actually made any progress... still at 0.0% after 8 years.

Then of course there are assignments that have had some progress made and are also pretty old (not like those two, but still ancient):
[URL="https://www.mersenne.org/M56487203"]M56487203[/URL]

That anon user has a handful like that. It was assigned as a first-time check over 6 years ago, expired and done by someone else nearly 4 years ago, yet it still checks in with its 26.4% done. I have no idea if it's stuck on that progress or if it's actually moving slowly... don't care. :smile:

In terms of assignments that have *not* expired and are also pretty old but still going, the winner there is:
[URL="https://www.mersenne.org/M332198957"]M332198957[/URL]

It'll be coming up on 8 years old, and since it's a 100M digit exponent it won't be expiring anytime soon. Although by the time this thing gets close to finishing, we may be expiring exponents in that range by then. Considering it took nearly 8 years to get about 1/3 of the way through, can we expect another ~16 years before it finishes? :smile:

[QUOTE=Madpoo;504444]It's too bad Kreisel did it as a PRP, because the assignment will probably finish, eventually, and now we'll have one PRP and one LL, and we'll need yet a 3rd one of either to verify. If we're going to poach, let's at least do it the same as the assignee so we're not creating extra work in case that assignment actually does finish.[/QUOTE]

Someone did an LL test so now we won't have that problem (hopefully Waki will finish before 2019-12-31)
[code][SIZE="1"]-Anonymous- Manual testing [COLOR="RoyalBlue"][B]81885841[/B][/COLOR] C 2019-01-01 02:34 0.0 255.5236 1FF0A07990679D__[/SIZE]
[/code]

[QUOTE=Madpoo;504529]There are a few interesting cases of machines that keep plugging away on expired assignments for a LOOONG time.
...
In terms of assignments that have *not* expired and are also pretty old but still going, the winner there is:
[URL="https://www.mersenne.org/M332198957"]M332198957[/URL]

It'll be coming up on 8 years old, and since it's a 100M digit exponent it won't be expiring anytime soon. Although by the time this thing gets close to finishing, we may be expiring exponents in that range by then. Considering it took nearly 8 years to get about 1/3 of the way through, can we expect another ~16 years before it finishes? :smile:[/QUOTE]Machine lifetime of ~24 years would be pretty remarkable. Eight years is getting long in the tooth.
It's unlikely that double checking or the first test wavefront will reach high enough to make a strong case for expiring it in 16 years.
The first-test milestones are advancing at 6 million/year lately (down from 8 million annually); double check milestones have been advancing at about 3-4 million/year. Using the more rapid figures, first test milestone would be at about 82+16*8=210M, double check milestone about 46+16*4=110M at the beginning of 2035.

[URL]https://www.mersenne.org/assignments/?exp_lo=82038613&exp_hi=82589933&execm=1&exp1=1&extf=1&exdchk=1[/URL]
Matthew Millar has two exponents expiring today, that are among the 30 exponents to go, in primality testing up to M51. They are each over 95% complete in LL test and stopped updating Dec 13. It would be a sad waste if they're not completed. The 97.7% completed one was also previously 87% by curtisc.
Latest current completion estimate for the remaining 30 is Feb 2 (but does not include expiration/reassign/complete times).

[QUOTE=kriesel;505709][URL]https://www.mersenne.org/assignments/?exp_lo=82038613&exp_hi=82589933&execm=1&exp1=1&extf=1&exdchk=1[/URL]
Matthew Millar has two exponents expiring today, that are among the 30 exponents to go, in primality testing up to M51. They are each over 95% complete in LL test and stopped updating Dec 13. It would be a sad waste if they're not completed. The 97.7% completed one was also previously 87% by curtisc.
Latest current completion estimate for the remaining 30 is Feb 2 (but does not include expiration/reassign/complete times).[/QUOTE]

Don't worry... I'm sure there is someone among us that has already completed them, and is quietly waiting the expiration to send them to the server... :smile:

and then there were 20...
 

Twenty to go on first-time primality tests up to M51, and two of them expire today, also with a high percentage completion before they stopped updating status.

[url]https://www.mersenne.org/assignments/?exp_lo=82038613&exp_hi=82589933&execm=1&exp1=1&extf=1&exdchk=1[/url]

[QUOTE=kriesel;506422]Twenty to go on first-time primality tests up to M51, and two of them expire today, also with a high percentage completion before they stopped updating status.

[url]https://www.mersenne.org/assignments/?exp_lo=82038613&exp_hi=82589933&execm=1&exp1=1&extf=1&exdchk=1[/url][/QUOTE]

How come an exponent has a[B] -1 day[/B] expiration expectation? Shouldn't they expire before that?

[Code][SIZE="1"]82545713 LL LL, 32.30% [COLOR="Red"][SIZE="3"][B][B]-1[/B][/B][/SIZE][/COLOR] -25 2018-11-18 2018-12-20 2018-12-21 2018-12-26 ANONYMOUS[/SIZE]
[/Code]

That column is Aaron's estimate of when they expire, not the actual variable that causes them to expire, and rarely that estimate is 1 day off for some reason.

[QUOTE=rudy235;506456]How come an exponent has a[B] -1 day[/B] expiration expectation? Shouldn't they expire before that?

[Code][SIZE="1"]82545713 LL LL, 32.30% [COLOR="Red"][SIZE="3"][B][B]-1[/B][/B][/SIZE][/COLOR] -25 2018-11-18 2018-12-20 2018-12-21 2018-12-26 ANONYMOUS[/SIZE]
[/Code][/QUOTE]

I don't know why that happened either. When I saw it, I thought maybe the daily expiration job had failed, but nope.

My best guess is that the expiration job runs at 23:25 UTC each night, not right at midnight. So there's still a chance that an assignment originally made between 23:25 and 00:00 might get an extra day reprieve. Not sure if that's what happened here but I re-ran the daily job which freed up that one as well as a couple others that would have expired later today.

and then there were 9
 

but three of them are going to take a while to get verified.

[code][SIZE="1"]There are 9 assignments

82154753 LL LL, 83.50% 1 5 2018-10-27 2019-01-23 2019-01-24 2019-01-29 -Anonymous-
82295879 LL LL, 17.40% 26 3 2019-01-20 2019-01-24 2019-01-25 2019-01-27 ATH
82342367 LL LL, 66.80% 14 11 2018-11-09 2019-01-24 2019-01-25 2019-02-04 David_Moral
82345519 LL LL, 13.90% 14 15 2018-11-09 2019-01-24 2019-01-25 2019-02-08 hsl
82375379 LL LL, 61.00% 12 4 2019-01-06 2019-01-24 2019-01-25 2019-01-28 ATH
[COLOR="Red"]82399433 LL LL, 79.40% 18 -8 2018-11-13 2019-01-13 2019-01-14 2019-01-16 phoniclynx
82541887 LL LL, 93.40% 24 2 2018-11-19 2019-01-24 2019-01-25 2019-01-26 lozcs
82547519 LL LL, 75.30% 24 8 2018-11-19 2019-01-24 2019-01-25 2019-02-01 -Anonymous-[/COLOR]
82574567 LL LL, 55.40% 19 3 2019-01-13 2019-01-22 2019-01-23 2019-01-27 Rich A
[/SIZE][/code]

Unless, of course, someone poaches them (improbable)

This is going much better than expected. 82375379 should start moving again soon.[code]There are 7 assignments
[SIZE="1"]
82154753 LL 6 5 2019-01-25 2019-01-26 2019-01-27 2019-01-31 ATH
82295879 LL LL, 72.60% 24 1 2019-01-20 2019-01-26 2019-01-27 2019-01-27 ATH
82342367 LL LL, 88.60% 12 4 2018-11-09 2019-01-26 2019-01-27 2019-01-30 David_Moral
82345519 LL LL, 14.30% 12 15 2018-11-09 2019-01-26 2019-01-27 2019-02-10 hsl[COLOR="red"]
82375379 LL LL, 61.00% 10 2 2019-01-06 2019-01-26 2019-01-27 2019-01-28 ATH
82399433 LL LL, 79.40% 16 -10 2018-11-13 2019-01-13 2019-01-14 2019-01-16 phoniclynx][/COLOR]
82547519 LL LL, 79.30% 22 6 2018-11-19 2019-01-25 2019-01-26 2019-02-01 -Anonymous-[/SIZE][/code]