Discussion related to cado-nfs

[canny georgina <cannysiska@gmail.com>]

Have tried tasks.sieve.allow_largesq = true, thank you for the suggestion, i think this one is more flexible than enlarging lpb0 and lpb1 (because i can't monitor the computation 24/7).

But, I do change the largesq option by interrupting the computation first (using Ctrl+C) then re-input the snapshot using tasks.sieve.allow_largesq = true option. Will it take any effect then?

However, I'm curious whether it's possible to resume the computation of a failed workunit. I mean, this "not-enabled-allow-largesq" option has made the computation stop in the middle (because of so many failed workunits) and when I tried to resume its snapshot using same trick (by adding tasks.sieve.allow_largesq = true option) , it failed to resume.

2018-07-23 21:57 GMT+07:00 Pierrick Gaudry <pierrick.gaudry@loria.fr>:

> well, I run it as a root because I need to install some dependencies first
> before using cado, so it feels kinda bothersome to reenter password
> everytime It needs root permission. But, will it be unsafe, then?

It depends if you trust the cado-nfs developpers ;-)

Even if we are honest, we could have made a mistake in the code that
implies doing "rm -rf /" or something similarly bad for the system, if
you gives weird, untested parameters.

At your own risk...

Regards,
Pierrick