More specifically, a CORS error that only appears for a certain number of embedded cells. This page will render in Safari with only nine embedded cells:
Visualizing the Principle of Least Action <http://analyticphysics.com/General%20Physics/Visualizing%20the%20Principle%20of%20Least%20Action.htm> Bizarre... On Monday, April 3, 2017 at 7:42:45 PM UTC-7, Paul Masson wrote: > > And now I finally looked at the JS console: it's a CORS error! This has > happened in the past, I believe. > > On Monday, April 3, 2017 at 7:31:24 PM UTC-7, Paul Masson wrote: >> >> Here's a couple: >> >> A Simple Deliciously Nonlinear Equation >> <http://analyticphysics.com/Differential%20Equations/A%20Simple%20Deliciously%20Nonlinear%20Equation.htm> >> Closed Orbit Conditions for Power Potentials >> <http://analyticphysics.com/Power%20Potential%20Orbits/Closed%20Orbit%20Conditions%20for%20Power%20Potentials.htm> >> >> >> On Monday, April 3, 2017 at 6:39:59 PM UTC-7, David Farmer wrote: >>> >>> >>> Whenever there is a problem with one of the browsers, usually it is >>> Safari. >>> >>> I think it is difficult/inefficient to say anything more without >>> seeing an actual page with these issues. >>> >>> Regards, >>> >>> David >>> >>> >>> On Mon, 3 Apr 2017, Paul Masson wrote: >>> >>> > I should have said "autoevaluated" in the first post, since that is >>> how I've been using >>> > SageMathCell for almost a year now. I've been careful not use any >>> individual calculations that I >>> > know will time out and try to use the fastest algorithms possible to >>> reduce evaluation time. I >>> > haven't seen any serious problems with autoevaluation before this. >>> > The number of CPUs can't be the direct cause, since the evaluations >>> just queue up and come back >>> > when a CPU is available, right? I can accept that, but when the >>> evaluations never return it's a >>> > problem. >>> > I just retested some pages with a dozen autoevaluated cells in Chrome, >>> FireFox and Safari on >>> > macOS. Evaluations eventually finish in the first two browsers but not >>> in the latter. I first >>> > noticed the problem on my iPad, where pages with more than ten cells >>> never autoevaluate. >>> > >>> > Whatever new "feature" of Jupyter is preventing Safari from >>> autoevaluting a certain number of >>> > cells is likely what is also slowing down other browsers. >>> > >>> > >>> > On Monday, April 3, 2017 at 5:23:20 PM UTC-7, Andrey Novoseltsev >>> wrote: >>> > A week ago we have updated to Sage-7.6 which means picking up >>> new Jupyter components >>> > etc. So it is possible that some change in them results in >>> slower behaviour. Are you >>> > taking about 10 embedded cells which are autoevaluated? Note >>> that we have only 4 CPU >>> > cores, so if someone opens a page with 5 automatic embedded >>> cells there is already a >>> > slow down, not to mention what happens with multiple users. So >>> in my opinion it makes >>> > sense to use automatic evaluation sparingly and make users press >>> a button whenever >>> > sensible. Restarting the server does not affect speed much and >>> is sensible only when >>> > nodes are down (in which case they usually restart >>> automatically). At the moment they >>> > are running for 3 days, and we are getting load spikes in the >>> afternoons. >>> > >>> > On Monday, 3 April 2017 15:10:01 UTC-6, Paul Masson wrote: >>> > Over the past few days, I've noticed that pages with >>> greater than around >>> > 10 embedded cells evaluate much more slowly in FireFox and >>> often fail to >>> > evaluate at all in Safari, both desktop and mobile. >>> > >>> > Has something changed recently that would affect multiple >>> evaluations? Or >>> > does the server simply need restarting? >>> > >>> > -- >>> > You received this message because you are subscribed to the Google >>> Groups "sage-cell" group. >>> > To unsubscribe from this group and stop receiving emails from it, send >>> an email to >>> > [email protected]. >>> > To view this discussion on the web visithttps:// >>> groups.google.com/d/msgid/sage-cell/1dbede41-f950-4e8f-9826-ef19f7a940af%40googlegroups.c >>> >>> > om. >>> > For more options, visit https://groups.google.com/d/optout. >>> > >>> > >>> >> -- You received this message because you are subscribed to the Google Groups "sage-cell" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-cell/1ee88f99-84c2-45a4-96a2-27f01e0b296b%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
