Want to take part in these discussions? Sign in if you have an account, or apply for one below
Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.
We have been given early warning by CMU that the building the nLab is in will be out of power all day on the 11th of August. We have enough time to put up a warning message on the nLab, but I wonder if we should use this as a spur to setup a second server on a virtual machine somewhere? We can use a pay-per-use machine, and then decide whether or not we keep it up after the downtime.
There will be quite a bit of work to do to set this up, so I will need a decision soon. If we go for it, we will also need to finance it. Maybe those willing can divide the sum between them, it should not be too much. I am willing to contribute.
If no-one objects too strongly, I would suggest on this occasion that we use Amazon, simply because I am already familiar with it. We can probably get by with a single EC2 instance of size something like t2.large (I can experiment a bit beforehand to get the right size machine). If we have it up for 48 hours, such a machine would cost 4.4544 dollars. I can pay for this myself, but it’d be nice if a few others also commit to paying, just so that we can split the costs if needed (the actual cost will be a little bit more, because I will need to test, and one pays every time one switches it on and off). We would also need a way to switch routing. To keep costs down, we could simply use an nginx server on the same EC2 instance that we hire, and manually switch. The more robust approach (but maybe too much for now) would be to use a load-balancer (which will come with some small additional costs).
There would be some side-benefits of doing this. One being that I will need to make the nLab easily deployable, e.g. build locally.
4.4544 dollars
Could you say that again? Is it four thousand dollars plus something?
We could of course ask whether Steve would be willing to foot the bill from the HoTT grant as well. But as I say, we’d need a quick response. Maybe it’d be easiest to finance it ourselves for now. Or if people don’t feel it’s right, we can just accept the downtime and put up a warning message.
Re #2: It’s literally 4 dollars and 45 cents! (Of course I can finance that myself, I am asking more if people are in principal willing to do this, or in case costs turn out to be a bit more).
All right, sorry. I am willing to contribute.
Yes, I don’t think it’s worth bothering about trying to get the grant to pay $4. I’d be happy to chip in part of that if needed. If you are really willing to do the work on setting this up in time for the outage, I think it would certainly be worth doing.
OK, great, thanks! I cannot promise I’ll have enough time, but I’ll give it a go. We’ll no doubt learn something anyhow. The cost might well be more than 4 dollars, but I’ll keep track of the cost while I’m working on it.
With the work on the new renderer, I unfortunately did not get the chance to look into this. I’ll put a warning message up instead.
One of the things that is troubling, me, though, is the purely practical matter that one needs to have a credit card registered to the account on AWS, for example, which is billed monthly. Whilst I am willing to make a financial contribution, I am uncomfortable with using a personal credit card in this way. Is there any way that we can, with a minimal amount of hassle, find some kind of financial arrangement which allows us to set up some kind of bank account for the nLab which we can pay into?
Presumably the nLab would need to be registered as a (not for profit) organisation in some country first. Could somebody volunteer to do that?
Put up the notice warning about the expected downtime now.
Re #9: I might possibly be willing to undertake this in Norway. However, I am not convinced this is a good idea, due to the fact that the documents will be in Norwegian, which will problematic should I be unavailable. I do think that registering the organisation somewhere and setting up a bank account in its name is the way forward, though.
I’m off to China tomorrow and won’t have time. I don’t think we have to rush things for this particular outage. The world won’t end with nLab unavailable for 20 hours.
By the way, in the warning there is ’Carnegie Mellom’ with an ’m’.
Thanks for all your work on the nLab, Richard. It’s working so much better now.
By the way, in the warning there is ’Carnegie Mellom’ with an ’m’.
Thank you, fixed now!
I’m off to China tomorrow and won’t have time. I don’t think we have to rush things for this particular outage. The world won’t end with nLab unavailable for 20 hours.
Absolutely, sorry for being unclear, #8 and #9 were more about the future; working towards a setup where we have some kind of cloud setup in addition to the server at CMU, so that we can avoid downtime and have better possibilities for backups, etc. If you would be willing to do the registration once you have time, that would be fantastic.
Thanks for all your work on the nLab, Richard. It’s working so much better now.
You’re welcome, thanks for the feedback! Still lots of things to do, but I feel we’re slowly moving in the right direction.
In case my grant is awarded including funding for nLab cloud servers, I could use my university expense card. But we won’t know about that for several more months.
That would be great!
I’m thinking, though, that it might be good in addition to grants to have some kind of setup which allows us to handle the finances of the nLab without any personal involvement. For example, I think many people would be willing to make a small financial contribution if it were needed every now and then, but we cannot really do that without some kind of organisation. In Norway at least, and I think the UK as well, it seems like it is not too much work to set it up for the kind of purposes we have in mind.
The downtime seemed to be more or less as expected. Have removed the warning now.
1 to 16 of 16