Server thoughts for renewal

[FishPants]

So we've been with our hosting provider for quite awhile now and it's not cheap (pretty much 2k Canadian a year) and things have been rock solid (which is why I try to avoid changing too much -- it's nice to not have to do much other than regular patching).

That said, our provider (OVH) is top tier -- they have lower tier companies under their umbrella that offer cheaper servers for a hell of a discount (basically cut our bill down to less than half) but that comes with a bit of a hit:

1. Less outgoing bandwidth, from 1gbps to 250mbs -- this should not be a problem for a forum. I've look at some traffic logs, and we should be fine.
2. Less memory - Go from 64gb to 32gb (no option to customize on lower tier). This one makes me maybe a little more concerned as I currently load the entire forum DB into memory with mysql. If we get into more disk access, there could be some lag on posting etc (probably more on editing previous posts vs inserting a new record). Maybe a non-event, won't know till we try.
3. Marginally less speed on the cpu (from the benchmarks online, it's a 10% reduction in CPU).

Knowing that this is community fund raised - and COVID is likely kicking a lot of our member's financial asses; maybe this is the time I put some work into migrating us? Everything is in virtual min -- so in THEORY, I just need to install virtualmin and copy across the backup and the magic internet things happen to restore the site elsewhere. I've tried this before at home on an old PC just to ensure the backups worked, and it did. That being said I would need to order the server sooner rather than later so I can begin testing backups and tuning the DB to the new memory size etc.

So thoughts on this? Poll questions created.

[Kraken]

I'd try the discounted service for a year. If it's unsatisfactory, you can always go back to top-shelf again the next year, right?

[Little Raven]

I'd try the discounted service for a year. If it's unsatisfactory, you can always go back to top-shelf again the next year, right?

I'm inclined to agree. If our current usage is well below what we're specced out for, (which seems probable) then there's probably no real harm is saving a few bucks.

[Skinypupy]

Most of that is technobabble to me, but if a cheaper version would still work for our little corner of the interwebs, I say go for it.

[Smoove_B]

Yeah, I'd need to defer to the people that know the technology here. Whatever makes the most sense, I'd support.

[Sudy]

The way you put it, the value would likely far outweigh the potential drawbacks. My only concern would be the amount of additional work it initially creates for you.

[Remus West]

Where are we financially? You list needing to fund raise the cheaper version soon but not when we would need to raise the more expensive - I assume soon as well given the disparity in cost.

[gbasden]

I voted to stay where we are. The small decrease in CPU isn't terribly concerning, but halving the RAM when we are using it for DB caching has the potential for some significant performance issues. I'm happy to kick in more to keep support and performance at a higher tier.

[Smoove_B]

Where are we financially?

I know the loss of Amazon link revenue was unfortunate, but is there any chance of figuring out a way to have that re-instated? Like, if there was a front page for the site and links to buying OO T-shirts via Cafe Press (or equivalent), would that meet Amazon's business requirements?

Sorry for off-topic commentary, but if there's anything to figure out so that Amazon referral cash can return I'd offer my assistance there as well (to whatever degree I could help).

[Daehawk]

I voted reduce it. I feel bad about this post because Im not able to help. If I win the lotto I cant afford to play Ill buy you a server though.. I should play once a week. I can send ya some canned goods though you can resell or bribe companies with..if they like beans. Got lots of beans. Lots.

[LordMortis]

I don't mind donating every year (which it's almost time for) but I fear if you don't get enough of us or if we aren't generous enough it comes out of your pocket or we shut down. I'm good with kraken's voice. Try it for a year, if it's noticeably worse then we know why we give. but I'm easy. I like what we have and I'm comfortable with what I give. I just don't know how it looks from your end.

[Kraken]

Yeah, if we have to wait an extra three seconds for a post to register, I can cope. OTOH, most of us are going to get another government stimulus check, so we could all send those to FishPants and buy him the company.

[Exodor]

Voted for the cheaper option - the risk seems reasonable and cutting costs in half helps ensure this place stays around.

[Jeff V]

The reality is this isn't a growing community; quite the opposite, we seem to be shedding a few active members every year. Do we have any load metrics that you can match up with the specs of the cheaper option and maybe take a guess as to the likely effect on our day-to-day experience. I would guess database searches might take longer, but for those of us not Isg, how big of an impact is that really? Now, if we're talking severe lags in day to day posting, then I'd agree with gbasden and would be willing to contribute a substantial amount if he is able to get me a job at Microsoft but until I'm gainfully employed, I can just wave from the sidelines.

[Baroquen]

I'm torn, in that the reduction in RAM sounds like it could be problematic, even with our shrinking community. But on the other hand, it could be fine and the reduction in costs would certainly be worth it. I guess I'm going to vote for the cheaper option and hope for the best on the tech back end?

[Moat_Man]

Why don't we do the fundraising drive early this year and see where we end up? If the doners pony up enough cash then we don't need to do anything and we keep our speedy, non-crashing server. If we come up short then FP has enough time to implement the contingency plan on the reduced service model.

[Blackhawk]

Need data!

With our current incoming funds, how close do we come to the full server costs each year? How does that compare to the cost of the new server? Also, I'm not an IT guy. I don't have any idea what a reduction in CPU means for a forum. And 'some lag' - are we talking milliseconds, or twenty seconds?

[Zarathud]

If we could keep the memory, I’m not worried about the rest. I’m overdue to chip in anyway.

[LordMortis]

Why don't we do the fundraising drive early this year and see where we end up? If the doners pony up enough cash then we don't need to do anything and we keep our speedy, non-crashing server. If we come up short then FP has enough time to implement the contingency plan on the reduced service model.

I can live with this. Early donation sent. One way or another it will come out in the wash. (or is it the Wash? Are we referring to sea in the UK or the fact that I sometimes don't empty my pockets.)

[Sudy]

Can we add those little pop-ups that ask you to enable notifications, accept cookies, and disable ad blockers? What about videos that minimize to the bottom corner of the screen when you close them? This site is feeling very behind the times, frankly.

[Archinerd]

Sounds like we could probably get by with the cheaper option, but I'll defer to the experts.

[TheMix]

I'll echo Kraken. He said it first. And said what I was thinking. With the caveat that I'm not a "power" user, so it's unlikely I'll see much difference. Poor Isg....

[Lassr]

I'm fine either way but do agree that this is not a growing community so moving cheaper may be smart.
Do you know how many people chipped in last year? I know we raised what was needed but was wondering if we had many smaller donations or a few larger donations. I think I did $50 last year. Just curious if we had like 30 give at an average of $70 or much more at a lower average. I remember wondering how much to give last year so I started low and would have done more if we wound up short.

[Blackhawk]

I put back five per month for it, then send it in a lump sum when the time comes, so that's $60 for me.

[Isgrimnur]

I'll echo Kraken. He said it first. And said what I was thinking. With the caveat that I'm not a "power" user, so it's unlikely I'll see much difference. Poor Isg....

Somehow, I will manage, no matter what the decision. But I do appreciate the concern.

[Skinypupy]

Kicked in my donation early, if it helps in the decision making process.

[Baroquen]

Kicked in my donation early, if it helps in the decision making process.

Same.

[FishPants]

The way you put it, the value would likely far outweigh the potential drawbacks. My only concern would be the amount of additional work it initially creates for you.

Well that's a consideration, it's not a small task to move us - but that's why I would want to start sooner than later so I'm not up against a crazy deadline.. I can move backups over and test -- figure out what I need to fine tune for the database with the reduction in ram etc. I don't mind doing it, I would mind doing it all in one weekend.

[FishPants]

Where are we financially?

I know the loss of Amazon link revenue was unfortunate, but is there any chance of figuring out a way to have that re-instated? Like, if there was a front page for the site and links to buying OO T-shirts via Cafe Press (or equivalent), would that meet Amazon's business requirements?

Sorry for off-topic commentary, but if there's anything to figure out so that Amazon referral cash can return I'd offer my assistance there as well (to whatever degree I could help).

Their message was "This decision is final and you are forever prohibited from having an affiliate account again". They cracked the whip hard on affiliates because they don't "need" to pay that money out anymore, a lot of the commercialized services (at least here in Canada) like Great Canadian Rebates (like ebates/rakuten) no longer offer Amazon links either because they got booted as well.

It was a great way to do it, but unfortunately that ship has sailed. Bezos is a dick.

[FishPants]

I'm fine either way but do agree that this is not a growing community so moving cheaper may be smart.
Do you know how many people chipped in last year? I know we raised what was needed but was wondering if we had many smaller donations or a few larger donations. I think I did $50 last year. Just curious if we had like 30 give at an average of $70 or much more at a lower average. I remember wondering how much to give last year so I started low and would have done more if we wound up short.

Unfortunately I don't know how many did, I had a spreadsheet somewhere that I can't seem to find in my mess. I'll dig a bit, but I want to say around 50-60 people if memory serves.

[Moat_Man]

Smack me if this is a dumb question (and frankly I may have asked this last year too so you can hit me twice) but for us Canadians do you have a C$ donation link so we can avoid the conversion hit? If the hosting costs are in C$ then it *appears* I will take a coversion hit donating in US$ then you will probably take a conversion hit again coverting the US$ back into Canadian. Or am I needlessly complicating things and should just shut up and donate?

[Default]

Kinda tapped myself out paying the mortgage off (yay!) but will kick some Ameros into the hat in August.

[FishPants]

Smack me if this is a dumb question (and frankly I may have asked this last year too so you can hit me twice) but for us Canadians do you have a C$ donation link so we can avoid the conversion hit? If the hosting costs are in C$ then it *appears* I will take a coversion hit donating in US$ then you will probably take a conversion hit again coverting the US$ back into Canadian. Or am I needlessly complicating things and should just shut up and donate?

I think you can just donate CDN $ no?

[FishPants]

Smack me if this is a dumb question (and frankly I may have asked this last year too so you can hit me twice) but for us Canadians do you have a C$ donation link so we can avoid the conversion hit? If the hosting costs are in C$ then it *appears* I will take a coversion hit donating in US$ then you will probably take a conversion hit again coverting the US$ back into Canadian. Or am I needlessly complicating things and should just shut up and donate?

Added a CDN dollar link.

[infinitelurker]

I'm a software/devops guy (but for large enterprise apps, never worked with phpBB), and am curious if you could share some of the stats around what is needed to host OO. Database size, compute needed, bandwidth, etc. Are other hosts cheaper (?), but too much of a pain to move to, etc. If it's a pain to gather the info - don't worry, really just curious.

I'm going to kick in some money too (done!), and vote to stay as is, just because I know what it's like in your shoes. Time is money, and tweaking something that ain't broke can bring unforeseen pain. That being said, a little disk access here and there never hurt anybody, but then again, I'm guessing these servers aren't specced out with ssd's?

Thanks for what you do!

[Lassr]

Just did my donation.

[Kraken]

I can't donate yet because I'm spending a ridiculous amount of money on drugs...for my cat. Things should stabilize in a week or so, but until then I won't know how much extra cash I can scare up.

[Moat_Man]

Smack me if this is a dumb question (and frankly I may have asked this last year too so you can hit me twice) but for us Canadians do you have a C$ donation link so we can avoid the conversion hit? If the hosting costs are in C$ then it *appears* I will take a coversion hit donating in US$ then you will probably take a conversion hit again coverting the US$ back into Canadian. Or am I needlessly complicating things and should just shut up and donate?

Added a CDN dollar link.

Thanks. Done.

[Gateguard]

Should I be worried about voter fraud?
Oh well, I voted to stay where we are and made a donation. Just promise to keep the lights on here regardless of the decision.

[noxiousdog]

Donated.