> On 20201001, at 22:08, Attilla de Groot <[email protected]> wrote:
> 
> I’m not familiar with all the details, but an F5 is probably an overkill and 
> you can acomplish the same thing with quite a simple nginx setup. ;-)

BIG-IP is indeed effectively an old Red-Hat, with Apache, Zebra (not even 
Quagga) and some other bits and pieces and a big API and UI around it.
The good old hardware boxes with SSL accelerator cards where quite cool, but we 
have modern software and hardware-accelerated crypto in CPUs nowadays, which is 
exactly what the 'virtual' version does.


As you say, you can just use F5's nginx (yes, they bought that) if you want 
load-balancing and whatever BGP daemon if you want to go there.

Using a proprietary box like that for a open source project is also odd, 
especially as it runs mostly on open software that is an apt-get and 
ansible/puppet/choose-your-poison away.

And as I understand Dirk is just using a NUC at the moment, if that is not 
sufficient, upgrading that single box is much much easier than using an 
overloaded VM for little gain...



Also BIG-IP does not solve data sync between an active/passive host which is an 
actual concern from what I understand (can be off there ;) .

Greets,
 Jeroen


>> On 1 Oct 2020, at 21:51, Chirana Gheorghita Eugeniu Theodor 
>> <[email protected]> wrote:
>> 
>> This can solve the issue of transaction loss if a switch from active to 
>> standby ocures. But I think there are also some open source variants of the 
>> F5
>> 
>> On Thu, Oct 1, 2020, 22:49 Chirana Gheorghita Eugeniu Theodor 
>> <[email protected]> wrote:
>> The virtual version of the F5 bigip load balacer. The vm can be run in any 
>> virtial enviroment.
>> 
>> On Thu, Oct 1, 2020, 22:06 Dirk Hohndel <[email protected]> wrote:
>> 
>> 
>>> On Oct 1, 2020, at 11:23 AM, Chirana Gheorghita Eugeniu Theodor  wrote:
>>> 
>>> Or we could sponsor a F5 bigip virtual appliance and that should solve the 
>>> load balancing. Correct me if I am wrong.
>> 
>> I don't even know what that is :-)
>> 
>> /D
> 

_______________________________________________
subsurface mailing list
[email protected]
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Reply via email to