New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
This was already addressed a while back... Once the cPanel price panic hit, we made a knee-jerk decision to offer "Owned" licenses. Owned was the same price ($299) but only a year of updates. It wasn't a true lifetime license. After that, $99/year to get updates.
But then guys like Rack911 tore us a new one, saying it would encourage people to run older (and possibly vulnerable) software in order to save money. We only sold a handful of these and quickly gave up on the idea. It wasn't very well thought-out.
To prove our intentions were good, anyone with this license type could change it to our traditional lifetime and get the free updates forever, therefore always having secure/up-to-date software.
My comment was about extended support cost being $99/y or $199/y? I thought it was $99/y.
Didn’t know you offered to change all Lifetime Licences over to the traditional Lifetime Licences.
I respectfully disagree. There is only so much we can do with one payment. However, we may never agree, so I will openly offer you a full refund on your lifetime license even if you purchased it 10 years ago.
Totally different things. That $99 you are seeing was that botched "Owned" product. It got taken off the shelves pretty quick. If you happened to be someone who grabbed one, we will ignore the $99 thing and just make it a traditional lifetime.
There is a $199 option for retail lifetime licenses (to get a year of access to our tech staff) but to be honest, barely anyone uses it. It wasn't even our idea but there were a few lifetime owners pressing hard for some kind of support option. That's why you've never seen it mentioned anywhere. It's just there as a last resort/option for those who want it.
Most would say, that's pretty expensive.. 2/3 of the lifetime price. But that's kind of the idea. Staff is not cheap and lifetime licenses are already a hell of a deal -- so we just didn't have any motivation to offer a hell of a deal on a support package for them also.
Sorry if that sounds blunt but that's the reasoning behind it. Imagine selling something of your own at a loss. It happens. But then imagine selling something else at a fair price, and the same guy comes back to you and says because he got a deal the first time, anything else he buys from you, you must sell at a loss too. He deserves it, because of the deal he got the first time around.... That's what we were trying to avoid
Thanks for the detailed reply @DA_Mark. Absolutely agree with you and everything you've written. No, your reply wasn't blunt in anyway. You need to be pay the bills at the end of the day and keep developing the product.
Personally, all my servers are DA and I wouldn't go back to cPanel. For me, what sets you guys apart is how approachable you are, receptive to user suggestions, feedback and enhancements and above all how great DA works for both admins and users. I am looking forward to the Pro Pack being officially released.
@DA_Mark any ETA on pro pack ?
Oakley has deep pockets. They have a history of way over paying for shitty properties just because they 'want' it. Everyone has their price. While I have no doubt that DA is friendly to both Cloudlinux/JetBackup, they need to do what's right to protect their product.
If suddenly Cloudlinux stopped supporting DA, well, no one is going to use DA for grand scale shared hosting, only for personal boxes.
People should be grateful that they're still honoring the lifetimes and not renamed the product, just so they can go "TECHNICALLY it's a different product now, your lifetime was for this now EOL DA."
It's a bunch of people that bought a license 10 - 15 years ago and thought they now hit the lottery with cPanel being dicks. You completely forgot you had the license up until this all happened and didn't care about the product or what happened. Now you're upset.
Francisco
Even thou they are on good terms DA got to develop that it will protect everyone
CL now as product without alternative is potential target for every VC that knows that as if they manage acquired it they can heavy milk the customers after that
So even the price is close I will spend some for unpolished product early adoption to secure my business and potentially all of the industry from milking
The official answer is no we don't. We've been friends with CL for a long time, and JetBackup took a huge risk developing for something like DA that has only a small % of the market share. So we are happy to put endorse them through a simple web link.
Behind the scenes, we don't sell their software or receive any affiliate revenue. We get ZERO benefit if you click through our site first, and then buy something from them.
I'll get @smtalk to provide more information when he is around. To be honest, adding the Pro Pack introduction to our docs is not for the reasons of building hype, or trying to pre-sell it. It's just to provide an explanation why retail Lite/Standard licenses start showing things other licenses don't. That's really it.
My heart is ready. My wallet is slowly opening to my heart.
So pro pack will have some extra perks for DA but not no per user resource control similar to CloudLinux and we may not see Jebbackup feature either.
I had hopes
https://docs.directadmin.com/other-hosting-services/pro-resource-throttling/general/
It's done, and yes it's per-user resource control. Keep in mind this is all pre-release stuff, so the crowning touch will be a Pro Pack interface in the Evo skin, which we guarantee in the next release.
Of course, it will include a nice resource management interface.
For anyone who wants to try the feature now now now, the enhanced skin will show the per-user resource controls.
Thank you!
Looking forward to removing cloudlinux!
Do we need pre release binaries?
Yes, it's there right now in the pre-release binaries (but only in the enhanced skin at the moment). You'll need to wait for the actual release to see a slick resource-management dashboard in the Evo skin.
Not every OS is supported, so check the doc again:
https://docs.directadmin.com/other-hosting-services/pro-resource-throttling/general/
While in that doc, take note of the CloudLinux section. CL is much more than just throttling and we don't want to market this as a replacement. Let's not go overboard here
Sir how very much is this much?
It seems I am very late to the party! I am excited about the development and if anything, DA deserves a huge thank you from the community.
Many wanted resource management built in to the control panel from before the price hike fiasco. Nobody listened or did anything to make it a reality. So thank you for listening to the community. It was a craving.
Though it is a work in progress, will there be any log for the resource usage for the past 30 days or so on a per user basis?
Can any admin make it a sticky?
It is really going to be beneficial to many. People need to know this development. A sticky could give it some more exposure.
Sent you an e-mail! The public answer is, this is an early-stage product.. a few features are completed but many are left. It's not ready to sell yet. The top-tier retail licenses can test it out, but that's it so far.
Thanks so much for the enthusiasm -- I honestly think this will be our fuel moving forward.
We want to do more, but we've always been a budget panel with limited resources. And we're certainly listening. I think this Pro Pack is going to be a great thing, because we can finally do more than just listen.
Regarding your logging question, would you mind if I let @smtalk answer that? I don't know the answer and I promise he'll pop in here as soon as he can. It's been a long week and I'm sure he needs some time away but I have a feeling he'll be here pretty soon.
You are very welcome!
For me, it is a pleasant experience seeing a piece of software evolve and experience the change as an end user/direct beneficiary for that matter.
I do think that this development might become monumental in DA's own journey, at the same time helping small and medium businesses to survive.
So yes, I am enthusiastic.
It seems waiting has its rewards.
That's a pleasant news.
Its not like that I am asking for more but as you mentioned on page about resource control, for mysql resource abuser, we have to look into CL as you dont' have any plan to implement it or do you ?
How does this affect machines already with cloudlinux? Do we need to remove cloudlinux first?
Yes, but we plan to only show current usage + any failures (limit hits) for 30 days. With a limit hit you'd see the value it has reached and the limit which was set at that time. I guess it should be enough?
MySQL isn't a user-level process, and it a bit different. Even with CloudLinux you only get CPU and read/write limitations, a separate configuration file (https://docs.cloudlinux.com/cloudlinux_os_components/#configuration-and-operation) and a different behaviour. We're not saying limitation for MySQL won't be there, but it's not there at this time.
Yes, it'd conflict with CloudLinux. Either CloudLinux or DA limitations would need to be in use.
What would you think about ability to limit these in MySQL:
Would this be sufficient for MySQL limits?
I especialy asked for the ETA and do not expect anything that soon
I added new servers on January 10-14 and up to a couple days ago one has zero customers but now it have couple so I got to wait till next one is in line for production
Also that interface changes should work on traditional theme as a lot of hosts use it
But maybe I'll get a license just to play with it for some time
That's plenty!
I suppose you will also log the full file paths that were causing those hits and a way to search through those logs on a per user/date range/resource when you initially release the pro pack?
Additional thought for the pro pack, DA doesn't have an easy way to enable disable PHP extensions from the control panel. A list of extensions with check boxes for each different PHP versions available on the server would be nice. Or, perhaps a similar interface with this level of convenience?
I agree with regards to PHP extensions.....
Guys selectable PHP Modules / Extensions will require pre-compiled php binaries
that will cut at minimum 10% performance from the server and 97% + of the customers do not select anything custom ...
here is benchmark of precompiled Cloud Linux one vs DA compiled 7.4 (opcache is on ) Latest available on the day of testing about 2 weeks ago ...
Even it is latest RH / CL 8 kernel that is old and not fully recognize the Ryzen 3900 CPU on a newer kernel like Ubuntu 20.04 compiled for the AMD vs CL8 it can be more drastic difference
Our main business is to sell shared access to CPU cores, RAM and Space if we can acheive 20% for the same server and bases that Ryzen 3900 is faster than Xeon silvers and some Xeon Golds and based on that 5 series are 12.5 faster on gcc compiling vs current one and up to 20% faster on benchmarks next year we will be killers on the market (scr*w the slectable modules) as difference will be visible with naked eays especially vs cPanel CL7 ones and older intel cores ...
CL one Default module + Opcache
`php bench.php
| PHP BENCHMARK SCRIPT |
Start : 2021-01-08 02:20:19
Server : Linux/4.18.0-147.8.1.el8.lve.1.x86_64 x86_64
Platform : Linux
System : Linux
CPU :
model : AMD Ryzen 9 3900 12-Core Processor
cores : 12
available : 12
MHz : 3094.186MHz
Memory : 256 Mb available
Benchmark version : 1.0.37
PHP version : 7.4.12
available modules :
mbstring : yes
json : yes
pcre : yes
Max execution time : 600 sec
Crypt hash algo : MD5
TEST NAME : SECONDS | OP/SEC | OP/SEC/MHz | MEMORY
01_math : 1.492 sec | 670.26 kOp/s | 216.62 Ops/MHz | 2 Mb
02_string_concat : 0.195 sec | 39.58 MOp/s | 12.79 kOps/MHz | 128.84 Mb
03_1_string_number_concat : 1.479 sec | 3.38 MOp/s | 1.09 kOps/MHz | 4 Mb
03_2_string_number_format : 1.334 sec | 3.75 MOp/s | 1.21 kOps/MHz | 4 Mb
04_string_simple_functions : 1.301 sec | 998.96 kOp/s | 322.85 Ops/MHz | 4 Mb
05_string_multibyte : 0.974 sec | 133.50 kOp/s | 43.14 Ops/MHz | 4 Mb
06_string_manipulation : 2.462 sec | 528.13 kOp/s | 170.68 Ops/MHz | 4 Mb
07_regex : 1.688 sec | 770.23 kOp/s | 248.93 Ops/MHz | 4 Mb
08_1_hashing : 2.172 sec | 598.50 kOp/s | 193.43 Ops/MHz | 4 Mb
08_2_crypt : 7.034 sec | 1.42 kOp/s | 0.46 Ops/MHz | 4 Mb
09_json_encode : 2.110 sec | 616.17 kOp/s | 199.14 Ops/MHz | 4 Mb
10_json_decode : 3.454 sec | 376.36 kOp/s | 121.64 Ops/MHz | 4 Mb
11_serialize : 1.716 sec | 757.75 kOp/s | 244.90 Ops/MHz | 4 Mb
12_unserialize : 2.141 sec | 607.19 kOp/s | 196.24 Ops/MHz | 4 Mb
13_array_fill : 1.510 sec | 33.12 MOp/s | 10.70 kOps/MHz | 12 Mb
14_array_range : 0.318 sec | 314.26 kOp/s | 101.57 Ops/MHz | 12 Mb
14_array_unset : 1.509 sec | 33.13 MOp/s | 10.71 kOps/MHz | 12 Mb
15_loops : 0.872 sec | 229.24 MOp/s | 74.09 kOps/MHz | 4 Mb
16_loop_ifelse : 0.653 sec | 76.53 MOp/s | 24.73 kOps/MHz | 4 Mb
17_loop_ternary : 1.087 sec | 46.01 MOp/s | 14.87 kOps/MHz | 4 Mb
18_1_loop_defined_access : 0.342 sec | 58.55 MOp/s | 18.92 kOps/MHz | 4 Mb
18_2_loop_undefined_access : 3.061 sec | 6.53 MOp/s | 2.11 kOps/MHz | 4 Mb
19_type_functions : 0.724 sec | 4.15 MOp/s | 1.34 kOps/MHz | 4 Mb
20_type_conversion : 0.723 sec | 4.15 MOp/s | 1.34 kOps/MHz | 4 Mb
21_0_loop_exception_none : 0.040 sec | 100.87 MOp/s | 32.60 kOps/MHz | 4 Mb
21_1_loop_exception_try : 0.047 sec | 84.72 MOp/s | 27.38 kOps/MHz | 4 Mb
21_2_loop_exception_catch : 1.332 sec | 3.00 MOp/s | 970.26 Ops/MHz | 4 Mb
22_loop_null_op : 1.014 sec | 49.32 MOp/s | 15.94 kOps/MHz | 4 Mb
23_loop_spaceship_op : 0.907 sec | 55.14 MOp/s | 17.82 kOps/MHz | 4 Mb
26_1_class_public_properties : 0.098 sec | 51.05 MOp/s | 16.50 kOps/MHz | 4 Mb
26_2_class_getter_setter : 0.338 sec | 14.81 MOp/s | 4.79 kOps/MHz | 4 Mb
26_3_class_magic_methods : 0.702 sec | 7.12 MOp/s | 2.30 kOps/MHz | 4 Mb
Total time: : 44.827 sec | 13.44 MOp/s | 4.34 kOps/MHz |
Current PHP memory usage: : 4 Mb
Peak PHP memory usage: : 125.47 Mb
`
Compiled from DA one IMAP, Ioncube + Imagic + GMP + xmlrpc
`php bench.php
| PHP BENCHMARK SCRIPT |
Start : 2021-01-08 02:14:04
Server : Linux/4.18.0-147.8.1.el8.lve.1.x86_64 x86_64
Platform : Linux
System : Linux
CPU :
model : AMD Ryzen 9 3900 12-Core Processor
cores : 12
available : 12
MHz : 3094.186MHz
Memory : 256 Mb available
Benchmark version : 1.0.37
PHP version : 7.4.13
available modules :
mbstring : yes
json : yes
pcre : yes
Max execution time : 600 sec
Crypt hash algo : MD5
TEST NAME : SECONDS | OP/SEC | OP/SEC/MHz | MEMORY
01_math : 1.366 sec | 732.00 kOp/s | 236.57 Ops/MHz | 2 Mb
02_string_concat : 0.207 sec | 37.25 MOp/s | 12.04 kOps/MHz | 128.84 Mb
03_1_string_number_concat : 0.022 sec | 229.26 MOp/s | 74.10 kOps/MHz | 4 Mb
03_2_string_number_format : 1.297 sec | 3.85 MOp/s | 1.25 kOps/MHz | 4 Mb
04_string_simple_functions : 1.254 sec | 1.04 MOp/s | 334.96 Ops/MHz | 4 Mb
05_string_multibyte : 0.917 sec | 141.81 kOp/s | 45.83 Ops/MHz | 4 Mb
06_string_manipulation : 2.221 sec | 585.25 kOp/s | 189.14 Ops/MHz | 4 Mb
07_regex : 1.620 sec | 802.31 kOp/s | 259.30 Ops/MHz | 4 Mb
08_1_hashing : 2.173 sec | 598.27 kOp/s | 193.35 Ops/MHz | 4 Mb
08_2_crypt : 7.077 sec | 1.41 kOp/s | 0.46 Ops/MHz | 4 Mb
09_json_encode : 1.952 sec | 666.13 kOp/s | 215.28 Ops/MHz | 4 Mb
10_json_decode : 3.338 sec | 389.46 kOp/s | 125.87 Ops/MHz | 4 Mb
11_serialize : 1.485 sec | 875.43 kOp/s | 282.93 Ops/MHz | 4 Mb
12_unserialize : 2.108 sec | 616.58 kOp/s | 199.27 Ops/MHz | 4 Mb
13_array_fill : 1.468 sec | 34.07 MOp/s | 11.01 kOps/MHz | 12 Mb
14_array_range : 0.321 sec | 311.42 kOp/s | 100.65 Ops/MHz | 12 Mb
14_array_unset : 1.460 sec | 34.24 MOp/s | 11.07 kOps/MHz | 12 Mb
15_loops : 0.883 sec | 226.53 MOp/s | 73.21 kOps/MHz | 4 Mb
16_loop_ifelse : 0.265 sec | 188.61 MOp/s | 60.96 kOps/MHz | 4 Mb
17_loop_ternary : 0.564 sec | 88.67 MOp/s | 28.66 kOps/MHz | 4 Mb
18_1_loop_defined_access : 0.243 sec | 82.41 MOp/s | 26.63 kOps/MHz | 4 Mb
18_2_loop_undefined_access : 2.886 sec | 6.93 MOp/s | 2.24 kOps/MHz | 4 Mb
19_type_functions : 0.620 sec | 4.84 MOp/s | 1.56 kOps/MHz | 4 Mb
20_type_conversion : 0.620 sec | 4.84 MOp/s | 1.56 kOps/MHz | 4 Mb
21_0_loop_exception_none : 0.017 sec | 231.04 MOp/s | 74.67 kOps/MHz | 4 Mb
21_1_loop_exception_try : 0.037 sec | 107.75 MOp/s | 34.82 kOps/MHz | 4 Mb
21_2_loop_exception_catch : 1.231 sec | 3.25 MOp/s | 1.05 kOps/MHz | 4 Mb
22_loop_null_op : 0.674 sec | 74.23 MOp/s | 23.99 kOps/MHz | 4 Mb
23_loop_spaceship_op : 0.738 sec | 67.73 MOp/s | 21.89 kOps/MHz | 4 Mb
26_1_class_public_properties : 0.082 sec | 61.03 MOp/s | 19.73 kOps/MHz | 4 Mb
26_2_class_getter_setter : 0.322 sec | 15.54 MOp/s | 5.02 kOps/MHz | 4 Mb
26_3_class_magic_methods : 0.680 sec | 7.36 MOp/s | 2.38 kOps/MHz | 4 Mb
Total time: : 40.147 sec | 15.00 MOp/s | 4.85 kOps/MHz |
Current PHP memory usage: : 4 Mb
Peak PHP memory usage: : 125.25 Mb
`
My point do you can keep a server with CL8 of the same CPU model where your customers will receive 20% slower service but will got selectable extensions ... Do not push DA in direction where it will loose advantages and most customers do not need
P.S CL version is older as of the moment of the test no newer binaries are available next day after the benchmark 7.4.14 become available for compiling where CL was still behind
I said 'meh' to CloudLinux's recent pro-style features but I'm definitely saying 'yeah' to DirectAdmin's. Looking forward to this. 😊