.htaccess patch/plugin

O

OC2PS

Guest
#1
Seems pretty clear that LiteSpeed Tech will not support .htaccess on OpenLiteSpeed.

Don't know if anyone has dug under the bonnet yet, but any opinions as to whether .htaccess support is something that could be achieved via a patch or plugin?
 
O

OC2PS

Guest
#3
Yeah, but I hope someone will see the pain of not having actual .htaccess as big enough to knock up a patch/plugin.
 

lsmichael

Active Member
#4
Howdy Octy,

It's still a ways off, but there's been some talk about partial support of .htaccess, like a feature that would search and use .htaccess settings when you told it to (but might require a restart). This means that when you update something, you run the feature and do a restart and all your .htaccess settings are pulled into OpenLiteSpeed.

That's one idea and we're still pretty early on, but it's a thought...

m
 
O

OC2PS

Guest
#5
partial support of .htaccess, like a feature that would search and use .htaccess settings when you told it to (but might require a restart). This means that when you update something, you run the feature and do a restart and all your .htaccess settings are pulled into OpenLiteSpeed.
On one hand, it's good and encouraging that you (LiteSpeed Tech, and not just you personally Mike) are listening and trying to work out solutions.

On the other hand, the quoted suggestion sounds like you are complicating things a fair bunch in an attempt to "not provide full .htaccess support" or "distinguish Enterprise edition from OLS" (take your pick of the phrasing).

You might recall that the primary reason wordpress.com moved from Apache to LiteSpeed all those years ago was because
Barry Abrahamson said:
http://www.wired.com/wiredenterprise/2013/09/nginx
Our problem was that deploying configuration changes which required a graceful restart under production traffic levels didn't work very well.
(Of course, ultimately, wordpress.com moved to nginx in 2008 as LiteSpeed wasn't OpenSource)

I strongly feel that instead of erecting artificial barriers to prevent people from using OLS to its full potential, you should try to build it as a robust, flxible platform.
 

lsmichael

Active Member
#6
Hey Octy,

I strongly feel that instead of erecting artificial barriers to prevent people from using OLS to its full potential, you should try to build it as a robust, flxible platform.
So, how do you feel about Nginx Plus?
 
O

OC2PS

Guest
#7
how do you feel about Nginx Plus?
I think we are definitely getting off-topic (vis-a-vis the thread), but here goes:

In one sentence:
nginX is doing the opposite of what LiteSpeed is doing.

Short answer

nginx Plus is built on top of nginx, which lots of people (somewhere around 20% of the web, in fact) are already using.
You are trying to build OLS as LiteSpeed Minus in the hope of attracting more folks.

Longer answer

nginX created a fantastic Open Source server, which came to be used by millions of sites worldwide. Now, to capitalize on the popularity (when Wired does an article on you, you are mainstream) they are building on top of this tried and tested platform. These additional services are commercial and of the millions of current users, even if a small proportion trade-up, nginX still stands to make decent amounts of money.

LiteSpeed created a scrappy, performance server as a commercial product. Now that OpenSource servers are eating into the niche, LiteSpeed are trying to put artificial barriers in place to block some functionality and release a watered-down version under OpenSource license.

Who would use this?

nginX's Freemium model is tried and tested. Of the millions of folks already using nginX, if there are some folks missing a few features, they'd be delighted to get something additional for a small fee. nginX doesn't even have to market the "Plus" product.

LiteSpeed is used by a few corporates. They are already paying. A fraction of them might convert to OLS (which is actually a bad thing from revenues perspective...but on the other hand if you don't cannibalize your product, nginX, Lighty and yes Apache will...what would you do then?). But that is unlikely to make OLS popular. Do you see the problem here? You might think that nginX and LiteSpeed are approaching the same end-result from different ends, but the fact is that you have "no" captive audience. In order to make "new" people want to use OLS, you need to make a splash. You need to provide a drop-in Apache replacement that is leaner, meaner, faster, easier. Anything short of that is likely to fizzle out.

Hence my suggestions on the other thread.
 

Slavik

Administrator
#8
On the other hand, the quoted suggestion sounds like you are complicating things a fair bunch in an attempt to "not provide full .htaccess support" or "distinguish Enterprise edition from OLS" (take your pick of the phrasing).

I have to agree with this @lsmichael

Requiring a restart, defeats the object of including the htaccess in the first place. Htaccess support should allow for people to alter the htaccess file and then see the changes immediately. If you need to restart after each change, it is no different than actually using the current re-write system.

I can see what your aiming at, that for example, if you upload wordpress witht he default htaccess, it will "work". But at the same time, consider it too limiting to make on-the-fly edits.
 

lsmichael

Active Member
#9
Yo,

I see what you guys are saying, but we're coming at it from a slightly different position. The reason we're so paranoid about cannibalizing our market is because it's not really being cannibalized by anyone else now. We have a niche — Apache users, especially people that use Apache with control panels, that need very easy maintenance and need more out of their current hardware. For the most part, we keep the users we get, but we'd like to have more people using LiteSpeed. So we figured we'd expand into a different niche.

It's because we're so happy (and optimistic for the future) in our current niche that we're so keen on keeping the niches separate. I know you guys would love to have .htaccess file changes taken up automatically, but that convenience is our Enterprise market. Enterprise is there for admins and companies that want to save money on hardware and man-hours on support/maintenance. We're interested in looking at ways to meld and co-mingle the niches, but, like bridging natural habitats, we have to be careful what we introduce. We want to make LiteSpeed stronger, we want more people using LiteSpeed, but we won't endanger Enterprise.

I know you guys are going to say we're not being daring enough, but we're trying to be prudent. I know, I know. I sound like an old man. Yes, if you think of OpenLiteSpeed as a watered-down Enterprise, it's not very appetizing. Maybe that's what it is. But it's also a high-performance web server that's easier to use than Nginx. We're hoping to slowly build a base of users that see that and make it its own vibrant habitat.

m
 
#10
Probably easier to find or develop more or new unique and exclusive additional features for LiteSpeed Enterprise product to further differentiate it from OpenLiteSpeed so to maintain that 'must have' factor that folks would be willing to pay for in Enterprise product (besides control panel support).

As opposed to crippling OpenLiteSpeed adoption by withholding .htaccess file support.

This way you are pushing forth potential customer interest and increased market share for both LiteSpeed Enterprise AND OpenLiteSpeed at the same time ;) Otherwise with current .htaccess file support lacking in OpenLiteSpeed, all you are doing is limiting OLS growth and adoption rates AND not growing the size of the pie for LiteSpeed Enterprise market share and adoption rates. Now that you have 2 products web server products, you have to grow both not at the expense of each other. To clarify that statement, LiteSpeed Enterprise needs to grow it's market share and product awareness through OLS but it won't be able to do that if OLS adoption is held back by that little .htaccess file :)

Just my AUD$0.02 cents :D
 
Last edited:

lsmichael

Active Member
#11
I'll take your Aussie dollars. Never know when they might come in handy.:D

So, what you're saying is that we need to find a new reason for people to buy Enterprise, other than .htaccess and control panel support. I agree that finding new features might be a good way to expand our market and compete against Nginx and Apache. I'm not sure I see this as a good reason to make .htaccess and control panel support OpenLiteSpeed features. If the reasoning behind making .htaccess part of OpenLiteSpeed is that it would attract more users, then why not make control panel support part of OpenLiteSpeed? Our reasoning is the same for keeping both features out of OpenLiteSpeed (or only adding them slowly) — adding them to OpenLiteSpeed might expand OpenLiteSpeed's user base, but possibly at the expense of Enterprise.

Ah, I feel like we're just going back and forth on this and not really getting anywhere...:(

I'm starting a new thread about other companies that have made the same jump from proprietary to open source... See y'all there.

m
 
#12
I know you guys are going to say we're not being daring enough, but we're trying to be prudent. I know, I know. I sound like an old man. Yes, if you think of OpenLiteSpeed as a watered-down Enterprise, it's not very appetizing. Maybe that's what it is. But it's also a high-performance web server that's easier to use than Nginx. We're hoping to slowly build a base of users that see that and make it its own vibrant habitat.
m
Actually, I find nginx much easier to configure than OLS. I can simply install the necessary PHP related PECL's and modules and off I go. I can get memcached/memcache to work under it fine as well as xCache. I STILL haven't been able to get xCache to work with xenForo yet and APC has problems that I really don't want to deal with and the other opcode cache won't even compile (guess I'm missing even MORE dev libraries).
The configuration file(s) are not that hard to figure out and the syntax is fairly simple.
 

lsmichael

Active Member
#13
Howdy Tracy,

To some degree, I feel your pain. We're still working on making OpenLiteSpeed easier to use. We need much more documentation and a bigger community, but, with the GUI and Apache-style rewrite rules, I think it has the potential to be much easier to use than nginx.

m
 
#14
Howdy Tracy,

To some degree, I feel your pain. We're still working on making OpenLiteSpeed easier to use. We need much more documentation and a bigger community, but, with the GUI and Apache-style rewrite rules, I think it has the potential to be much easier to use than nginx.

m
That's why I want it to work so bad. My forums are strictly hobby related so I can't justify going out and spending the big bucks on LS when nginx is doing just fine. OLS would be a good compromise if I can ever get it to work correctly... and nope, not a neophyte to setting stuff up. I'm pretty good with both Apache2 and nginx. ;)
 
L

lisandi

Guest
#15
Hmm - just came over this thread while setting up a Nginx-varnish-apache-Virtualmin hosting surrounding which creates automatic sites and setting. I totally disagree for what Michael is trying to make us all believe here. OK their business model is something like fishing for fish with a bad nonfunctional hook in a see full of small and big fishes.
  1. Going Open Source was a good decision and making it available under GPL 3 was even better as it avoids the problems currently some GPL2 CMS products have to create more income for the development.
  2. Building an Enterprise and a reduced OLS Version is simply crab - sorry to say this. WHY?
Out there are Millions of Apache Fishes of all size. and they swim around slowly in comparison to what litespeed could offer them. But instead of simply throwing out a net which will catch them more or less all you threw out an fihingline with no hook - well perhaps a hook but with a bad worm. On some finshinglines thow you rpowide a very delicious worm. The .htaccess and Hosting Worm. Unfortunately all Major Hosting Companies are based on Apache and don't care about speed so much as still there is a lot they can do with a one time investment into hardware or the cloud to improove speed. Which means into hardware. The litespeed Fishers with their nice delicious worms can see all those fishes but they are much cleverer than you! They actually attract their small fishes surrounding them with a nicer "Hardware" surrounding - or in other words - no litepseed needed!

What would be the alternative! Here is the Vision and I hope it will become true even this month! - which means FAST as LiteSpeed. Here the steps I would do if I were LiteSpeed and would take Lite Speed really serious! Attention: Below is a VISION until now - and hopefully will become true asap!
  1. Litespeed announces to merge there Enterprise and there OLS product to one single an the best fastest most reliable Open Source Hosting Webs Server in the Open Source Universe.
  2. As a reason for this move you will tell the public that you realized that out there are Millions of Apaches moving around very slowly looking for a refreshing speed improvement but most of them can't afford and other won't be willing to make a major financial investment. Your move to make LITE SPEED FOR ALL - the one and Only Server - which is a combination of your current products Enterprise Webserver and OLS will result in a at least 50% move of all Major Hosting Companies to Lightspeed. WHY this will happen?
    1. Google is valuating loading times more and more
    2. The customers of the Hosting Companies will need faster website
    3. A Hosting Company providing a 100% Apache compatible (on place replacement for Apache!) Lite Speed Server will be more than interested to switch their complete infrastructure to use now LiteSpeed Instead of Apache.
    4. As it is a "on place replacemet" another scenario would be that Hosting Companies will provide customers with a SWITCH - Do you want to use APACHE or LITESPEED FOR ALL. As it would be a 100% compatible replacement the customers will for sure start testing out the performances and soon will realise that Apache is to be send into the Museum and to be preserved for History.
    5. For sure NO customer and NO Hosting Company which attracks Masses i.e. would even try to use litespeed enterprise if they would need to pay for it. NONE of the customers will use LiteSpeed Enterprise and not even OLS because it will be a hazzle to be setup and no real community support and beside this the BIG Problem and WORRY that the LiteSpeed People might lock down LiteSpeed OLS or nomore support it when they got all their pockets full of money from the EnterPrise.
  3. The CMS TYPO3 is currently doing something similar but just in the opposite way. They were the best Open Enterprise CMS since 2002 and now their people are running away - WHY? IMHO because they focused to much on a small group of people and companies who might be able to pay their very high prices for a "Developer" Product. So TYPO3 which was very userfriendly was getting more and more a headache and - we need a developer like - so people went i.e. to Drupal - which is getting now lots of festures in 8 what TYPO3 had since 2002 but wont have in TYPO3 NEOS i.e. Multilingual Sites etc.
  4. Litespeed will end up like TYPO3 witha few people being able to afford Lite Speed. Some LiteSpeed Developer might be happy to benefit out of this personally but over all I guess that they simply IGNORE what MAJOR CHANGE could actually bring them into the world of REAL ENTERPRISE!
  5. OK I continue the Vision I have: After LiteSpeed hopefully announced in January 2014 to support from now on ONLY LiteSpeed for All as FREE OPEN SOURCE SOFTWARE licensed under GPLv3, many Hosting Providers started switching to provide their customers a replacement for Apache. Just about 2 weeks Later LiteSpeed for All will increase their userbase to more than 400%. More and more people switching to LiteSpeed. for All. End of February 2014 the LiteSpeed team announced their certification Program. Together with Pearson they had setup a worldwide online certification program which will bring them more revenue.
  6. In March 2014 LitSpeed for All has already conquered abot 50% of the Market shares of all Apache Servers and it is now fast approaching the Rang Number 1 in Webserver Usage. The LiteSpeed team meanwhile was able to double their staff as many hosting companies asked them in helping to convert all their sites to LitSpeed For All.
  7. In April 2014 LiteSpeed For All is holding Rang No 1 in Webserver Usage used because most Hosting Companies had switched to use LiteSpeed For All in their Hosting environment instead of apache. Some even without asking the customers . Only a few former Apache Customers actually complain as the gain in speed is so tremendous.
  8. In May people from NGINx and the Varnish branches switch to use LiteSpeed For All too. Now it is LiteSpeed for All time!
  9. While the company around Litespeed increased their own staff they now pull in Major projects from HUGE customers. LiteSpeed For all is in the Mouth of All and everybody is talking about it. The popularity of LiteSpeed For All can't be stopped anymore. A very nice effect of the move in January 2014 to release LiteSpeed For All under GPLv3 (in our Vision unfortunately only until now) brought the developer community thousands of voluntary workers, module providers, caching speacialists and and and which now contribute their time, energy and code to the fast growing LiteSpeed For All Open Source Community. Thanks to GPLv3 some of those very specialized Modules can be provided for a fee to special customers. A LiteSpeed For All Market starts to grow around the product itself. The LiteSpeed for All APP STORE grows nearly as fast as the community.
  10. Another major move brought LiteSpeed For All lots of people using it. They started cooperating with the people who build Virtualmin/Webmin - a Free OpenSource Hosting Environment.
FAZIT of this Vison.
LiteSpeed For All will be affordable and feasible and manageable for all. It will bring you a huge community and lots of highly qualified Free Open Source Software Developers which will contribute to make LiteSpeed For All even much faster, much more secure and much more user friendly than right now. The Services provided by the comany behind LiteSpeed are asked now by many clients, one reason why they started their certification program. People are proud to be LiteSpeed For All Certified.
----
Instead of trying to reinvent wheels like the former litespeed Enterprise and the OLS Version tried to do - the Free Open Source LiteSpeed for All release conquered them all!
----
Well all the above is a vision which ONLY will become true if you will release really "LiteSpeed For All under GPLv3 in the next couple of weeks. You have the product the world is actually looking for so simply make it available for all BEFORE the people of Apache or NGINx will even have time to start thinking about how they could improve their own product to be like litespeed! The longer you will wait with this move the more people will actually stay with their former products as they will analyze LiteSpeed and improve their own Products with the findings! SURE!

Don't Hide your talents - SHOW IT! as it will be worth doing it! Perhaps you should found a OpenSourceCompany with shares people can buy in parallel. Try to pull in other CMSs other Products to be 100% using and recommending LiteSpeed and no more Apache - This and ONLY this approach will really bring You and actually all of us a benefit! Evertything else is nothing more than a waste of talents.
----
This said here my question to the LiteSpeed Community.

If the people behind LiteSpeed don't make this move why don't WE make this move ourself and FORK the LiteSpeed Project???
Why would be a fork a good idea:
  1. We would NOT start from scratch with a server project
  2. The server would already able to replace Apache
  3. We could easily analyses and adjust the code bases to what would be needed to integrate .htaccess and to be 100% hosting plattform compliant with cPanel, WHM, Virualmin and for sure we will get the full support of those Companies and Developers behind those hosting panels!
  4. Similar thing happened recently to the great ERP System OpenERP - when they started to go more commecrial they lost lots of great people to the Free and OpenSource Alternative. The same happened to Redmine and the Chilli Project.
  5. Think about it and let us know - it would be great to create actually LiteSpeed For All in Reality and in litespeed.
CU there.
 
L

laapsaap

Guest
#16
I registered on this forum especially for this thread. I admin a few hundred Linux servers for about 13-14 years and use many opensource software to achieve that. We all started out using Apache only, then came lighttpd and then Nginx. Now we run a mix of Apache and Nginx.

And how we choose between Apache or Nginx on a new server is quite simple. When we dont need .htaccess support, we go for Nginx, when we do need .htaccess we go for Apache. When we need high performance, we just cannot support .htaccess. This is the missing link and where Litespeed can fill the gap. But Apache and Nginx are both free full features products and (open)Litespeed isnt, this is why you will never get any mainstream interest

You should look at Nginx as an example, how they capture so much market and funding. Dont think Nginx will get so far going Litespeed way.

And look at mariaDB, not much longer everybody will be using mariaDB in stead of MySQL.
 

lsmichael

Active Member
#17
Howdy,

Just figured I'd put in my two cents when I have time. I'm mostly interested in why people think OpenLiteSpeed is not a "full-featured product". It is a new product, of course, and thus we're still working on adding features. But, if your argument is that OpenLiteSpeed is not a "full-featured product" because it does not use .htaccess files, well, where does that put nginx?

We're still working on adding more features and making OpenLiteSpeed an even better (and more usable) web server than nginx. This will almost certainly involve making it easier to get your .htaccess settings into OpenLiteSpeed. Unfortunately, we are only human and have not gotten to developing this yet. Time will tell if we have what it takes.

Cheers,

Michael
 
Top