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
Familiarity is one aspect. CVE-2019-11043 illuminates in detail the consequences of configuring something you're not intimately familiar with. I'm not sure why anyone would dispatch a file to PHP-FPM without first seeing if the file actually exists, but the configuration was widespread enough to garner attention.
Other issue is with keepalives on, testing locally with a static file, NGINX is ~27k req/sec. Apache is 10k req/sec spread across 10 concurrent requests. ~1.5 ms per request won't make a world of difference when your WordPress blog takes 4000 ms to load up because it's stuffed with unnecessary plugins. Your bottleneck is in PHP. Handoff from NGINX/Apache is thus insignificant.
Now, if you're building a CDN, then NGINX no question. Always pick the right tool for the job. Sometimes knowing how to use a tool is more important than picking a tool.
Guy I work with pronounced it with a long i.
En-GYE-Necks. Sounded like an anatomical term.
I told him he had to stop because it was driving me nuts. He still says LYE-NUCKS. Surprisingly, he's a crackerjack Linux admin.
Because you can't replace your father. No matter how cool and rich your father-in-law
Apache is the father of all webserver
Actually, Apache replaced its father, the father of all web servers: NCSA httpd
And NCSA httpd replaced its father, httpd.
This. In most cases the webserver isn't the bottleneck.
It should be noted that it's now possible to use MPM event: it's not as bad at it has been at some point, even if it's not as efficient as nginx.
I did say 'used to'
I use both, for me Apache is better and easy to config... Apache work out of the box.
Um, your deank plugin that rewrites everything he says to mean the opposite isn't working. Until then, just mentality flip whatever he says for it to make sense.
why not litespeed/openlitespeed?
Apache is not slow or inefficient when used in MPM_EVENT mode and not the legacy Prefork... the world just needs to grow (Debian 10 dropped prefork by default)
Secondly... there are legacy apps that need htaccess support and yes you just can't convince the owners to upgrade them. I personally have legacy old apps that use hundreds of 4+ levels nested .htaccess rules that would be impossible to parse through Nginx efficiently.
Integration, legacy and marketshare. Take for example, Bitcoin, it's worth the most but compared to newer tech, it's considered a dinosaur (slow tx, mempool congestion). The reason why it's number #1 is because it had first mover advantage and is widely supported. Apache is the same imho. However, in saying that, Apache has kept up with development and is not much worse after you tune it, when compared to Litespeed, Nginx.
For most end users, choosing between Apache and Nginx is like choosing between blue and red underwear. They both will do the job well. I just happen to learn Apache first, and I don't see a compelling reason to change based on my experience.
Why are you spamming?
Ask @Raindog307, he started it.
Ignore that poser. Nothing but fake news spewing out of his wannabe face.
If an admin moves away from Apache, that admin would have to adapt to new configurations and new management; just as humanity would need to adapt to new gravity and atmosphere on other planets (keeping the comparison).
Oh, mi Gosh, @Raindog306 is actually flaming me!
Or raining me?
He is? Flag the post and I'll mod his ass with extreme prejudice.
I'll make it rain on you. Meet me out back.
Cannot.
You see, I wanna be a space cowboy. Luna, Mars, Venus, Pluto, I wanna go there.
However, our dear thug, @Raindog305, is a true and thorough Earthian nord. He doesn't wanna leave mother Earth.
We are not meant to be together.
It seems that .htaccess files is a major reason for people to stay with apache. So: why doesn't nginx have a module to work with .htaccess files?
I'm asking because on one side .htaccess seems to be sooooo important but on the other side not important enough to bring them to nginx.
(FWIW I'm unbiased and no particular fan of either one)
Is it because Apache has a more process-driven approach than the event-driven based that NGINX does? sing event-driven architecture to handle multiple requests has a few downsides too I guess. Anyone else thinks otherwise, please let me know
there is somewhat of a difference between .htaccess and mod_rewrite use. i doubt 99% of users touch any real .htaccess directives allowed by apache, but they all deploy "clean URL" mod_rewrite rules.
nginx requires being able to edit the conf file for your virtualhost to rewrite.
iirc, htaccess files slow down processing a decent bit on apache, especially if there are multiple in folders.
And that's why apache is still used a lot even if in some cases it's slower than nginx: if makes it easy to setup shared hosting and to let users have some control over the behavior of the webserver for their domain, without them ever touching the config file. For low traffic sites, in most cases you don't need the best performance (and if that's the case and MPM event is not enough for you, you can add varnish or nginx in front of apache).
Because apache is from friends and nginx is not!
Apache is easier to config (for me), but I am starting to move towards OpenLitespeed
Simple, because:
I do (think otherwise). In fact, both processes/threads and event based have their reason to exist. To summarize it, processes/threads allows to spread the load over multiple cores while event based gets the most out of one core and is much cheaper.
But both can be combined in diverse ways and it's not really a one or the other situation (although that usually makes the configuration for the end user quite complicated).
FWIW, my personal reason to not even look at apache since many years is the fact that their config is XML based, which I consider an abomination except for rare exceptions. It seems to me that the nginx people have thought more and better about their design. But then, to be fair, apache has been designed some millenia earlier (tongue in cheek) and maybe at that time XML and processes seemed to be the smart was to go (or imperator Nero forced them to take that route).
i wouldn't really call it xml
It's not. I've never thought apache and nginx conf were that different that one had to exclude it. On quick glance, it's basically {}'s and ;'s vs <>'s and structure is more or less the same.