What's new

pixelserv pixelserv - A Better One-pixel Webserver for Adblock

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

I guess the ultimate question is how much real world noticeable diff would a home user see, given the speed of pixelserv already. I realize that is likely yet to be determined by testing.

It's possible to quantify it by tests but I wonder anyone will bother to do it. 'cos the process will be tedious and yet the measurement might fall with statistical errors.

With that said, a massively multithreaded server like pixelserv-tls will definitely benefit from more cores. Clustering two or more machines is one way to give pixelserv-tls more cores.

Modern browsers are heavily multi-threaded too. Paraell execution in nature. Hence, pixelserv-tls fits in well and fast. With more instances, responses are going to be even faster. That's for sure.

Routers/APs are abundant resources for users at home. Low hanging fruits shall get harvested. Other than initial necessary setup of pixelserv-tls just like running on a single machine, the key prerequisite is that: your adblock script shall be adapted to generated host files on two IP addresses fro a cluster of two pixelserv-tls.
 

Updated: new beta version 2.1.0-rc.3

Thanks for all the QUICK feedback on memory use!

I uploaded a rebuilt of rc.3 with further optimisation that help on websites with lots of POST requests.

I recommend people update if you're bored with the previous build. From my stress tests on RT-AC56U, the peak memory use is about 10-13MB or 4%-5.2%. Apparently everyone's mileage will vary.

Please continue to provide feedback on memory use.

We're perhaps hitting the limits as quite a few variables are out of an application's control.

Entware (ARMv7, ARMv8 and mipsel) users could run the one liner below or otherwise to install the test version.

Code:
sh -c "$(wget -qO - https://kazoo.ga/pixelserv-tls/install-beta.sh)"
 
Updated to latest Beta:
pixelserv-tls 2.1.0-rc.3 (compiled: Mar 30 2018 17:10:59) options: 10.0.1.3 -l 2
629 uts, 2 log, 1 kcc, 1 kmx, 3.30 kvg, 16 krq, 33 req, 564 avg, 602 rmx, 2 tav, 50 tmx, 27 slh, 0 slm, 0 sle, 5 slc, 0 slu, 75 sct, 9 sch, 0 scm, 0 scp, 6 sst, 3 ssh, 1 ssm, 0 ssp, 27 nfe, 0 gif, 0 ico, 0 txt, 0 jpg, 0 png, 0 swf, 0 sta, 2 stt, 0 ufe, 0 opt, 0 pst, 0 hed, 0 rdr, 0 nou, 0 pth, 0 204, 0 bad, 0 tmo, 5 cls, 0 cly, 0 clt, 0 err

htop- 0.6
 
a massively multithreaded server like pixelserv-tls will definitely benefit from more cores. Clustering two or more machines is one way to give pixelserv-tls more cores.

Modern browsers are heavily multi-threaded too. Paraell execution in nature. Hence, pixelserv-tls fits in well and fast. With more instances, responses are going to be even faster. That's for sure.

Routers/APs are abundant resources for users at home. Low hanging fruits shall get harvested. Other than initial necessary setup of pixelserv-tls just like running on a single machine, the key prerequisite is that: your adblock script shall be adapted to generated host files on two IP addresses fro a cluster of two pixelserv-tls.

Uh, forgive me, but you need to be somewhat more vocal about this. no, a LOT more vocal.

And I'll install the latest beta shortly. If it frees up RAM as well as it appears across all platforms, my LAN might (should?) speed up some.
 
Uh, forgive me, but you need to be somewhat more vocal about this. no, a LOT more vocal.

Well, it takes time. When word of mouth gets to work, more people will follow. I recommend people with their own adblock scripts to try first.

When I adapted my adblock script for two instances of pixelserv-tls, I only need to add one line (2nd ip address) and change one line (make use of the 2nd ip address).

And I'll install the latest beta shortly. If it frees up RAM as well as it appears across all platforms, my LAN might (should?) speed up some.

More resources available for other tasks in theory shall give better overall system responses. In practice, people's perception is very poor. But you're guaranteed to have a robust LAN with more resources at freely disposable.
 
Code:
pixelserv -v

appears undocumented (but totally makes sense)

I was about to reply that

Code:
pixelserv --help

outputs the version number and compile date as well :D
 
houston, we have a problem: 192.168.1.2/servstats is reporting v35.HZ12.Kk compiled: Sep 26 2017

the install beta option in amtm seems to have failed...or did I miss/forget something?
 
16 hours in

Code:
pixelserv-tls 2.1.0-rc.3 (compiled: Mar 29 2018 19:28:58) options: 192.168.1.3 -l 2

uts    0d 16:37    process uptime
log    2    critical (0) error (1) warning (2) notice (3) info (4) debug (5)
kcc    5    number of active service threads
kmx    10    maximum number of service threads
kvg    1.17    average number of requests per service thread
krq    43    max number of requests by one service thread
req    2239    total # of requests (HTTP, HTTPS, success, failure etc)
avg    488 bytes    average size of requests
rmx    1967 bytes    largest size of request(s)
tav    7 ms    average processing time (per request)
tmx    282 ms    longest processing time (per request)
slh    448    # of accepted HTTPS requests
slm    2    # of rejected HTTPS requests (missing certificate)
sle    0    # of rejected HTTPS requests (certificate available but bad)
slc    1043    # of dropped HTTPS requests (client disconnect without sending any request)
slu    703    # of dropped HTTPS requests (other TLS handshake errors)
sct    75    cert cache: # of certs in cache
sch    1512    cert cache: # of reuses of cached certs
scm    0    cert cache: # of misses to find a cert in cache
scp    0    cert cache: # of purges to give room for a new cert
sst    2    sess cache: # of cached TLS sessions (for older non-RFC5077 clients)
ssh    181    sess cache: # of reuses of cached TLS sessions
ssm    177    sess cache: # of misses to find a TLS session in cache
ssp    0    sess cache: # of purges to give room for a new TLS session

Memory Usage at 2.9%
 
houston, we have a problem: 192.168.1.2/servstats is reporting v35.HZ12.Kk compiled: Sep 26 2017

the install beta option in amtm seems to have failed...or did I miss/forget something?

Might not be amtm's fault. Can you try the one-liner script directly.

In SSH, copy & paste:
Code:
sh -c "$(wget -qO - https://kazoo.ga/pixelserv-tls/install-beta.sh)"
 
Might not be amtm's fault. Can you try the one-liner script directly.

In SSH, copy & paste:
Code:
sh -c "$(wget -qO - https://kazoo.ga/pixelserv-tls/install-beta.sh)"
done...same result. should I have rebooted the router?
 
There is something newer than 2.1.0-rc.3?

A rebuild of rc.3 (a different timestamp) that included changes related to POST http method. Helps using less RAM on websites with collecting lots of your privacy info such as Daily Mail.

The detailed change is described in #1525 while the original rc.3 is a few postings back.

If you re-run the one-liner, you'll get the latest timestamp. You can tell as the compile timestamp is different.
 

Similar threads

Support SNBForums w/ Amazon

If you'd like to support SNBForums, just use this link and buy anything on Amazon. Thanks!

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!

Staff online

Top