nuclei multi threading

[molitona]

Hi, why isn't there -t option for multi threading like httpx have ?

nuclei is slow compared to httpx

[olearycrew]

@ehsandeep what are your thought here?

[tarunKoyalwar]

@molitona https://nuclei.projectdiscovery.io/nuclei/get-started/#mass-scanning-using-nuclei

[molitona]

Hi @tarunKoyalwar @olearycrew @ehsandeep

I scan 16 million list with only 1 template, that's my case, is below command the right one for my case for faster work ?

nuclei - -l 16_milion_list -id xxxx -stats -rl 400 -bs 400 

NB: that dash after nuclei word is steam according to https://nuclei.projectdiscovery.io/nuclei/get-started/#mass-scanning-using-nuclei 'cause i saw ---> This option should only be enabled if targets > 10k . This skips any type of sorting or preprocessing on target list. on that link and it's my case i think ?

i run on vps machine with 500gb speed.

can you suggest the perfect tuned command args for the above one or it's perfect ? also is that steam dash correct syntax to put it after nuclei keyword ?

[tarunKoyalwar]

@molitona , interesting and thanks for sharing detailed information this is helpful . the ideal command for your setup should be

also its not a dash we will replace it with NA , it means it does not have short flag

nuclei -rl 0 -bs 10000 -l 16_milion_list -id xxx -stats -elog errors.txt -stream -o output.txt -nh

i set the -bs value to 10000 but i would recommend to test it for more (i think you can even set its value to 50000 or even 100000 if you have enough RAM)

Some Explaination

-rl 0 : you are only running 1 template so no need of ratelimit (anything other than 0 will affect your scan speed since ratelimit is global)
-bs 10000: also known as host concurrency (this is parameter you should tune for faster scans)
-stream : skips preprocessing like sorting etc
-nh : we can skip this if we are only sending one request per target (if more than 1 not using the flag is more helpful)

PS: I would like to hear back from you about stats like scan time etc if possible

[whereveryouare666]

thanks for info !

[molitona]

Hi @tarunKoyalwar,

Thank you for the detailed response.

I have 4GB RAM. So what is the max -bs value that i don't have to exceed according to your experience ?

Other than that i'll get back to you once I run a test.

Again, thanks for help.

[tarunKoyalwar]

@molitona , if its 4GB i think you might have to drop down bs to -bs 5000 . this is my estimation but you will get a better understanding after doing 1 or 2 scans whether to increase or decrease concurrency (just keep an eye out on RAM consumption and RPS in stats)

[molitona]

that was the case. what about 1gb ram ? how you calculate it ? Apr 15, 2023, 16:31 by ***@***.***:

…

@molitona <https://github.com/molitona>> , if its 4GB i think you might have to drop down bs to > -bs 5000> . this is my estimation but you will get a better understanding after doing 1 or 2 scans whether to increase or decrease concurrency (just keep an eye out on RAM consumption and RPS in stats) — Reply to this email directly, > view it on GitHub <#3534 (comment)>> , or > unsubscribe <https://github.com/notifications/unsubscribe-auth/AIPC4MO667RCIRG6B6XXP2TXBK5MJANCNFSM6AAAAAAW4IWUQQ>> . You are receiving this because you were mentioned.> Message ID: > <projectdiscovery/nuclei/repo-discussions/3534/comments/5624142> @> github> .> com>