-
Notifications
You must be signed in to change notification settings - Fork 26
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Get some of your cpu back when being hammered (Discussion) #5
Comments
this is because of the way graphics are drawn. I have found the issue and almost fix it by changing the framework.
… On 13 Nov 2017, at 22:29, Gateway ***@***.***> wrote:
Hey guys, I wanted to say that we got hammered yesterday and had 8 cores pegged on a few servers. This was due to attacks and a lot of miners.
I was able to bring back some cpu by removing the graph stats container and the related javascript..
this is more of a discussion on server load help.
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub <#5>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADrjrcbUxetKMQB1WVjarFO8cqZf2eozks5s2KaTgaJpZM4QcXUs>.
|
Are the stats being pulled every tick? Could we just do it every x interval to update the chart.. and in between cache the chart data so its not always pulling? .. I think if switched to another graph package it might be wort doing this.. have you looked at highcharts? |
I have made changes to this, but they are experimental. I would share this with all the people eventually if I decide to continue my contribution on the open source z-nomp after all the attacks from the last 2 days.
… On 14 Nov 2017, at 21:08, Gateway ***@***.***> wrote:
Are the stats being pulled every tick? Could we just do it every x interval to update the chart.. and in between cache the chart data so its not always pulling? .. I think if switched to another graph package it might be wort doing this.. have you looked at highcharts?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#5 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADrjrQXXXHcOmiI9c3ExXlR5WUF5Lvy8ks5s2eUsgaJpZM4QcXUs>.
|
That brings up a bigger topic which I think we should open a discussion thread here on how to deal with attacks. I felt the same and their are a few things people should have in place to help. (fail2ban), their is a znomp filter for this. proper ip tables, cloudflare. On the ports for the miners filter html data POST/Get request.. all this crap ties stuff up in the event loop. Here is some data from newrelic from one of my servers.. a lot of time spent hammering live stats.. I think that should be cached more and not so live.. thoughts? Put in a variable to only get data every x seconds? oh I would be up for funding for any help.. I think the BTG peeps using should kick in a bit to help move this project forward.. but most importantly optimizations.. |
We use cloudflare and with specific settings for caching. We can open threads for these problems, because there is no universal approach but we can share opinion.
… On 14 Nov 2017, at 21:24, Gateway ***@***.***> wrote:
That brings up a bigger topic which I think we should open a discussion thread here on how to deal with attacks. I felt the same and their are a few things people should have in place to help. (fail2ban), their is a znomp filter for this. proper ip tables, cloudflare. On the ports for the miners filter html data POST/Get request.. all this crap ties stuff up in the event loop.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#5 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADrjrZ0kgNbr8C1lw16AOZt_wnB0XdFwks5s2ejegaJpZM4QcXUs>.
|
Hey guys, I wanted to say that we got hammered yesterday and had 8 cores pegged on a few servers. This was due to attacks and a lot of miners.
I was able to bring back some cpu by removing the graph stats container and the related javascript..
this is more of a discussion on server load help.
The text was updated successfully, but these errors were encountered: