Question Xtream UI main server down after LB down

[Sami_boxy]

I m having an issue i have xtream ui running on Ubuntu 18 , with a main server that has a good CPU RAM....., everything work normal , until i get example an LB down or a specific channels down with lot of viewers , once those users start to request the channels that are off , the main server CPU goes up and i cant access , is like my own customers simulate a flood attack requesting the channels down , do you have any idea how i can fix this ,

 

[Robin-Hood]

I m having an issue i have xtream ui running on Ubuntu 18 , with a main server that has a good CPU RAM....., everything work normal , until i get example an LB down or a specific channels down with lot of viewers , once those users start to request the channels that are off , the main server CPU goes up and i cant access , is like my own customers simulate a flood attack requesting the channels down , do you have any idea how i can fix this ,

I have the same problem.

 

[Sami_boxy]

I have the same problem.

Any thought how it can be fixed or avoided

 

[alqui82]

everything work normal ..
How can everything work normal if your LB goes down??!!

 

[Sami_boxy]

everything work normal ..
How can everything work normal if your LB goes down??!!

LB can be down and main server still working no relation betwen them , the issue if the channels is down or LB down when many users start requesting stream from this LB off / channels off , the main server cpu went very high and crush , is like our own users are flooding the main server

 

[akawi11]

yes normal
example
main server with 1 gbit
2x LB 1 gbit
so it's 1 gbit + 1 gbit

it's like 2 gbit try to access from 1 gbit main server to redirect to LB
imagine when there is many LB 1+1+1+1 try to accessat same min or same seconds

it's not about CPU or RAM it's also about network interface

ideas:
you can separate/duplicate the same channel in others LB to split the traffic when one channel down, separate bouquets to have many ID's to same channel, but if all them have same source, so it's same issue.
like put SD in LB and HD in another LB, so if SD channel down, they can go to HD and back to SD when it's online again
many reverse proxy behind main server, so you can stop an proxy until some users connected

as we know it's happen on peak time or event time
do not worry about that issue, just do not do more many things on the main server when you have same issue and keep restart services with command to drop some new connection's ( like stop it ) and let others connection's connect to LB

 

[Robin-Hood]

Any thought how it can be fixed or avoided

I don't have any solution for this problem, but through ssh I see that the high cpu consumption comes from mysql.

 

[keyvpn]

I don't have any solution for this problem, but through ssh I see that the high cpu consumption comes from mysql.

check error log in xtream logs, i think youll find youve been hacked

 

[b4dreligion]

Hi, anyone found solution for this?

 

[guetteur]

Hi, anyone found solution for this?

I don't have any solution for this problem, but through ssh I see that the high cpu consumption comes from mysql.

Read again, and use he search ?

Question - Xtream UI main server down after LB down

I m having an issue i have xtream ui running on Ubuntu 18 , with a main server that has a good CPU RAM....., everything work normal , until i get example an LB down or a specific channels down with lot of viewers , once those users start to request the channels that are off , the main server CPU...

www.worldofiptv.com

 

[b4dreligion]

Read again, and use he search ?

Question - Xtream UI main server down after LB down

I m having an issue i have xtream ui running on Ubuntu 18 , with a main server that has a good CPU RAM....., everything work normal , until i get example an LB down or a specific channels down with lot of viewers , once those users start to request the channels that are off , the main server CPU...

www.worldofiptv.com

I don't think there's a solution, but a way to try to soften the impact. My question was whether anyone already had a solution.