Jump to content
Not connected, Your IP: 216.73.217.99
Sign in to follow this  
Lucan Niewhoner

ANSWERED What is is causing these sporadic throttle events on Torcular?

Recommended Posts

I've got a device connected to the Torcular server with - ah - a quite reliable upload stream. In the history graph you can see the server's traffic has dipped sharply and then quickly recovered several times recently, and is cratering again as I write this. It's especially annoying because my Plex server (which I'm exposing via an AirVPN port, long story) becomes basically unusable during these throttle events.

Side note, the dip is by more than just my contribution, lol. The dip is ~2Gbps whereas my usage rarely exceeds 200Mbps.

I can switch to another server that isn't doing this cratering, but I'm curious to know what's causing these throttles that doesn't result in a "Status Log" issue.

Thanks
image.png.51ac456276f80619ed0c2ee8033d5527.png
image.png.30c8389251adf3ac017524397c1ed7a8.png
 

image.png

Share this post


Link to post
10 hours ago, Lucan Niewhoner said:

what's causing these throttles that doesn't result in a "Status Log" issue.


Hello!

In a couple of cases, DDoS / flood attack was ongoing but was mitigated and caused no packet loss or other critical issue that could be detected by the monitoring system. However, they caused some problem: bandwidth impact and inability to communicate with the backend servers, which in turn did not have data to fill all the stats. However, the monitoring system did not put the server on "limited access" or "closed access" and did not log the issue mainly because IPv4 and IPv6 addresses were all reachable with no packet loss and not other failures were detected.

Another case is simply congestion on any upstream. Although we have a guaranteed bandwidth inside the datacenter and on interconnection with the main IP transit, some routes can be occasionally congested, even when this problem is not caused by the datacenter network and/or by the transit provider. However, this is a rare case, whereas flood attacks are more common throughout the whole infrastructure.

The last case is occasional. When the server needs an urgent kernel upgrade, it will be rebooted, and therefore will go completely down for a couple of minutes. However, this event is always recorded on the stats as it causes 100% packet loss on all addresses.

Kind regards
 

Share this post


Link to post

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Security Check
    Play CAPTCHA Audio
    Refresh Image
Sign in to follow this  

×
×
  • Create New...