Solved SBC 100% CPU utilisation

[null0]

I've noticed since an auto-upgrade to 3CX v15.5.8801.3 that my SBC's CPU utilisation is sitting at 100%.

It seems rebooting the 3CX server relieves the load temporarily, however utilisation ramps back up with in a few minutes.

I have tried re-installing the SBC, even tried it on a fresh install of Debian 9.

Wondering if anybody else is experiencing this?

 

[null0]

I've noticed this when I turn on VERBOSE logging on the SBC:

debug:3CXTunnel[1470]::2018-02-15T14:43:52.882687+00:00: TUNL | 140283480742080 | /home/repomaster/root15/Sources/Projects/3CXSBC/TunnelTcp.cpp:457 | FdSet timeout = 716
debug:3CXTunnel[1470]::2018-02-15T14:43:52.882724+00:00: TUNL | 140283480742080 | /home/repomaster/root15/Sources/Projects/3CXSBC/TunnelTcp.cpp:457 | FdSet timeout = 716
debug:3CXTunnel[1470]::2018-02-15T14:43:52.882760+00:00: TUNL | 140283480742080 | /home/repomaster/root15/Sources/Projects/3CXSBC/TunnelTcp.cpp:457 | FdSet timeout = 716
debug:3CXTunnel[1470]::2018-02-15T14:43:52.882797+00:00: TUNL | 140283480742080 | /home/repomaster/root15/Sources/Projects/3CXSBC/TunnelTcp.cpp:457 | FdSet timeout = 716
debug:3CXTunnel[1470]::2018-02-15T14:43:52.882833+00:00: TUNL | 140283480742080 | /home/repomaster/root15/Sources/Projects/3CXSBC/TunnelTcp.cpp:457 | FdSet timeout = 716
debug:3CXTunnel[1470]::2018-02-15T14:43:52.882870+00:00: TUNL | 140283480742080 | /home/repomaster/root15/Sources/Projects/3CXSBC/TunnelTcp.cpp:457 | FdSet timeout = 716
debug:3CXTunnel[1470]::2018-02-15T14:43:52.882906+00:00: TUNL | 140283480742080 | /home/repomaster/root15/Sources/Projects/3CXSBC/TunnelTcp.cpp:457 | FdSet timeout = 716
debug:3CXTunnel[1470]::2018-02-15T14:43:52.882943+00:00: TUNL | 140283480742080 |

Looks like the SBC might be missing keepalives to the 3CX server, then starts flooding them?

 

[null0]

Think I've figured this out, our upstream provider use Sonicwall firewalls and are a bit keen on filtering & IPS. I've tunnelled the traffic inside a VPN and it's absolutely fine now. Looks like the retries were valid.

 

[YiannisH_3CX]

Glad to see the issue has been resolved and thank you for sharing your solution.