50% interrupt
-
MSS clamping is on?
-
No
-
Something's beating it to death on em1 from the looks of it, the "100.00% [intr{irq259: em1:rx0}". Huge amount of load on receive. What traffic you see on that NIC? 'tcpdump -ni em1' on console while it's looking like that. em1 your LAN NIC or?
-
Yeah, em1 is LAN nic correct.
Funny thing is, there is NO traffic at all :)
-
What hardware's that?
-
em1: <intel(r) 1000="" pro="" network="" connection="" 7.4.2="">port 0x2000-0x201f mem 0xd0120000-0xd013ffff,0xd0100000-0xd011ffff,0xd0140000-0xd0143fff irq 17 at device 0.0 on pci1
em1: Using MSIX interrupts with 3 vectors
em1: netmap queues/slots: TX 1/1024, RX 1/1024Intel D2500CEE mobo.</intel(r)>
-
Is this still replicable for you? Haven't heard of it elsewhere.
-
This happened 2X when I streamed video over ipsec.
Disabled hardware stuff in advanced settings and hasnt happened since. -
Disabled hardware stuff in advanced settings and hasnt happened since.
Could you please the friendly and tell us all exactly what you have done?
Perhaps it might be that others also find this useful and/or over Google as a solution. -
Disabled hardware offload, tso all 3 options in advanced section of the config…
-
Which of those did you have enabled before? The defaults, TSO and LRO off, hardware checksum offloading on, should be fine there.
-
I had all enabled:
Disable hardware checksum offload
Disable hardware TCP segmentation offload
Disable hardware large receive offload