Apologies for leaving that out - yes, all the other servers were Debian
Wheezy, and all have been upgraded to Debian Jessie. All are running
The server in question does get a lot of traffic, and notably 127.0.0.1
has significant traffic - After two days uptime, there's 30 GiB of data
through lo. However, even that doesn't add up to greater than 3GB/hour.
On 10/18/16 07:55, Martin Pala wrote:
as you mentioned you have upgraded other servers in the farm and didn't
experienced this problem ... what was the version on these other servers? (both
Debian and monit?)
On 17 Oct 2016, at 22:55, Paul Theodoropoulos <p...@anastrophe.com> wrote:
I recently upgraded a couple of servers from Debian Wheezy to Debian Jessie.
Since then, I'm getting frequent alarms from my network bandwidth test.
check network network-bandwidth with interface eth0
if total upload > 3 GB in last 1 hours then alert
if total download > 5 GB in last 2 hours then alert
if total download > 10 GB in last day then alert
I wasn't getting alarms before the upgrade, so to try to be thorough, I rebuilt
monit from source. No change. This is 5.19.
The problem is that there's definitely not bandwidth anywhere near my testing
values taking place. With an uptime of 1 day, 14 hours, eth0 shows:
eth0 Link encap:Ethernet HWaddr 00:50:56:a7:3e:8e
inet addr:10.124.74.68 Bcast:10.124.74.127 Mask:255.255.255.192
inet6 addr: fe80::250:56ff:fea7:3e8e/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:27941718 errors:0 dropped:21 overruns:0 frame:0
TX packets:59281394 errors:0 dropped:0 overruns:0 carrier:0
RX bytes:333591539 (318.1 MiB) TX bytes:4005580141 (3.7 GiB)
Thoughts? I upgraded a number of other servers in our farm, and have not