The snmpd service keeps crashing and this time I got an error in syslog.
KERNEL : [156341.707801] snmpd[2967]: segfault at 0 ip 00007f4e56bf3783 sp 00007fff8497ab70 error 4 in libnetsnmpmibs.so.30.0.2[7f4e56b8f000+15e000]
Are you polling the device? If so, able you able to tie the crashes to specific events/polls? Which version of VyOS are you running? Is this a new install, or an existing where SNMP was not crashing initially? Any other log events at all around the same time?
I am polling with observium every 5 mins. The version is 1.1.6. Yes this is a new install starting happening after I made it prod. The router does no nat/firewall, ospf just 1 static route and bgp. I can start the snmpd service and then as soon as observium polls it it will crash.
config
interfaces {
ethernet eth0 {
address xxxxxxxxxx/30
address xxxxxxxx::1/64
description link-to-fw-public
duplex auto
hw-id 00:50:56:98:55:36
smp_affinity auto
speed auto
}
ethernet eth1 {
address xxxxxxxx/29
description xxxxxx_fiberlink
duplex auto
hw-id 00:50:56:98:5e:77
smp_affinity auto
speed auto
}
ethernet eth2 {
address xxxxxxxx/30
description xxxxxx_fiberlink
duplex auto
hw-id 00:50:56:98:0f:3c
smp_affinity auto
speed auto
}
ethernet eth3 {
address xxxxxxxxxxx/126
description xxxxxx_ipv6
duplex auto
hw-id 00:50:56:98:19:05
smp_affinity auto
speed auto
}
ethernet eth4 {
description spare
disable
duplex auto
hw-id 00:50:56:98:2a:10
smp_affinity auto
speed auto
}
ethernet eth5 {
description spare
disable
duplex auto
hw-id 00:50:56:98:2b:ee
smp_affinity auto
speed auto
}
loopback lo {
}
}
protocols {
bgp xxxxx {
address-family {
ipv6-unicast {
network xxxxxx:/32 {
}
}
}
neighbor xxxxxxx {
remote-as xxxxx
route-map {
}
soft-reconfiguration {
inbound
}
}
neighbor xxxxxx {
remote-as xxxxx
route-map {
}
soft-reconfiguration {
inbound
}
}
neighbor xxxxxxxx9 {
address-family {
ipv6-unicast {
soft-reconfiguration {
inbound
}
}
}
remote-as 2711
soft-reconfiguration {
inbound
}
}
network xxxxxx/20 {
}
network xxxxxxx0/20 {
}
network xxxxxxxxx/22 {
}
parameters {
disable-network-import-check
}
}
static {
route xxxxx/19 {
blackhole {
}
}
route xxxxx/20 {
blackhole {
}
}
route xxxxx/20 {
blackhole {
}
}
route xxxxx/32 {
blackhole {
}
}
route xxxxxxx/19 {
next-hop xxxxxx {
}
route6 xxxxx {
next-hop xxxxx {
}
}
}
}
service {
snmp {
community xxxx_public {
authorization ro
}
listen-address xxxxxxx {
port xx161
}
}
ssh {
listen-address xxxxxx
listen-address xxxxxx
listen-address xxxxxx
port xxxx22
}
}
system {
config-management {
commit-revisions 20
}
console {
device ttyS0 {
speed 9600
}
}
host-name 301bgp-rt
login {
user vyos {
authentication {
encrypted-password xxxxx
plaintext-password ""
}
level admin
}
}
name-server xxxxx
name-server xxxxx
ntp {
server 0.pool.ntp.org {
}
server 1.pool.ntp.org {
}
server 2.pool.ntp.org {
}
}
package {
auto-sync 1
repository community {
components main
distribution helium
password ""
url http://packages.vyos.net/vyos
username ""
}
}
syslog {
global {
facility all {
level notice
}
facility protocols {
level debug
}
}
host xxxxxx{
facility all {
level debug
}
}
}
time-zone US/Eastern
}Have you tried to remove polls one by one to see if you can find any relation?
It is still a little soon but I disabled “ucd-mib” and I was able to poll 2x manual and will give it 15-20 mins to see if the auto poll crashes it.
It crashed again over night wit the same error as above. This time there was also error about
SNMPD : ioctl 35123 returned -1Which OIDs are being polled? Is this a physical or virtual environment?
I will have to run a debug poll tonight. This is hosting in vmware with e1000 nics on vmware 5.5.
I know that this doesn’t help too much, but: happens to me as well. It started a few days ago, whereas the other router with the exact same VyOS version and hardware doesn’t have any problems. SNMP crashes suddenly with snmpd[3989]: segfault at 0 ip 00007f9b98eb1783 sp 00007fffbaa7c930 error 4 in libnetsnmpmibs.so.30.0.2[7f9b98e4d000+15e000]
Polling of the router didn’t change, it’s the same as before and on the other router. Did anyone find a solution for this crashing symptom?
No I did not find a fixed. I just wrote a script that checks to see if snmpd is running if not restart it. Then made a cron job to run the script every 5 mins.