#1392 Duffy seems to be inaccessible from CentOS CI OCP #3
Closed: Fixed 3 months ago by mrc0mmand. Opened 3 months ago by mrc0mmand.

I noticed that since ~4 hours ago all our CI jobs are stuck since the OCP pods can't reach Duffy:

sh-4.4$ hostname
cico-workspace-swgxj
sh-4.4$ ip a s dev eth0
3: eth0@if48076: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8951 qdisc noqueue state UP group default 
    link/ether 0a:58:0a:83:01:60 brd ff:ff:ff:ff:ff:ff link-netnsid 0
    inet 10.131.1.96/23 brd 10.131.1.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 fe80::c033:cbff:fec1:b812/64 scope link 
       valid_lft forever preferred_lft forever
sh-4.4$ ping duffy.ci.centos.org -c 10
PING duffy.ci.centos.org (172.19.0.18) 56(84) bytes of data.

--- duffy.ci.centos.org ping statistics ---
10 packets transmitted, 0 received, 100% packet loss, time 9236ms

This seems to be limited to that specific network, since I can reach Duffy just fine from my machines.


Metadata Update from @arrfab:
- Issue assigned to arrfab

3 months ago

Metadata Update from @arrfab:
- Issue tagged with: centos-ci-infra, medium-gain, medium-trouble

3 months ago

yes, there was a Fiber cut (I read the discussion today on the internal community-cage list) and while there should be redundant path through another provider/link, it seems it wasn't routed correctly so "flapping".
AFAICS it was all resolved, and zabbix is also happy now about all nodes back under control.
Can you just confirm and so close ticket ?

Metadata Update from @arrfab:
- Issue priority set to: Waiting on Reporter (was: Needs Review)

3 months ago

Yup, it looks like everything is running again, thanks!

Metadata Update from @mrc0mmand:
- Issue close_status updated to: Fixed
- Issue status updated to: Closed (was: Open)

3 months ago

Log in to comment on this ticket.

Metadata