Grenade jobs failing due to failed ping to the test instance

[Slawek Kaplonski]

It happens in all kinds of our grenade jobs. Examples of failures:

    https://58140b717b0fd2108807-1424e0f18ed1d65bb7e7c00bb059b2d8.ssl.cf2.rackcdn.com/872830/1/gate/neutron-ovs-grenade-multinode/6b0f11b/controller/logs/grenade.sh_log.txt

    https://bca8357d212b699db2ea-90e26954923e94ff0df2f7d3073d9fdf.ssl.cf2.rackcdn.com/871983/3/check/neutron-ovn-grenade-multinode-skip-level/fd1c1fa/controller/logs/grenade.sh_log.txt

    https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_01d/873615/2/check/neutron-ovs-grenade-dvr-multinode/01d90b6/controller/logs/grenade.sh_log.txt

Error snippet from grenade.sh.log:
https://paste.opendev.org/show/bMXLQ5lFsV2Nj1QGcq1J/

Every time it happens in the same stage, after smoke tests run and pass.

[yatin (yatinkarel)]

All recent runs looks good, Just dropping importance as we need to root cause the issue in bug description if it still happens:-
- https://zuul.openstack.org/builds?job_name=neutron-ovs-grenade-multinode
- https://zuul.openstack.org/builds?job_name=neutron-ovn-grenade-multinode-skip-level
- https://zuul.openstack.org/builds?job_name=neutron-ovs-grenade-dvr-multinode

[yatin (yatinkarel)]

Seen in https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_19d/874232/1/check/grenade/19d599c/controller/logs/grenade.sh_log.txt

[Rodolfo Alonso (rodolfo-alonso-hernandez)]

Patch for grenade: https://review.opendev.org/c/openstack/grenade/+/874417

[yatin (yatinkarel)]

Pushed https://review.opendev.org/c/openstack/grenade/+/874417 to collect server console log if ping fails.

[yatin (yatinkarel)]

Ok it reproduced and we have console logs now:-
https://79513b8e59d721f7ba79-3211a8aa52f1a73b6f8e95a0078f36dc.ssl.cf1.rackcdn.com/874232/1/gate/neutron-ovs-grenade-multinode/82d0db0/controller/logs/grenade.sh_log.txt

So based on this:-
- Server ACTIVE at 13:49:02.205
- It tried 30 seconds waiting for ping succeed and timed out at 13:49:39.975
- From console log dhcp and metadata requests succeed, public-key fetch failed
    udhcpc: lease of 10.1.1.31 obtained, lease time 86400
    successful after 1/20 tries: up 39.70. iid=i-0000000b
    failed to get http://169.254.169.254/2009-04-04/meta-data/public-keys(expected as keypair is not passed)
- From neutron dhcp 13:49:41.463183 np0033217712 dnsmasq-dhcp[95819]: DHCPACK(tap080dcbda-71) 10.1.1.31 fa:16:3e:31:64:be nova-server1

So seems guest vm just slow to boot and increasing timeout from 30s to 40+second should help, but can check few more fails to see if it's always the same issue or another.

[yatin (yatinkarel)]

Check few more failures:-
- [1][2][3] all same issue as comment #5(info: /etc/init.d/rc.sysinit: up at 33.36)
info: /etc/init.d/rc.sysinit: up at 40.04
info: /etc/init.d/rc.sysinit: up at 40.58
info: /etc/init.d/rc.sysinit: up at 35.83

[4] slight difference server ACTIVE at 15:42:16.096 and pinged timeout at 15:42:53.729
info: /etc/init.d/rc.sysinit: up at 14.80
udhcpc: sending select for 10.1.3.219

no more messages after ^ in console log run at 15:42:56.518 even DHCPACK was 15 seconds before.

Feb 22 15:42:38.531241 np0033220059 dnsmasq-dhcp[93999]: DHCPACK(tap060de802-ca) 10.1.3.219 fa:16:3e:2c:6d:b1 nova-server1

[1] https://0fe40d3c8f5f65d37e16-935e188dfa3a1a49945ee8787d93930a.ssl.cf5.rackcdn.com/871104/6/check/manila-grenade/f4ff59f/controller/logs/grenade.sh_log.txt
[2] https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_9d5/874811/1/check/grenade/9d5bd12/controller/logs/grenade.sh_log.txt
[3] https://81b633e2c5fe858f8400-d324a81a71d524d51ede3dc5aee27774.ssl.cf5.rackcdn.com/850499/21/check/grenade-skip-level/c068cb6/controller/logs/grenade.sh_log.txt
[4] https://cf3dfb14062b562daedb-2ff8f9467c39881ef83876a3ff9f514b.ssl.cf1.rackcdn.com/873699/4/check/neutron-ovs-grenade-multinode/c5f681f/controller/logs/grenade.sh_log.txt

[yatin (yatinkarel)]

Pushed https://review.opendev.org/c/openstack/grenade/+/874822 to bump ping timeout to 60 seconds.