I'm running iceccs on a cluster where each machine is a member of the same public net, and a private net.
I'm trying to force iceccd to only look for private network for listening to scheduler server broadcasts, but it still tries to use interface connecting to a public net.
root@SRVMT6CTXB6P5:~# iceccd -i ens192 -v --help
usage: iceccd [-n <netname>] [-m <max_processes>] [--no-remote] [-d|--daemonize] [-l logfile] [-s <schedulerhost[:port]>] [-v[v[v]]] [-u|--user-uid <user_uid>] [-b <env-basedir>] [--cache-limit <MB>] [-N <node_name>] [-i|--interface <net_interface>] [-p|--port <port>]
root@SRVMT6CTXB6P5:~# ip address show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: ens160: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
link/ether 00:50:56:01:5e:0b brd ff:ff:ff:ff:ff:ff
inet 178.170.221.41/24 brd 178.170.221.255 scope global ens160
valid_lft forever preferred_lft forever
inet6 fe80::250:56ff:fe01:5e0b/64 scope link
valid_lft forever preferred_lft forever
3: ens192: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000
link/ether 00:50:56:01:5e:0e brd ff:ff:ff:ff:ff:ff
inet 10.0.0.24/24 brd 10.0.0.255 scope global dynamic ens192
valid_lft 6589sec preferred_lft 6589sec
inet6 fe80::250:56ff:fe01:5e0e/64 scope link
valid_lft forever preferred_lft forever
root@SRVMT6CTXB6P5:~# iceccd -i ens192 -vvv
[1677] 2020-06-13 16:50:41: ICECREAM daemon 1.3.1 starting up (nice level 5)
[1677] 2020-06-13 16:50:41: 8 CPU(s) online on this server
[1677] 2020-06-13 16:50:41: allowing up to 8 active jobs
[1677] 2020-06-13 16:50:41: supported features: env_xz env_zstd
[1677] 2020-06-13 16:50:41: not detaching
[1677] 2020-06-13 16:50:41: already a process group leader
[1677] 2020-06-13 16:50:41: ignoring localhost lo for broadcast
[1677] 2020-06-13 16:50:41: broadcast ens160 178.170.221.255
[1677] 2020-06-13 16:50:41: broadcast ens192 10.0.0.255
[1677] 2020-06-13 16:50:41: Netnames:
[1677] 2020-06-13 16:50:41: ICECREAM
[1677] 2020-06-13 16:50:41: starting to listen on interface ens192
[1677] 2020-06-13 16:50:41: starting to listen on interface lo
[1677] 2020-06-13 16:50:41: ignoring localhost lo for broadcast
[1677] 2020-06-13 16:50:41: broadcast ens160 178.170.221.255
[1677] 2020-06-13 16:50:41: broadcast ens192 10.0.0.255
[1677] 2020-06-13 16:50:41: scheduler not yet found/selected.
[1677] 2020-06-13 16:50:41: Suitable scheduler found at 178.170.221.39:8765 (version: 42)
[1677] 2020-06-13 16:50:41: scheduler not yet found/selected.
[1677] 2020-06-13 16:50:44: scheduler is on 178.170.221.39:8765 (net ICECREAM)
[1677] 2020-06-13 16:50:44: Connected to scheduler (I am known as 178.170.221.41)
Hello,
I'm running iceccs on a cluster where each machine is a member of the same public net, and a private net.
I'm trying to force iceccd to only look for private network for listening to scheduler server broadcasts, but it still tries to use interface connecting to a public net.