3

So this is really odd to me. First, I guess, here is my setup:

root@kh13-9:/var/log/radosgw# cat /etc/*release*
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=14.04
DISTRIB_CODENAME=trusty
DISTRIB_DESCRIPTION="Ubuntu 14.04.3 LTS"

I have a 2x10Gib SFP+ port card and I am using 1 port on that card. 

Settings for p7p1:
    Supported ports: [ FIBRE ]
    Supported link modes:   10000baseT/Full 
    Supported pause frame use: No
    Supports auto-negotiation: No
    Advertised link modes:  10000baseT/Full 
    Advertised pause frame use: No
    Advertised auto-negotiation: No
    Speed: 10000Mb/s
    Duplex: Full
    Port: Direct Attach Copper
    PHYAD: 0
    Transceiver: external
    Auto-negotiation: off
    Supports Wake-on: d
    Wake-on: d
    Current message level: 0x00000007 (7)
                   drv probe link
    Link detected: yes

Here is the config i'm using for p7p1:

auto p7p1
iface p7p1 inet static
  address 10.64.64.152
  netmask 255.255.192.0
  network 10.64.64.152.0
  broadcast 10.64.127.255
  gateway 10.64.64.1
  dns-nameservers 10.100.100.251 10.100.100.252
  dns-search osdc.io
  mtu 9000
  post-up  /sbin/ip link set $IFACE txqueuelen 10000 || /bin/true
  post-up  /sbin/iptables-restore /etc/iptables.conf &>/dev/null || /bin/true

I have an address and I have network connectivity but I can not download/upload anything large without locking up my ssh session.

root@kh13-9:/var/log/radosgw# ip addr show p7p1
5: p7p1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 9000 qdisc mq state UP     group default qlen 10000
    link/ether 0c:c4:7a:bc:2c:de brd ff:ff:ff:ff:ff:ff
    inet 10.64.64.152/18 brd 10.64.127.255 scope global p7p1
       valid_lft forever preferred_lft forever

root@kh13-9:/var/log/radosgw# ping -c1 -w1 10.64.64.1 -I p7p1
PING 10.64.64.1 (10.64.64.1) from 10.64.64.152 p7p1: 56(84) bytes of     data.
64 bytes from 10.64.64.1: icmp_seq=1 ttl=64 time=0.195 ms

--- 10.64.64.1 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.195/0.195/0.195/0.000 ms

root@kh13-9:/var/log/radosgw# curl -s www.google.com >/dev/null && echo $?; echo 0

and by large I mean 100mib.bin from a local mirror.

root@kh13-9:/var/log/radosgw# wget    http://speedtest.dallas.linode.com/100MB-dallas.bin
--2016-08-31 16:31:10--  http://speedtest.dallas.linode.com/100MB-   dallas.bin
Resolving speedtest.dallas.linode.com (speedtest.dallas.linode.com)...    50.116.25.154, 2600:3c00::4b
Connecting to speedtest.dallas.linode.com    (speedtest.dallas.linode.com)|50.116.25.154|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: ‘100MB-dallas.bin.1’

0% [                                                                                                                                             ] 17,146      --.-K/s  eta 3d 5h

The file will never download and until I press ctrl+c the host doesn't seem to accept any more ssh connections other than my initial one.

Restarting the host fixes it but the issue comes back after some time. Everything on the switch seems fine to me. Everything on this host looks fine. There isn't any load, ram is fine, no swapping happening right now. I really have no idea what is going on right now.

I have ceph radosgw on this host and it seems to happen with any 14.04 node with radosgw running. The thing is, after I stop radosgw the issue persists until I reboot the server. I am lost. Does anyone have any idea on what this may be? I think this is a bug.

Lookcrabs
  • 31

0 Answers0