Thanks, Joshua and Jerry Jelinek for your quick fix.

Very happy to see this  https://smartos.org/bugview/OS-5126 bug have been
fixed and it will be contained in next bi-weekly release. I will test on
this release later.

I digged more about aerospike on lx brand zone, following is what i have
found:

After aerospike server started, listening ports :

   - Service 3000 => Application, Tools, and Remote XDR use the Service
   port for database operations and cluster state.
   - Fabric 3001 => Intra-cluster communication port. Replica writes,
   migrations, and other node-to-node communications use the Fabric port.
   - Multicast port 9918 on 239.1.99.222 => Heartbeat protocol ports are
   used to form and maintain the cluster. (Only one heartbeat port may be
   configured.)
   - Info 3003 => Telnet port that implements a plain text protocol for
   administrators to issue info commands. For more information, see asinfo
   documentation.

Below is my test scenario:

Tested on SmartOS Live Image: 20160204T173339Z

   1. lx brand zone:
      - With image ubuntu-15.04 20160201,  8GB memory, install aerospike
      version: aerospike-server-community-3.7.3-ubuntu12.04
      - Install and starting aerospike success, but when i run `asinfo`
      randomly got error: can't connect 127.0.0.1: 3000 return error . Try
      10 times about 3 time success.
      - `netstat -g`  outputs nothing, but all ports listened (`netstat
      -an`).
   2. kvm:
      - With image ubuntu-certified-15.10 20160114, same vm configuration,
      install same aerospike installation package.
      - Everything worked as well.
      - `netstat -g` outputs,  multicast address works.

       IPv6/IPv4 Group Memberships

Interface       RefCnt Group

--------------- ------ ---------------------

lo              1      all-systems.mcast.net

eth0            1      239.1.99.222

eth0            1      all-systems.mcast.net

lxcbr0          1      all-systems.mcast.net

lo              1      ip6-allnodes

lo              1      ff01::1

eth0            1      ff02::1:ff9a:eaa1

eth0            1      ip6-allnodes

eth0            1      ff01::1

lxcbr0          1      ff02::1:ff9f:4935

lxcbr0          1      ip6-allnodes

lxcbr0          1      ff01::1


> I tried to use dtrace script lxunsup.d and lxsocketopt.d (
https://wiki.smartos.org/display/DOC/LX+Branded+Zones) on global zone to
probe the syscalls. i got `28824 systemd prctl option 30\0` when starting
aerospike server lx brand zone.  Wiki says `prctl` is harmless.

*So, is there any ideas on aerospike server 3000 port randomly return error
issue?*


On Thu, Jan 28, 2016 at 12:53 PM, Joshua M. Clulow <[email protected]> wrote:

> On 27 January 2016 at 20:23, Joshua M. Clulow <[email protected]> wrote:
> > It's possible that you could work around this by removing that line
> > from the startup script, but in order to reduce the friction for users
> > I imagine we can try and emulate (or at least silently ignore) this
> > one.
> 
> I have filed an RFE:
> 
> https://smartos.org/bugview/OS-5126
> 
> Thanks for reporting, Tim.
> 
> --
> Joshua M. Clulow
> UNIX Admin/Developer
> http://blog.sysmgr.org
> 



-- 

*Blog -> http://timtang.me <http://timtang.me>*
*Email -> [email protected] <[email protected]>*



-------------------------------------------
smartos-discuss
Archives: https://www.listbox.com/member/archive/184463/=now
RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb
Powered by Listbox: http://www.listbox.com

Reply via email to