Open Stack

This should be in the ops guide!


-------- Original Message --------
Subject: 	Re: [Openstack] Very slow connectivity from within tenant
network - GRE
Date: 	Wed, 23 Oct 2013 01:34:26 -0200
From: 	Martinx - ジェームズ <thiagocmartinsc at gmail.com>
To: 	Geraint Jones <geraint at koding.com>
CC: 	openstack at lists.openstack.org <openstack at lists.openstack.org>



WOW! Nice move!

But, upgrading from `openvswitch 1.10.2-0ubuntu2~cloud0` to `1.11.0-1`
did not solved my issue.

Tenant Instances still have a very slow Internet connectivity.

Thanks anyway! Nice to see your charts, pretty good improvement!

Regards,
Thiago


On 22 October 2013 22:48, Geraint Jones <geraint at koding.com
<mailto:geraint at koding.com>> wrote:

    I have just had to tweak our grizzly network node the biggest
    impacts were seem from doing changing root wrap to only use sudo –
    not the python wrapper (its super slow) and upgrading openvswitch to
    1.11

    This smoke ping shows the latency to one of our instances from
    europe before and after the openvswitch upgrade : http://d.pr/i/36v0

    And this graph shows the load avg on our network node, the first
    drop is from disabling root wrap the second is after the OVS upgrade
    : http://d.pr/i/xhFc

    I would suggest you do the same, and just make sure all MTU’s are
    correct.

    --
    Geraint Jones
    Director of Systems & Infrastructure
    Koding
    https://koding.com
    geraint at koding.com <mailto:geraint at koding.com>
    M (NZ) +64 22 123 4626 <tel:%2B64%2022%20123%204626>
    M (US) +1 415 316 8027 <tel:%2B1%20415%20316%208027>

    From: Martinx - $B%8%'!<%`%:(B <thiagocmartinsc at gmail.com
    <mailto:thiagocmartinsc at gmail.com>>
    Date: Tuesday, 22 October 2013 9:00 am
    To: Rick Jones <rick.jones2 at hp.com <mailto:rick.jones2 at hp.com>>
    Cc: "openstack at lists.openstack.org
    <mailto:openstack at lists.openstack.org>"
    <openstack at lists.openstack.org <mailto:openstack at lists.openstack.org>>
    Subject: Re: [Openstack] Very slow connectivity from within tenant
    network - GRE

    Hi Rick!

    Back with Grizzly, I faced that problem and I was able to detect it,
    at the Network Node with tcpdump and fix it by running "ip link set
    mtu 1454 dev eth0" within the Instance.

    Not this time... This is another problem...   ;-/


    On 22 October 2013 13:25, Rick Jones <rick.jones2 at hp.com
    <mailto:rick.jones2 at hp.com>> wrote:

        On 10/22/2013 01:32 AM, Martinx - $B%8%'!<%`%:(B wrote:

            Stackers,

            I'm trying to put my Havana into production and I'm facing a
            very
            strange problem.

            The Internet connectivity from tenant's subnet is very, very
            slow. It is
            useless in fact... I can not even use "apt-get update" from
            a Instance.

            The following command works (apt update from the tenant
            namespace):

            ---
            root at net-node-1:~# ip netns exec qrouter-XXXXXXXXX aptitude
            update
            ---

            But not from the tenant subnet...

            I'm following this topology:


http://docs.openstack.org/__trunk/install-guide/install/__apt/content/section_use-cases-__tenant-router.html

<http://docs.openstack.org/trunk/install-guide/install/apt/content/section_use-cases-tenant-router.html>

            Already tried to change MTUs (via DHCP agent)... Nothing had
            fixed this
            weird issue.

            Any thoughts?!

            Right now, my "aptitude safe-upgrade" will take 2 days to
            download
            60MB... During this network outages, even the SSH session stops
            responding for a few seconds...

            Everything else seems to be working as expected, as for
            example, DHCP,
            Floating IPs, Security Groups...

            Sometimes, even the first ssh connection to the Instance
            Floating IP,
            have a lag.


        It is but a guess, but I wonder if, even with changing MTUs (to
        what values?) you may still be experiencing a PathMTU+ICMP
        blackhole problem accessing nodes on the Internet.  Can you
        access something that is a bit "closer" but still outside your
        stack so you have a shot at looking at netstat statistics on the
        sender and/or get packet traces on the sender?

        You could still try taking packet traces at the instance or
        perhaps the namespace and try to discern packet losses at the
        receiving side, though it can be a bit more difficult.

        rick jones


    _______________________________________________ Mailing list:
    http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack Post
    to : openstack at lists.openstack.org
    <mailto:openstack at lists.openstack.org> Unsubscribe :
    http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack




-------------- next part --------------
_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to     : openstack at lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack