Got it.

You're all set.



On 08/08/2017 02:22 PM, Neale Ranns (nranns) wrote:

Hi Steve,

No worries.

We use gerrit for our git interactions. In the gerrit link for that patch: <>

you’ll see a button for cherry pick, select it and enter the branch stable/1707. That’s it JWhen the CI jobs finish I can merge.



    *From: *Steve Heckman <>
    *Date: *Tuesday, 8 August 2017 at 19:16
    *To: *"Neale Ranns (nranns)" <>,
    "" <>
    *Subject: *Re: [vpp-dev] vpp v17.07 ping assert


    Sorry, I'm a bit new to this.

    I'm working with git cli. What GUI are you referring to? Eclipse/EGit?

    This is the change:

    mage removed by sender.

    Mohammed Hawari< <>>
    07/18/2017 09:25:01 AM +0200

    Committed by:John Lo< <>>
    07/27/2017 10:10:26 PM +0000

    ping: fixing wrong value when there are worker threads - the
    echo_reply_node is now notifying the cli process on the main
    thread/vlib_main - the timestamp for the icmp reply is now
    acquired in the echo_reply_node and not in the cli process to
    avoid an off by 10ms error (see 【vpp-dev】delay is error in ping
    with multi worker thread) Change-Id:
    I21d37002b0376b4f2ccab08d8f04c2f2944b9b39 Signed-off-by: Mohammed

    I've seen references to gerrit-cherry-pick. Should I use that?
    Thanks, -Steve

    On 08/08/2017 01:00 PM, Neale Ranns (nranns) wrote:

        Hi Steve,

        Could you please press the cherry pick button on those review[s] to get 
them into stable/1707.



        -----Original Message-----

        From:<> <>  on behalf of 
"Heckman, Steve"<> <>

        Date: Tuesday, 8 August 2017 at 17:01

        To:"" <>  <> 

        Subject: Re: [vpp-dev] vpp v17.07 ping assert

             FWIW...I back-merged the changes to ping.[ch] from master to 17.07 
locally and now it works.


             -----Original Message-----

<>  [] On 
Behalf Of Steve Heckman

             Sent: Tuesday, August 8, 2017 10:56 AM


             Subject: [vpp-dev] vpp v17.07 ping assert

             I've been running into issues on certain servers where I get an 

             Aug 08 09:51:52 p5s3 systemd[1]: vpp.service: Main process exited, 
code=killed, status=6/ABRT Aug 08 09:51:52 p5s3 systemd[1]: vpp.service: Unit 
entered failed state.

             Aug 08 09:51:52 p5s3 systemd[1]: vpp.service: Failed with result 

             here in ping.c

                vlib_process_signal_event (vm, pr->cli_process_id, event_type, 

             which calls vec_elt, where the assert actually occurs.

             /** \brief Get vector value at index i checking that i is in 
bounds. */

             #define vec_elt_at_index(v,i)                   \

             ({                                              \

                ASSERT ((i) < vec_len (v));                   \

                (v) + (i);                                    \


             /** \brief Get vector value at index i */ #define vec_elt(v,i) 

             pr->cli_process_id is set in run_ping_ip46_address() and always 
seems to

             be 344.

                uword curr_proc = vlib_current_process (vm);


                pr->cli_process_id = curr_proc;

             vec_len(v) seems to be 32 (Does that make sense?).

             Everything else seems to work OK. I can do "show interface (addr)", 
"show ip arp".

             Any ideas?

             This is stock apt install on Ubuntu 16.04 LTS on HPG9 and SMC 
servers with Intel 82599 NICs.


             Steve Heckman

             Principal Software Engineer



             vpp-dev mailing list




             vpp-dev mailing list



vpp-dev mailing list

Reply via email to