[dpdk-dev] Secondary process crash in rte_eal_init

[Burakov, Anatoly]

On 06-Apr-19 3:26 PM, Jim Holland (jimholla) wrote:
> Hi,
> 
> We're seeing something similar to what is described in thread below. Our product uses dpdk 17.08. Was there ever a resolution to Souvik's issue?
> 
> Thanks...Jim
> 
> On 01/04/2015 06:00, Dey, Souvik wrote:
>> Hi All,
>>                   We have a single primary application with multiple secondary applications launched on the same cpu core. When the system boots up for the first time ,the primary comes up followed by the secondary process everything works fine. But in between if I try to restart the secondary processes without restart the primary, the secondary process crashes with the below stack trace. Any idea what is going wrong out here.
>>
>> ============================================
>> Stack Trace of the Crash :
>> (gdb) bt
>> #0 0x00007f4dad6e01b5 in raise () from /lib/libc.so.6
>> #1 0x00007f4dad6e2fc0 in abort () from /lib/libc.so.6
>> #2 0x0000000000402545 in __rte_panic ()
>> #3 0x00000000007353f4 in rte_eal_init ()
>> #4 0x0000000000403133 in main (argc=10, argv=0x7fff2c52fea8)
>> ============================================
>>
>>
>> --
>> Regards,
>> Souvik
>>
> Hi Dey,
> 
> Could you try to reproduce the issue with one of the multi-process
> examples from DPDK?
> Also you could provide some info such as command line options when
> running both primary/secondary apps, dpdk version, etc...
> 
> Sergio
> 

The crash you are seeing is a panic, but i don't see *what* is 
panic'ing, so i cannot diagnose the issue from stack trace alone.

In general, restarting secondary processes during primary's runtime is a 
supported case, and is indeed the basis for one of our unit tests 
(granted, most secondary process spinups in our unit tests are supposed 
to fail, but some will pass), and i have at various points done 
"secondary process spam" tests firing up tens of thousands of 
secondaries as well. So, crashes on secondary initialization with valid 
parameters is certainly not normal behavior.

A more detailed crash log would be good to have.

-- 
Thanks,
Anatoly