Hi,
I want to know if anybody is having problems with fault tolerant job using
infiniband. When I run my job with tcp if anything happens with one node, my
job keeps running, but if I change my job to use infiniband if anything
happens with the infiniband (i.e cable problems) my job fails.
Anybody knows if there is something different that need to be done when
using openib instead tcp?
Bellow a example of the message I'm receiving from the mpi.
Regards,
Guilherme
--------------------------------------------------------------------------
The OpenFabrics stack has reported a network error event. Open
MPI
will try to continue, but your job may end up
failing.
Local host: XXXXX
MPI process PID: 23341
Error number: 10 (IBV_EVENT_PORT_ERR)
This error may indicate connectivity problems within the fabric;
please contact your system administrator.
--------------------------------------------------------------------------
[ZZZZ:23320] 15 more processes have sent help message
help-mpi-btl-openib.txt / of error event
[WWW:23320] Set MCA parameter "orte_base_help_aggregate" to 0 to see all
help / error messages
[[4089,1],144][btl_openib_component.c:3227:handle_wc] from XXXXX to: YYYYY
error polling LP CQ with status RETRY EXCEEDED ERROR status number 12 for
wr_id 214283560 opcode 51 vendor error 129 qp_idx 3
[[4089,1],147][btl_openib_component.c:3227:handle_wc] from XXXXX to: YYYYY
error polling LP CQ with status RETRY EXCEEDED ERROR status number 12 for
wr_id 490884096 opcode 1 vendor error 129 qp_idx 0
--------------------------------------------------------------------------
The InfiniBand retry count between two MPI processes has been
exceeded. "Retry count" is defined in the InfiniBand spec 1.2
(section 12.7.38):
The total number of times that the sender wishes the receiver to
retry timeout, packet sequence, etc. errors before posting a
completion error.
This error typically means that there is something awry within the
InfiniBand fabric itself. You should note the hosts on which this
error has occurred; it has been observed that rebooting or removing a
particular host from the job can sometimes resolve this issue.
Two MCA parameters can be used to control Open MPI's behavior with
respect to the retry count:
* btl_openib_ib_retry_count - The number of times the sender will
attempt to retry (defaulted to 7, the maximum value).
* btl_openib_ib_timeout - The local ACK timeout parameter (defaulted
to 10). The actual timeout value used is calculated as:
4.096 microseconds * (2^btl_openib_ib_timeout)
See the InfiniBand spec 1.2 (section 12.7.34) for more details.
Below is some information about the host that raised the error and the
peer to which it was connected:
Local host: XXXX
Local device: mlx4_0
Peer host: YYYY
You may need to consult with your system administrator to get this
problem fixed.
--------------------------------------------------------------------------
