In RoCE network, packet loss and dealy due to congestion can happen
quite often. We need to tolerate such event. So increase retry_count
and rnr_retry_count to 7 to allow NIC to retry operations when an
error happens, instead of returning the error directly which causes
the connection to be aborte
Ack.
What 0 values did mean? No retry or some default value?
On Thu, Aug 17, 2023 at 11:47 PM Kui Liu wrote:
>
>
>
> In RoCE network, packet loss and dealy due to congestion can happen
>
> quite often. We need to tolerate such event. So increase retry_count
>
> and rnr_retry_count to 7 to allow
Do we need it for vz9 as well?
--
Best regards,
Konstantin Khorenko,
Virtuozzo Linux Kernel Team
On 17.08.2023 17:45, Kui Liu wrote:
In RoCE network, packet loss and dealy due to congestion can happen
quite often. We need to tolerate such event. So increase retry_count
and rnr_retry_count to
I believe 0 values mean no retry based on test. With both values as 0, It
kept receiving IBV_WC_RETRY_EXC_ERR for RDMA READ operations, which aborts the
connection. Changing to 7 (which I find to be commonly used value in several
examples from RDMA Core library), the error goes away and we can
Yes, I think we need it for vz9, the patch can be applied directly to vz9.
-Original Message-
From: Konstantin Khorenko mailto:khore...@virtuozzo.com>>
Date: Friday, 18 August 2023 at 1:40 AM
To: Kui Liu mailto:kui@acronis.com>>, Devel
mailto:devel@openvz.org>>
Cc: Alexey Kuznets
