Comment 2 for bug 1954763

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to neutron (master)

Reviewed: https://review.opendev.org/c/openstack/neutron/+/821727
Committed: https://opendev.org/openstack/neutron/commit/82aabb0aa962a3c5c5ce5ad1067952d8f3d9f992
Submitter: "Zuul (22348)"
Branch: master

commit 82aabb0aa962a3c5c5ce5ad1067952d8f3d9f992
Author: Slawek Kaplonski <email address hidden>
Date: Tue Dec 14 12:44:31 2021 +0100

    Allocate IPs in bulk requests in separate transactions

    In the ML2 plugin in create_port_bulk method, we are iterating over
    list of the ports to be created and do everything for all ports in
    single DB transaction (which makes totally sense as this is bulk
    request).
    But one of the things which was done during that huge transaction was
    allocation of the IP addresses for all ports. That action is prone for
    race conditions and can fail often, especially when there is no many IP
    addresses available in the subnet(s) for the ports.
    In case of the error while allocating IP address even for one port from
    the whole bulk request, whole create_port_bulk method was retried so
    allocations (and everything else) for all ports was reverted and started
    from scratch. That takes a lot of time so some requests may be processed
    very long time, like e.g. 2-3 minutes in my tests.

    To reproduce that issue I did simple script which created network with
    /24 subnet and then sent 24 requests to create 10 ports in bulk in each
    request. That was in totall 240 ports created in that subnet.
    I measured time of the creation of all those ports in the current master
    branch (without this patch) and with the patch. Results are like below:

    +-----+---------------+------------+---------------------------+
    | Run | Master branch | This patch | Simulate bulk by creation |
    | | [mm:ss] | [mm:ss] | of 10 ports one by one |
    +-----+---------------+------------+---------------------------+
    | 1 | 01:37 | 01:02 | 00:57 |
    | 2 | 02:06 | 00:40 | 01:03 |
    | 3 | 02:08 | 00:41 | 00:59 |
    | 4 | 02:14 | 00:45 | 00:55 |
    | 5 | 01:58 | 00:45 | 00:57 |
    | 6 | 02:37 | 00:53 | 01:05 |
    | 7 | 01:59 | 00:42 | 00:58 |
    | 8 | 02:01 | 00:41 | 00:57 |
    | 9 | 02:39 | 00:42 | 00:55 |
    | 10 | 01:59 | 00:41 | 00:56 |
    +-----+---------------+------------+---------------------------+
    | AVG | 00:02:07 | 00:00:45 | 00:58 |
    +-----+---------------+------------+---------------------------+

    Closes-Bug: #1954763
    Change-Id: I8877c658446fed155130add6f1c69f2772113c27