* [MPTCP] Re: [PATCH mptcp-next v2] selftests: mptcp: add link failure test case
@ 2020-10-24 10:05 Matthieu Baerts
0 siblings, 0 replies; 4+ messages in thread
From: Matthieu Baerts @ 2020-10-24 10:05 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 1639 bytes --]
Hi Florian,
On 09/10/2020 17:11, Florian Westphal wrote:
> Add a test case where a link fails with multiple subflows.
> The expectation is that MPTCP will transmit any data that
> could not be delivered via the failed link on another subflow.
Thank you for these modifications!
(...)
> @@ -279,14 +308,32 @@ run_tests()
> listener_ns="$1"
> connector_ns="$2"
> connect_addr="$3"
> - rm_nr_ns1="${4:-0}"
> - rm_nr_ns2="${5:-0}"
> - speed="${6:-fast}"
> + test_linkfail="$4"
> + rm_nr_ns1="${5:-0}"
> + rm_nr_ns2="${6:-0}"
> + speed="${7:-fast}"
> lret=0
> + oldin=""
> +
> + if [ "$test_linkfail" -eq 1 ];then
> + size=$((RANDOM%1024))
> + size=$((size+1))
> + size=$((size*128)) > +
> + oldin=$(mktemp)
> + cp "$cin" "$oldin"
> + make_file "$cin" "client" $size
This will create one file up to 1024*128 KB = ~131MB when later using
this command:
dd if=/dev/urandom of="$name" bs=1024 count=$size
This file will be duplicated a few times later when doing the transfer:
( cat "$cin" ; (...) ; cat "$cin" ) | \
tee "$cinsent" | \
(...) $mptcp_connect (...) > "$cout"
If I am not mistaken, we then have:
- $cin = max 131MB: initial file
- $cinsent = 2 times $cin
- $sout = $cinsent
→ total: 5*$cin = max 655MB
Was it the intension? Could we reduce the maximum size while keeping the
test valid?
On my side, /tmp is mounted in tmpfs and currently limited to 490M. Just
to know if I need to increase the size of modify the selftest script :)
Cheers,
Matt
--
Tessares | Belgium | Hybrid Access Solutions
www.tessares.net
^ permalink raw reply [flat|nested] 4+ messages in thread
* [MPTCP] Re: [PATCH mptcp-next v2] selftests: mptcp: add link failure test case
@ 2020-10-24 12:17 Florian Westphal
0 siblings, 0 replies; 4+ messages in thread
From: Florian Westphal @ 2020-10-24 12:17 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 614 bytes --]
Matthieu Baerts <matthieu.baerts(a)tessares.net> wrote:
> ( cat "$cin" ; (...) ; cat "$cin" ) | \
> tee "$cinsent" | \
> (...) $mptcp_connect (...) > "$cout"
>
>
> If I am not mistaken, we then have:
>
> - $cin = max 131MB: initial file
> - $cinsent = 2 times $cin
> - $sout = $cinsent
>
> → total: 5*$cin = max 655MB
>
> Was it the intension? Could we reduce the maximum size while keeping the
> test valid?
Yes, I think its safe to reduce the size though.
Another option is to use /dev/zero as source and set the "sparse" flag for dd,
this would also reduce /tmp usage.
^ permalink raw reply [flat|nested] 4+ messages in thread
* [MPTCP] Re: [PATCH mptcp-next v2] selftests: mptcp: add link failure test case
@ 2020-10-17 7:31 Matthieu Baerts
0 siblings, 0 replies; 4+ messages in thread
From: Matthieu Baerts @ 2020-10-17 7:31 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 1037 bytes --]
Hi Florian, Mat,
On 15/10/2020 02:09, Mat Martineau wrote:
> On Fri, 9 Oct 2020, Florian Westphal wrote:
>
>> Add a test case where a link fails with multiple subflows.
>> The expectation is that MPTCP will transmit any data that
>> could not be delivered via the failed link on another subflow.
>>
>> Signed-off-by: Florian Westphal <fw(a)strlen.de>
>> ---
>> v2: rebase and add 'link loss' to test-case comment.
>> make sure we exit with 1 if file was corrupt or server/client
>> exited nonzero.
>>
>> .../testing/selftests/net/mptcp/mptcp_join.sh | 135 +++++++++++++-----
>> 1 file changed, 97 insertions(+), 38 deletions(-)
>>
>
> Thanks for the v2, looks good for the export branch.
Thank you for the patch and the review!
Just applied with Mat's Reviewed-by.
- 36221baa8712: selftests: mptcp: add link failure test case
- Results: 3fd01a5ab17b..0e70ef7be280
Tests and export will be started soon!
Cheers,
Matt
--
Tessares | Belgium | Hybrid Access Solutions
www.tessares.net
^ permalink raw reply [flat|nested] 4+ messages in thread
* [MPTCP] Re: [PATCH mptcp-next v2] selftests: mptcp: add link failure test case
@ 2020-10-15 0:09 Mat Martineau
0 siblings, 0 replies; 4+ messages in thread
From: Mat Martineau @ 2020-10-15 0:09 UTC (permalink / raw)
To: mptcp
[-- Attachment #1: Type: text/plain, Size: 661 bytes --]
On Fri, 9 Oct 2020, Florian Westphal wrote:
> Add a test case where a link fails with multiple subflows.
> The expectation is that MPTCP will transmit any data that
> could not be delivered via the failed link on another subflow.
>
> Signed-off-by: Florian Westphal <fw(a)strlen.de>
> ---
> v2: rebase and add 'link loss' to test-case comment.
> make sure we exit with 1 if file was corrupt or server/client
> exited nonzero.
>
> .../testing/selftests/net/mptcp/mptcp_join.sh | 135 +++++++++++++-----
> 1 file changed, 97 insertions(+), 38 deletions(-)
>
Thanks for the v2, looks good for the export branch.
--
Mat Martineau
Intel
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2020-10-24 12:17 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-10-24 10:05 [MPTCP] Re: [PATCH mptcp-next v2] selftests: mptcp: add link failure test case Matthieu Baerts
-- strict thread matches above, loose matches on Subject: below --
2020-10-24 12:17 Florian Westphal
2020-10-17 7:31 Matthieu Baerts
2020-10-15 0:09 Mat Martineau
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.