yuzu-emu/mbedtls
1
0
Fork 0
mirror of https://github.com/yuzu-emu/mbedtls.git synced 2024-11-26 18:45:43 +01:00
Code Issues Packages Projects Releases Wiki Activity

Add delay in test to avoid race condition

Browse Source 
We previously observed random-looking failures from this test. I think they
were caused by a race condition where the client tries to reconnect while the
server is still closing the connection and has not yet returned to an
accepting state. In that case, the server would fail to see and reply to the
ClientHello, and the client would have to resend it.

I believe logs of failing runs are compatible with this interpretation:
- the proxy logs show the new ClientHello and the server's closing Alert are
  sent the same millisecond.
- the client logs show the server's closing Alert is received after the new
  handshake has been started (discarding message from wrong epoch).

The attempted fix is for the client to wait a bit before reconnecting, which
should vastly enhance the probability of the server reaching its accepting
state before the client tries to reconnect. The value of 1 second is arbitrary
but should be more than enough even on loaded machines.

The test was run locally 100 times in a row on a slightly loaded machine (an
instance of all.sh running in parallel) without any failure after this fix.
This commit is contained in:
Manuel Pégourié-Gonnard 2018-08-21 12:17:54 +02:00
parent 661103595e
commit 2f2d9020cd
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tests/ssl-opt.sh
View File

@ -5139,6 +5139,8 @@ run_test "DTLS fragmenting: proxy MTU, simple handshake" \
# Since we don't support reading fragmented ClientHello yet, # Since we don't support reading fragmented ClientHello yet,
# up the MTU to 1450 (larger than ClientHello with session ticket, # up the MTU to 1450 (larger than ClientHello with session ticket,
# but still smaller than client's Certificate to ensure fragmentation). # but still smaller than client's Certificate to ensure fragmentation).
# reco_delay avoids races where the client reconnects before the server has
# resumed listening, which would result in a spurious resend.
not_with_valgrind # spurious resend due to timeout not_with_valgrind # spurious resend due to timeout
requires_config_enabled MBEDTLS_SSL_PROTO_DTLS requires_config_enabled MBEDTLS_SSL_PROTO_DTLS
requires_config_enabled MBEDTLS_RSA_C requires_config_enabled MBEDTLS_RSA_C
@ -5152,7 +5154,7 @@ run_test "DTLS fragmenting: proxy MTU, resumed handshake" \
"$P_CLI dtls=1 debug_level=2 \ "$P_CLI dtls=1 debug_level=2 \
crt_file=data_files/server8_int-ca2.crt \ crt_file=data_files/server8_int-ca2.crt \
key_file=data_files/server8.key \ key_file=data_files/server8.key \
mtu=1450 reconnect=1" \ mtu=1450 reconnect=1 reco_delay=1" \
0 \ 0 \
-S "resend" \ -S "resend" \
-C "resend" \ -C "resend" \