Intermittent failure of "Session resume .*, DTLS: openssl server"

[gilles-peskine-arm]

The ssl-opt.sh test case Session resume using tickets, DTLS: openssl server has a rare intermittent failure.

21:54:48 Session resume using tickets, DTLS: openssl server ..................... RETRY(client-timeout) FAIL
21:54:48   ! bad client exit code (expected 0, got 143)
21:54:48   ! outputs saved to o-XXX-148.log

• test_m32_o2, nightly job (private, will expire soon) on 9a34b60 (3.1.0+). Logs:
  all_u16-test_m32_o2-o.zip
• test_memory_buffer_allocator, PR job on TLS record HMAC #5573. Logs: all_u16-5573-o.zip
• test_malloc_0_null, PR job on lms: Move merkle tree generation to heap allocation #6595. Logs: all_u16-6595.zip

This may or may not be related to #5012.

[mpg]

Ran into an instance on the Open CI today:

[2022-03-16T13:19:17.573Z] Session resume using cache, DTLS: openssl server ....................... RETRY(client-timeout) FAIL
[2022-03-16T13:19:17.573Z]   ! bad client exit code (expected 0, got 143)
[2022-03-16T13:19:17.573Z]   ! outputs saved to o-XXX-170.log

And another.

[AndrzejKurek]

Related to #5279, might be fixed by #5686 - adding an additional retry (as originally intended).

[gilles-peskine-arm]

Seen once again in test_zlib_cmake in https://jenkins-mbedtls.oss.arm.com/blue/rest/organizations/jenkins/pipelines/mbed-tls-pr-head/branches/PR-8115-head/runs/2/nodes/511/steps/5195/log/ .

[gilles-peskine-arm]

Seen in test_memory_buffer_allocator in https://jenkins-mbedtls.oss.arm.com/blue/organizations/jenkins/mbed-tls-pr-head/detail/PR-7649-head/9/pipeline