Must restart CI multiple times because of an aesm error

[clangenb]

It has been a while since I filed a few PRs, so I can't say how recent this is:

[init ../../../psw/ae/aesm_service/source/core/ipc/UnixCommunicationSocket.cpp:225] Failed to connect to socket /var/run/aesmd/aesm.socket
[2023-08-01T05:15:25Z ERROR integritee_service::ocall_bridge::ffi::init_quote] [-]  Failed to init quote: InitQuote(SGX_ERROR_SERVICE_UNAVAILABLE)
[2023-08-01T05:15:25Z ERROR itp_stf_executor::executor] Fatal Error. Failed to attempt call execution: Sgx(SGX_ERROR_SERVICE_UNAVAILABLE)

It occurred on #1398 and on #1399 often enough so that I am not 100% sure that my change has nothing to do with it. On the other hand, #1375 was merged quite recently, it could also just be that it is not stable. My belief about instability is further strengthened by the fact that it mostly seems to occur with the long-running indirect invocation tests, where the socket is just crashes at some point.

[clangenb]

Ah, just saw this comment, which hardens my suspicion that it is unrelated to my PRs regarding the provisioning of the light client: #1375 (comment)

[brenzi]

this could be partially mitigated (or at least made less frequent) by #1405