hbbr using unexpected amount of memory and error out after a day in self-hosted setup #438
Description
Describe the bug
When I use the self hosted server, it works for a day but the server would consistently fail after some time (1-2 days). When looking into it I found the following.
Jun 28 20:34:19 ip-172-31-0-49 hbbr[422]: memory allocation of 488135971 bytes failed
Jun 28 20:34:19 ip-172-31-0-49 systemd[1]: rustdesk-hbbr.service: Main process exited, code=killed, status=6/ABRT
Jun 28 20:34:19 ip-172-31-0-49 systemd[1]: rustdesk-hbbr.service: Failed with result 'signal'.
Jun 28 20:34:19 ip-172-31-0-49 systemd[1]: rustdesk-hbbr.service: Consumed 8.722s CPU time.
The server works after a restart.
Describe the environment
Installed on AWS EC2 instantce. Since rustdesk is quite lightweight, I used t3.nano instance with only 0.5GB of memory allocated. During normal operation this seems to be perfectly sufficient, but occationally I get the above error message, it seems hbbr is asking for a lot of memory sometimes.
Is this expected behabior? Also this is quite similar to issue #96 which was not fully explained.