SIGSEGV when after forcing to install in remote docker

[ljubomirb]

I guess this is unrelated, since it maybe one off thing, or it may be docker thing, but if someone gets the same error, such as:

time="2020-04-30T22:24:54.720064266Z" level=error msg="failed to cleanup up stale network sandbox for container 30
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x18 pc=0x5651b5824ad5]

then the solution is just to delete containers AND YOUR VOLUMES of other docker containers will still be saved by doing: rm /var/lib/docker/containers/ -rf That is the short version. The longer version is - I was not able to find strait anwser - where and what to install in docker, so I found several docker files online, and tried some of them. Most expected to have docker on localhost (or better to say - in the same subnet), and probably all that you have certificates and server name. There is an option to run without certs, but I did not get to try that. One expected that you have mariadb running, and the trick is - log into that docker shell, and start app/cells install. When asked, you must choose "terminal" as installation method, since "browser" does not work, since it cannot be pointed to host mariadb server (something like 172.17.0.1). Anyway, in my case, docker crashed and I was not able even to restart easily, since mariadb stopped responding. I also observed memory usage being close to 1GB limit of that clud VM. My suggestion could be - please, add instructions for dummies of how to install to docker, and check that "broser install" part. I also must admit that I did not get to see if I was accessing pydio trough nginx proxy of a host, or directly from IP:8080 Again, this is probably just one off, but... let it be recorded if someone else also bumps here.

EDIT: after trying a few more hours, I come to see that it may have something to do with overuse of memory. For instance, at the moment, during install process, as started in docker console, shows that pydio cells eats 815MB/1GB, and all of swap (1GB), and the following line keeps repeating in install console: pydio.rest.config Ladon blocked GET request at /config/discovery. Ladon Response: DefaultDeny:true Also, mariadb is kind of unable to restart, and needs to be killed manually, though, this could be related to memory starvation. So, it may have to do with memory being low, or perhaps, single CPU on digital ocean does not have enough threads to run enough goroutines? Just a guess, but that should not be a problem. I have different go apps running on MIPS router with 64mb ram, so... I give up