The Success Rate is abysmal. I’ve read a number of threads and the reason seems to be that I am losing races to other nodes. What I am looking for here are possible suggestions that could help win more of these races. The Rejected rate also seems high to me, is that expected?
Anyway, the following are my OS and CPU information:
storj@pine64:~$ uname -a
Linux pine64 3.10.105-bsp-1.2-ayufan-136 #1 SMP PREEMPT Sat Oct 27 21:18:35 UTC 2018 aarch64 aarch64 aarch64 GNU/Linux
storj@pine64:~$ lscpu
Architecture: aarch64
Byte Order: Little Endian
CPU(s): 4
On-line CPU(s) list: 0-3
Thread(s) per core: 1
Core(s) per socket: 1
Socket(s): 4
CPU max MHz: 1344.0000
CPU min MHz: 480.0000
This topic may be helpful to find out what parts are slow.
Also look at use of system resources to see if there are any bottlenecks. It seems you still have a max concurrent requests limit set, otherwise you would have no rejected transfers. What did you set it to?
The issue linked by @Cmdrd will also have a big impact though. So looking at other things should be done in the context of trying to optimize, not because there is necessarily anything wrong with your node.
Did not know about those debug functions to get this, thanks for the info there, will have to get that enabled on my nodes, more information to work with for node hardware performance improvements.
Thx, the debug port should come in handy, I will add the debug port.
I do have the max concurrent limit set, should I remove that? It is currently set to the default value of 7 because when I had it set higher it was complaining so I figured I would just set it to what the config.yaml says is the default value.
If you changed the config, you need to restart the node. You can do it as stop, then start or as just restart, depending on the version the command will be different. You can take this guide as a base: https://documentation.storj.io/resources/faq/system-maintenance