Question Pagespeed with Redis directory+permission problem

[Dukemaster]

Hi at all,
I have installed ngx_pagespeed with Ubuntu 16 built-in redis on amazing PLESK.
Since I installed Redis and integrated it with pagespeed a new cache directory was created

/var/cache/pagespeed

under root:root (755). As I changed permissions to nginx:nginx (755) pagespeed created the main shm_metadata_cache subfolder.
/var/cache/pagespeed/shm_metadata_cache/snapshot/,2Fvar/cache/pagespeed/,2Fmetadata_cache/1/64. But with a kind of double directory path?. Maybe it's okay. For the reason that it's my first installation and all searches in docus and internet (pagespeed) were not successful perhaps someone knows if it's correct or not.

My main problems are the missing v3 directory and !clean!time! file in /var/cache/pagespeed. They were not created and in the old ngx_pagespeed nothing has changed since yesterday evening (start to use Redis).

And the third problem in this directory/permission problem is that /var/log/pagespeed directory was created (last modified or created(?) is also yesterday evening).
This directory is empty, it leads to the fact that also wrong permissions are the cause. But I don't want to change it stupid to nginx, it is also possible that Redis is the owner. Does somebody knows which owner the pagespeed log has, and which permissions? I read the docu of redis, but I wasn't able to find owner and also not the permissions.
Would be nice if someone could help me.

Greets

 

[Dukemaster]

Console is working, all graphs and metrics. Dummies are created in /var/log/pagespeed.
For this I only changed the owner and group to redis, naturally without success.
Then back again to root:root and also changed permission from 755 to 777.
This worked.
But with 777 I'm not satisfied and so I experimented (sorry to say) with permissions. Now I changed both to nginx with 755.
Looking if the files get actualized, yet with an look to other logs like nginx, redis and the messages
Last night I made a experiment and mount /var/ngx_pagespeed_cache/ as tmpfs.
it works

root@server:~# df | grep tmpfs
tmpfs                   1639864    20580   1619284   2% /run
tmpfs                   8199312        0   8199312   0% /dev/shm
tmpfs                      5120       20      5100   1% /run/lock
tmpfs                   8199312        0   8199312   0% /sys/fs/cgroup
tmpfs                   1073152     3180   1069972   1% /var/ngx_pagespeed_cache
tmpfs                   1639864        0   1639864   0% /run/user/0
root@server:~# nginx -t
nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: configuration file /etc/nginx/nginx.conf test is successful
root@server:~# service nginx restart

Greets