Carl
Carl
It's a shit ... Worst than a raspberry ... But it's my shit ... I did not have a screen to install so i used the 14" tv ..
dapit
Two hardrives?
Carl
Two hardrives?
yeah ... the small one is from a toshiba laptop 500 gb for system and data backups and another 500 gb will be filled with music to make a personal stream radio
dapit
With MPD?
Carl
but i'm preparing the environment because openbsd have some differences and i must make some changes
dapit
My plan is to make a Plex server
Carl
nice ... hope you have more time than me ...my job steal me a lot of time right now and i go steeeep by steeeep too slow
Carl
the job, the girl, the dog plus the time i spend in sleep and eat ... so free time for my own = almost nothing
./pascal.sh
dapit
Yes..but first I need to make the case.
dapit
Lol...I supposed to do it this weekend. But my lazyass rather watch movies all day.
dapit
Also I'm not going to build from scratch like Carl did, I'm looking for something that can fit the parts.
Phil
Hi, wich is the fastest db ? (to use with php). I was wondering if awk is good choice ? But in my database I want also to store some bytes (then i dont think it's good for this purpose)
Julian
What is the use case?
Julian
Without that we cannot give any suggestion.
Julian
Caching with key value I would say redis. But if you need relations obviously not. Then something more sql based like postgresql. Also depends on the architecture of your application. Distributed? Cassandra maybe or replicated/sharded mariadb... So many options
Julian
But reading your question tells me that you did not think about these things as you do it as a hobby. In that case I would suggest picking the one you favor and would like to learn :)
Phil
For the moment I have'nt already fixed in my mind the type of architecture
Phil
But the main use case is reading fast in multi rows (there will be a small number of columns about 2)
Julian
What is the data that you will store?
Julian
Do you need more tables than one?
Phil
3 tables
Julian
Could you store the object simply in JSON?
Phil
only strings
Julian
Do you need to persist or is it for runtime only?
Phil
no I need to persist
Julian
I see. How do the tables differ? Is there any relation between them? Do you work with foreign keys?
Phil
only for one table where the structure is 3 columns : first strings and second strings and third bytes
Phil
yes I need a foreign key
Phil
only one to connect users
Julian
Interesting usecase. May I ask why you want to persist binary?
Phil
cause I want to store messagePack datas
Phil
which works like JSON
Phil
but it's smaller
Julian
Ah protobuf alternative
Phil
due to the conversion in binary mode
Julian
In that case I would go with Postgres (easy expansion in the future, simple to setup)... MongoDB would be a little overkill for my taste but what would also work is Redis with FS persistence. Allows you to persist binary and offers more than a simple key value store.
The choice is up to you
Julian
Postgres may be a little more work initially compared to redis with FS
Phil
Are they fast in read mode ?
Julian
Indeed.
Julian
For reading almost every database is as fast as you could think
Julian
How long do you intend to store the data? Maybe your Architecture is suited for eventing?
Phil
I though for long term storing at least more than a year
Phil
thanks for your time and expertise
Julian
Julian
And you're welcome :)
./pascal.sh
the syncthing daemon is eating up over 100% cpu usage
is that normal?
./pascal.sh
ok now just about 20-30%
./pascal.sh
Mh so i tried adding temporary directories to the ignore list
I hope that helps
Mr.
hello everyone
./pascal.sh
Hi
./pascal.sh
Can someone tell me how I can upgrade my whole ports ?
./pascal.sh
ok i now did portmaster -af
dapit
portsnap fetch update
./pascal.sh
Yes i did that before
./pascal.sh
But it just fetches the latest ports
Ulf
Hi there,
Can QEMU/KVM be installed on FreeBSD 12.1 without any problems?
Ulf
I ask because VirtualBox seem to be "broken".
Ulf
I need to set up a few virtual machines on my server.
Jay
FreeBSD has native virtualization -> bhyve
Ulf
Yes
But it seems advanced for a FreeBSD noob like me...
😉
Ulf
Can you recommend any good guides for bhyve?
Phil
Hi I want to use postgresql with FreeBSD, is the choice of filesystem will impact performances (see there https://wiki.postgresql.org/wiki/FreeBSD)
Phil
wich filesystem do you recomand for : UFS or ZFS ?
./pascal.sh
./pascal.sh
i would go with ZFS
Phil
Thank you
Denis
Denis
You may want to disable journalling, not sure if it will ease io load much
Denis
But dont trust guides which tells to turn off softupdates.
./pascal.sh
I'm going to install freebsd on an ssd on my laptop
Should I really rather go with ufs than zfs ?
Phil
I know that if you use about 80% of the partition than ZFS will start to lag
./pascal.sh
Oh OK
Denis