[[Support] MariaDB Official]

On 8/21/2023 at 5:25 PM, mgutt said:

Maybe using the command mariadb and not mysql?

 

On 8/22/2023 at 9:17 AM, thunderclap said:

Well I feel stupid, though in my defense the official container info was wrong so I don't feel quite as stupid. Thanks.

 

I just wanted to bring up that the official MariaDB docker container still has this incorrect information so I can see it being confusing for people (including myself).

 

 

I found the correct information at https://mariadb.com/kb/en/mysqldump/ 

[Performance issue with Unraid with external Plex library scans]

7 hours ago, Energen said:

Or perhaps you're just not using Plex correctly / optimally.  If you have Plex set up to do a partial scan on detected changes (or whatever the option is exactly called) there would never really be a need to perform a full scan.  And even with a full scan, why would you need to do it so often where that would make Unraid unusable for you?  Seems like you are the problem, not Unraid.

 

I personally never have to force a full library update because Plex always picks up my changes as they happen, and maybe I have a full scan scheduled by default but day to day it just works as expected and my library is never out of date.

I have Plex configured to do partial scans, but unfortunately due to limitations (in SMB if I remember correctly), you can only have so many folders before partial scans no longer detect changes, which means that some changes in my setup are not detected until a scheduled scan goes through every folder. To make things worse, there is also a bug in Plex (regarding Curl) that large libraries cause TCP port exhaustion, which was made worse in 1.19.X versions. I would love to not do scheduled scans, but it would involve breaking up my libraries and file shares past the segmentation that I already do (tv shows, movies, cartoons, anime, etc). After I broke about 70-80TB of video, things stopped scaling so nicely.

 

Also, your "seems like you are the problem, not Unraid" comment was not necessary. I only stated that Unraid wouldn't work for me, which it seems like it won't. I never bashed Unraid as a product, which I'm sure works great for many people, and don't feel I should be bashed in return.

7 hours ago, trurl said:

And generally, Unraid will not read/write as fast as any striped system. But many people have chosen Unraid precisely because it is NOT striped. Not striping means each disk can be read independently of any others on any Linux, you can mix differently sized disks in the array, and you can easily add disks without rebuilding the whole array.

One of the biggest selling points for me on Unraid was the ability to dynamically grow my arrays, which is why I was testing it to see the feasibility of moving over my fleet of servers to it over time. Having to purchase drives 8-12 new drives at a time can get rather expensive, especially because it usually ends with me buying another server as well. I didn't mind the reduced read/write speeds as only one server in my setup actually needs high performance read/write and I would have just left that one on FreeNAS, but the latency I saw above just would have killed it for me. Moving the entire fleet over to Unraid would mean I would be looking at 1.5+ hours for each forced scan, which would be unfeasible for what I would require.

[Performance issue with Unraid with external Plex library scans]

Thank you for the reply. I guess Unraid is just not for me.

[Performance issue with Unraid with external Plex library scans]

I've set up an environment to test out Unraid to see if I want to move over to it and leave FreeNAS, but I've hit a bit of a snag.

 

In this test environment, I have two Dell R510 servers that are virtually identical (same CPU, 16gb ram, same HBA, similar disks). One is running FreeNAS 11.3-U3.2 and the other I am trying out Unraid 6.8.3. Each server has 12 hard drives in a 2 parity scenario (FreeNAS is running Raidz2), no cache drive or anything. I already know about the read/write speed limitations of Unraid (i'm getting 100MBps read / 30MBps write on average over SMB, which I'm OK with).

 

For my test, I have folder with about 5tb of tv shows that are properly structured for use in Plex duplicated on both the FreeNAS server and the Unraid server, each shared through SMB. Each of these folders are created as separate libraries on a 3rd server running Windows 10 and Plex. When I force a library scan on the FreeNAS folder, it takes maybe 60-90 seconds to complete. When I force a library scan on the Unraid folder, it takes maybe 15-20 minutes. To my understanding, a cache drive isn't going to help with that kind of performance issue. In the forums, I saw someone recommend "Dynamix Cache Dirs" to cache the directories for scanning, but it didn't seem to help me at all. I tried the "max protocol = SMB2_02" for Samba, but didn't notice any difference. I've also notice things like it taking much longer to calculate storage used in "Properties" on the Windows box. Any ideas on how I can speed this up? I had FreeNAS on this server until recently so I doubt it is a hardware/storage issue.