toptable

Members
  • Posts

    8
  • Joined

  • Last visited

toptable's Achievements

Noob

Noob (1/14)

2

Reputation

  1. Hello, I've just noticed today tat I can't reach the Web UI of my Unifi controller any more. No idea what's going on, but it just times out. When I try to ping the ip (including port) on the network it says there is no host. pinging the main ip of the server still works, as expected. Things I've tried: Updating the container. Restarting the container Changing tag to LTS and updating. No dice on any of these. Does anyone have any idea what the issue may be? EDIT: I also CAN'T reach the web UI of my Syncthing container, but I CAN reach the UI of Plex and Netdata, so the issue is obviously not across the board. EDIT2: Attached diagnostics old-hen-diagnostics-20210902-1033.zip
  2. Hello! I'm checking back in for completeness. Hopefully this will help someone with the same issue in the future. The advice I was given was bang on. My crappy ADATA SSDs were to blame. Use decent SSD drives with 3D TLC NAND. I'm using WD Blues. My cache drives are now running at line speed, all the time. See attached screengrab for the results. Thanks so much for the help.
  3. Righto, advice taken. A set of Crucial MX500s are on their way and the ADATAs are going back. I'll let you know how I get on when I receive and install them. Many thanks to you!
  4. Well I can certainly return them. It's not too much more to get some WD blues as you mentioned. So everything else with my config looks fine then? How about that trim thing, is that normal? Thanks again for all the help man
  5. Fair enough, they are economical SSDs, however I'd expect them to manage more than 20MB/s. I'd also expect them not to drop to 350KB/s pretty much ever! Wouldn't you? I've used my fair share of cheap SSDs in my time and never seen performance like that! Thanks for taking the time to help me!
  6. I turned on reconstruct write and copied a 13GB file to a share with cache set to "no" to test this. The copy averaged about 95MB/s. Stable speeds throughout. I've also attached a screengrab of me running trim 3 times in a row on the SSDs. The results are always the same. Is this right?
  7. Sorry, RAID1 is what it is! Will edit...
  8. Hello. Hoping someone can offer some help. I'm seeing extremely slow writes to the cache of my new unraid server. I'll try to set out what I've been doing below, along with things I've tried. I'll gladly provide any more info if you need it. I've attached my diagnostics as seems to be the norm, and also a screengrab of a large transfer (c. 80GB) to show the issue. I took the diagnostics during the transfer in the screenshot. I also noticed that the CPU is pretty much pinned while the transfer is going on (screenshot also attached), but when I looked in the processes tool, I couldn't see anything with high usage. Is this normal? CPU is an i5-4590. My setup is: 2x 240GB SATA SSDs for cache. These are set to the typical Raid 1. I have the trim plugin installed and set to daily. I also ran it from the terminal. I have the cache pool set to maintain 20GB of free space, and there is plenty left on the SSDs. Cache is set to "Yes" in the share I'm writing to. I've also confirmed it is indeed being written to the cache pool. Speeds were ok for the first couple of transfers I made, about 85MB/s, then suddenly dropped mid transfer to 20MB/s and have pretty much stayed there since. Many thanks for any help you can offer! old-hen-diagnostics-20201218-1602.zip