Jump to content

JorgeB

Moderators
  • Posts

    67,415
  • Joined

  • Last visited

  • Days Won

    705

Everything posted by JorgeB

  1. It won't hurt, especially if you get more similar errors in the future.
  2. With PCIe 3.0 controllers like the ones I recommended you'll get around 2200MB/s usable per port, then divide that by the number of connected disks, you can see some more performance numbers here.
  3. If you have UD installed you can mount each disk read only and keep your parity 100% in sync.
  4. That looks about right to me considering the SSDs used, you need faster 3D TLC SSDs (860EVO, MX500, WD Blue 3D, etc), also higher capacity models, at least 250GB to get better than that.
  5. Then 3200Mhz is fine and supported, assuming no failing DIMMs.
  6. Then I would suggest something like the 9207-8e or 9300-8e, for expanders you can use Intel SAS2, there's a 24 and 36 port model, both can be molex powered.
  7. Forgot to say, if/when the hardware problem is resolved you can run scrub, it will identify the corrupt files, those will need to be deleted/recovered from backups.
  8. It's still overclock, except for 3rd gen Ryzen with only 2 DIMMs. If the corruption is happening on reads it will be fixed once the hardware problem is resolved, but note that there could still be undetected corruption that happened at write time, since the checksum will match the corrupted data.
  9. It's dropping offline, likely a SATA/power cable problem, try replacing or swapping them, you could also use a different controller, since trim won't work on the LSI HBA.
  10. Docker image is corrupt (likely from the cache running out of space) and needs to be recreated, also not that usable cache space is the same as the smallest device, despite what is shown on the GUI.
  11. Checksum failures are usually a sign of data corruption, start by looking a this to see if your RAM is overclocked, which is a known source of data corruption with Ryzen.
  12. Not clear if you mean really external ports or just horizontal ports on the right edge like this:
  13. JorgeB

    Squid is 50!

    Happy birthday Squid!! 🎂🎂 No idea you were such an old guy , though I'm not that far behind you.
  14. If you install the UD plugin it will spin them out after some time.
  15. You should run an extended SMART test.
  16. There's a problem with the HBA: Mar 21 13:07:18 Molar kernel: mpt2sas_cm0: fault_state(0x7c35)! Make sure it's well seated or try a different slot if available, you should also update the firmware since it's on a very old one.
  17. Unlikely the problem went away on its own, you'll probably get more errors on future checks.
  18. Try deleting network-rules.cfg (flash/congfig)) then reboot.
  19. Similar thing was also happening to me on one of my servers also with a dual port Mellanox, every few boots a new eth interface would be detected for port2, i.e., NIC that was eth4 became eth5, then eth6 and so forth, each time this happened I needed to reconfigure the last NIC detected as the previous one stopped working, my "solution" was to swap that NIC to a server with a different motherboard and it never happened again, firmware update might also help.
  20. The cable wasn't the problem, the different number of errors (on different sectors) suggest very strongly that there's a hardware problem, like bad RAM.
  21. Check output of: cat /sys/class/sas_host/host11/device/port-11\:0/sas_port/port-11\:0/num_phys 4 = single link 8 = dual link P.S.: If the command doesn't work post new diags since host# is based on the last ones and it can change.
  22. There are SAS expanders, but they require a SAS HBA, like the 9207.
  23. Settings -> VM Manager -> Default Windows VirtIO driver ISO (optional) -> Click download on latest version
×
×
  • Create New...